aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/net/ethernet/intel
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/net/ethernet/intel')
-rw-r--r--drivers/net/ethernet/intel/e100.c2
-rw-r--r--drivers/net/ethernet/intel/e1000/e1000_main.c2
-rw-r--r--drivers/net/ethernet/intel/e1000e/netdev.c4
-rw-r--r--drivers/net/ethernet/intel/ice/ice.h59
-rw-r--r--drivers/net/ethernet/intel/ice/ice_adminq_cmd.h5
-rw-r--r--drivers/net/ethernet/intel/ice/ice_common.c117
-rw-r--r--drivers/net/ethernet/intel/ice/ice_common.h24
-rw-r--r--drivers/net/ethernet/intel/ice/ice_ethtool.c28
-rw-r--r--drivers/net/ethernet/intel/ice/ice_hw_autogen.h12
-rw-r--r--drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h46
-rw-r--r--drivers/net/ethernet/intel/ice/ice_lib.c180
-rw-r--r--drivers/net/ethernet/intel/ice/ice_lib.h4
-rw-r--r--drivers/net/ethernet/intel/ice/ice_main.c450
-rw-r--r--drivers/net/ethernet/intel/ice/ice_sched.c96
-rw-r--r--drivers/net/ethernet/intel/ice/ice_switch.c339
-rw-r--r--drivers/net/ethernet/intel/ice/ice_switch.h25
-rw-r--r--drivers/net/ethernet/intel/ice/ice_txrx.c694
-rw-r--r--drivers/net/ethernet/intel/ice/ice_txrx.h23
-rw-r--r--drivers/net/ethernet/intel/ice/ice_type.h5
-rw-r--r--drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c612
-rw-r--r--drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h20
-rw-r--r--drivers/net/ethernet/intel/igb/igb_ethtool.c4
-rw-r--r--drivers/net/ethernet/intel/igb/igb_main.c5
-rw-r--r--drivers/net/ethernet/intel/igc/igc.h68
-rw-r--r--drivers/net/ethernet/intel/igc/igc_base.h4
-rw-r--r--drivers/net/ethernet/intel/igc/igc_defines.h17
-rw-r--r--drivers/net/ethernet/intel/igc/igc_ethtool.c839
-rw-r--r--drivers/net/ethernet/intel/igc/igc_main.c447
-rw-r--r--drivers/net/ethernet/intel/igc/igc_regs.h16
-rw-r--r--drivers/net/ethernet/intel/ixgbe/ixgbe_main.c7
30 files changed, 3406 insertions, 748 deletions
diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c
index 0fd268070fb4..a65d5a9ba7db 100644
--- a/drivers/net/ethernet/intel/e100.c
+++ b/drivers/net/ethernet/intel/e100.c
@@ -2797,7 +2797,7 @@ static int e100_set_features(struct net_device *netdev,
netdev->features = features;
e100_exec_cb(nic, NULL, e100_configure);
- return 0;
+ return 1;
}
static const struct net_device_ops e100_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index 8fe9af0e2ab7..a7c76732849f 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -820,7 +820,7 @@ static int e1000_set_features(struct net_device *netdev,
else
e1000_reset(adapter);
- return 0;
+ return 1;
}
static const struct net_device_ops e1000_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 7acc61e4f645..745c1242a2d9 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -7003,7 +7003,7 @@ static int e1000_set_features(struct net_device *netdev,
else
e1000e_reset(adapter);
- return 0;
+ return 1;
}
static const struct net_device_ops e1000e_netdev_ops = {
@@ -7350,7 +7350,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP);
- if (pci_dev_run_wake(pdev))
+ if (pci_dev_run_wake(pdev) && hw->mac.type < e1000_pch_cnp)
pm_runtime_put_noidle(&pdev->dev);
return 0;
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 89440775aea1..b819689da7e2 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -42,10 +42,21 @@
extern const char ice_drv_ver[];
#define ICE_BAR0 0
-#define ICE_DFLT_NUM_DESC 128
#define ICE_REQ_DESC_MULTIPLE 32
#define ICE_MIN_NUM_DESC ICE_REQ_DESC_MULTIPLE
#define ICE_MAX_NUM_DESC 8160
+/* set default number of Rx/Tx descriptors to the minimum between
+ * ICE_MAX_NUM_DESC and the number of descriptors to fill up an entire page
+ */
+#define ICE_DFLT_NUM_RX_DESC min_t(u16, ICE_MAX_NUM_DESC, \
+ ALIGN(PAGE_SIZE / \
+ sizeof(union ice_32byte_rx_desc), \
+ ICE_REQ_DESC_MULTIPLE))
+#define ICE_DFLT_NUM_TX_DESC min_t(u16, ICE_MAX_NUM_DESC, \
+ ALIGN(PAGE_SIZE / \
+ sizeof(struct ice_tx_desc), \
+ ICE_REQ_DESC_MULTIPLE))
+
#define ICE_DFLT_TRAFFIC_CLASS BIT(0)
#define ICE_INT_NAME_STR_LEN (IFNAMSIZ + 16)
#define ICE_ETHTOOL_FWVER_LEN 32
@@ -114,6 +125,23 @@ extern const char ice_drv_ver[];
#define ice_for_each_q_vector(vsi, i) \
for ((i) = 0; (i) < (vsi)->num_q_vectors; (i)++)
+#define ICE_UCAST_PROMISC_BITS (ICE_PROMISC_UCAST_TX | ICE_PROMISC_MCAST_TX | \
+ ICE_PROMISC_UCAST_RX | ICE_PROMISC_MCAST_RX)
+
+#define ICE_UCAST_VLAN_PROMISC_BITS (ICE_PROMISC_UCAST_TX | \
+ ICE_PROMISC_MCAST_TX | \
+ ICE_PROMISC_UCAST_RX | \
+ ICE_PROMISC_MCAST_RX | \
+ ICE_PROMISC_VLAN_TX | \
+ ICE_PROMISC_VLAN_RX)
+
+#define ICE_MCAST_PROMISC_BITS (ICE_PROMISC_MCAST_TX | ICE_PROMISC_MCAST_RX)
+
+#define ICE_MCAST_VLAN_PROMISC_BITS (ICE_PROMISC_MCAST_TX | \
+ ICE_PROMISC_MCAST_RX | \
+ ICE_PROMISC_VLAN_TX | \
+ ICE_PROMISC_VLAN_RX)
+
struct ice_tc_info {
u16 qoffset;
u16 qcount_tx;
@@ -247,6 +275,7 @@ struct ice_vsi {
u8 irqs_ready;
u8 current_isup; /* Sync 'link up' logging */
u8 stat_offsets_loaded;
+ u8 vlan_ena;
/* queue information */
u8 tx_mapping_mode; /* ICE_MAP_MODE_[CONTIG|SCATTER] */
@@ -257,26 +286,33 @@ struct ice_vsi {
u16 num_txq; /* Used Tx queues */
u16 alloc_rxq; /* Allocated Rx queues */
u16 num_rxq; /* Used Rx queues */
- u16 num_desc;
+ u16 num_rx_desc;
+ u16 num_tx_desc;
struct ice_tc_cfg tc_cfg;
} ____cacheline_internodealigned_in_smp;
/* struct that defines an interrupt vector */
struct ice_q_vector {
struct ice_vsi *vsi;
- cpumask_t affinity_mask;
- struct napi_struct napi;
- struct ice_ring_container rx;
- struct ice_ring_container tx;
- struct irq_affinity_notify affinity_notify;
+
u16 v_idx; /* index in the vsi->q_vector array. */
- u8 num_ring_tx; /* total number of Tx rings in vector */
u8 num_ring_rx; /* total number of Rx rings in vector */
- char name[ICE_INT_NAME_STR_LEN];
+ u8 num_ring_tx; /* total number of Tx rings in vector */
+ u8 itr_countdown; /* when 0 should adjust adaptive ITR */
/* in usecs, need to use ice_intrl_to_usecs_reg() before writing this
* value to the device
*/
u8 intrl;
+
+ struct napi_struct napi;
+
+ struct ice_ring_container rx;
+ struct ice_ring_container tx;
+
+ cpumask_t affinity_mask;
+ struct irq_affinity_notify affinity_notify;
+
+ char name[ICE_INT_NAME_STR_LEN];
} ____cacheline_internodealigned_in_smp;
enum ice_pf_flags {
@@ -355,8 +391,9 @@ struct ice_netdev_priv {
* @vsi: pointer to vsi struct, can be NULL
* @q_vector: pointer to q_vector, can be NULL
*/
-static inline void ice_irq_dynamic_ena(struct ice_hw *hw, struct ice_vsi *vsi,
- struct ice_q_vector *q_vector)
+static inline void
+ice_irq_dynamic_ena(struct ice_hw *hw, struct ice_vsi *vsi,
+ struct ice_q_vector *q_vector)
{
u32 vector = (vsi && q_vector) ? vsi->hw_base_vector + q_vector->v_idx :
((struct ice_pf *)hw->back)->hw_oicr_idx;
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 242c78469181..8ff438968199 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -953,8 +953,9 @@ struct ice_aqc_set_phy_cfg_data {
__le64 phy_type_low; /* Use values from ICE_PHY_TYPE_LOW_* */
__le64 phy_type_high; /* Use values from ICE_PHY_TYPE_HIGH_* */
u8 caps;
-#define ICE_AQ_PHY_ENA_TX_PAUSE_ABILITY BIT(0)
-#define ICE_AQ_PHY_ENA_RX_PAUSE_ABILITY BIT(1)
+#define ICE_AQ_PHY_ENA_VALID_MASK ICE_M(0xef, 0)
+#define ICE_AQ_PHY_ENA_TX_PAUSE_ABILITY BIT(0)
+#define ICE_AQ_PHY_ENA_RX_PAUSE_ABILITY BIT(1)
#define ICE_AQ_PHY_ENA_LOW_POWER BIT(2)
#define ICE_AQ_PHY_ENA_LINK BIT(3)
#define ICE_AQ_PHY_ENA_AUTO_LINK_UPDT BIT(5)
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 63f003441300..5e7a31421c0d 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -262,7 +262,7 @@ static enum ice_media_type ice_get_media_type(struct ice_port_info *pi)
*
* Get Link Status (0x607). Returns the link status of the adapter.
*/
-static enum ice_status
+enum ice_status
ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse,
struct ice_link_status *link, struct ice_sq_cd *cd)
{
@@ -331,7 +331,7 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse,
/* flag cleared so calling functions don't call AQ again */
pi->phy.get_link_info = false;
- return status;
+ return 0;
}
/**
@@ -358,22 +358,22 @@ static void ice_init_flex_flags(struct ice_hw *hw, enum ice_rxdid prof_id)
*/
case ICE_RXDID_FLEX_NIC:
case ICE_RXDID_FLEX_NIC_2:
- ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_PKT_FRG,
- ICE_RXFLG_UDP_GRE, ICE_RXFLG_PKT_DSI,
- ICE_RXFLG_FIN, idx++);
+ ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_PKT_FRG,
+ ICE_FLG_UDP_GRE, ICE_FLG_PKT_DSI,
+ ICE_FLG_FIN, idx++);
/* flex flag 1 is not used for flexi-flag programming, skipping
* these four FLG64 bits.
*/
- ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_SYN, ICE_RXFLG_RST,
- ICE_RXFLG_PKT_DSI, ICE_RXFLG_PKT_DSI, idx++);
- ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_PKT_DSI,
- ICE_RXFLG_PKT_DSI, ICE_RXFLG_EVLAN_x8100,
- ICE_RXFLG_EVLAN_x9100, idx++);
- ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_VLAN_x8100,
- ICE_RXFLG_TNL_VLAN, ICE_RXFLG_TNL_MAC,
- ICE_RXFLG_TNL0, idx++);
- ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_TNL1, ICE_RXFLG_TNL2,
- ICE_RXFLG_PKT_DSI, ICE_RXFLG_PKT_DSI, idx);
+ ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_SYN, ICE_FLG_RST,
+ ICE_FLG_PKT_DSI, ICE_FLG_PKT_DSI, idx++);
+ ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_PKT_DSI,
+ ICE_FLG_PKT_DSI, ICE_FLG_EVLAN_x8100,
+ ICE_FLG_EVLAN_x9100, idx++);
+ ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_VLAN_x8100,
+ ICE_FLG_TNL_VLAN, ICE_FLG_TNL_MAC,
+ ICE_FLG_TNL0, idx++);
+ ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_TNL1, ICE_FLG_TNL2,
+ ICE_FLG_PKT_DSI, ICE_FLG_PKT_DSI, idx);
break;
default:
@@ -1100,8 +1100,9 @@ const struct ice_ctx_ele ice_tlan_ctx_info[] = {
*
* Dumps debug log about control command with descriptor contents.
*/
-void ice_debug_cq(struct ice_hw *hw, u32 __maybe_unused mask, void *desc,
- void *buf, u16 buf_len)
+void
+ice_debug_cq(struct ice_hw *hw, u32 __maybe_unused mask, void *desc, void *buf,
+ u16 buf_len)
{
struct ice_aq_desc *cq_desc = (struct ice_aq_desc *)desc;
u16 len;
@@ -1415,13 +1416,15 @@ void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res)
}
/**
- * ice_get_guar_num_vsi - determine number of guar VSI for a PF
+ * ice_get_num_per_func - determine number of resources per PF
* @hw: pointer to the hw structure
+ * @max: value to be evenly split between each PF
*
* Determine the number of valid functions by going through the bitmap returned
- * from parsing capabilities and use this to calculate the number of VSI per PF.
+ * from parsing capabilities and use this to calculate the number of resources
+ * per PF based on the max value passed in.
*/
-static u32 ice_get_guar_num_vsi(struct ice_hw *hw)
+static u32 ice_get_num_per_func(struct ice_hw *hw, u32 max)
{
u8 funcs;
@@ -1432,7 +1435,7 @@ static u32 ice_get_guar_num_vsi(struct ice_hw *hw)
if (!funcs)
return 0;
- return ICE_MAX_VSI / funcs;
+ return max / funcs;
}
/**
@@ -1512,7 +1515,8 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count,
"HW caps: Dev.VSI cnt = %d\n",
dev_p->num_vsi_allocd_to_host);
} else if (func_p) {
- func_p->guar_num_vsi = ice_get_guar_num_vsi(hw);
+ func_p->guar_num_vsi =
+ ice_get_num_per_func(hw, ICE_MAX_VSI);
ice_debug(hw, ICE_DBG_INIT,
"HW caps: Func.VSI cnt = %d\n",
number);
@@ -1617,8 +1621,8 @@ ice_aq_discover_caps(struct ice_hw *hw, void *buf, u16 buf_size, u32 *cap_count,
* @hw: pointer to the hardware structure
* @opc: capabilities type to discover - pass in the command opcode
*/
-static enum ice_status ice_discover_caps(struct ice_hw *hw,
- enum ice_adminq_opc opc)
+static enum ice_status
+ice_discover_caps(struct ice_hw *hw, enum ice_adminq_opc opc)
{
enum ice_status status;
u32 cap_count;
@@ -1929,6 +1933,15 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport,
if (!cfg)
return ICE_ERR_PARAM;
+ /* Ensure that only valid bits of cfg->caps can be turned on. */
+ if (cfg->caps & ~ICE_AQ_PHY_ENA_VALID_MASK) {
+ ice_debug(hw, ICE_DBG_PHY,
+ "Invalid bit is set in ice_aqc_set_phy_cfg_data->caps : 0x%x\n",
+ cfg->caps);
+
+ cfg->caps &= ICE_AQ_PHY_ENA_VALID_MASK;
+ }
+
ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_phy_cfg);
desc.params.set_phy.lport_num = lport;
desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
@@ -2027,8 +2040,10 @@ ice_set_fc(struct ice_port_info *pi, u8 *aq_failures, bool ena_auto_link_update)
/* clear the old pause settings */
cfg.caps = pcaps->caps & ~(ICE_AQC_PHY_EN_TX_LINK_PAUSE |
ICE_AQC_PHY_EN_RX_LINK_PAUSE);
+
/* set the new capabilities */
cfg.caps |= pause_mask;
+
/* If the capabilities have changed, then set the new config */
if (cfg.caps != pcaps->caps) {
int retry_count, retry_max = 10;
@@ -2136,6 +2151,32 @@ ice_aq_set_link_restart_an(struct ice_port_info *pi, bool ena_link,
}
/**
+ * ice_aq_set_event_mask
+ * @hw: pointer to the HW struct
+ * @port_num: port number of the physical function
+ * @mask: event mask to be set
+ * @cd: pointer to command details structure or NULL
+ *
+ * Set event mask (0x0613)
+ */
+enum ice_status
+ice_aq_set_event_mask(struct ice_hw *hw, u8 port_num, u16 mask,
+ struct ice_sq_cd *cd)
+{
+ struct ice_aqc_set_event_mask *cmd;
+ struct ice_aq_desc desc;
+
+ cmd = &desc.params.set_event_mask;
+
+ ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_event_mask);
+
+ cmd->lport_num = port_num;
+
+ cmd->event_mask = cpu_to_le16(mask);
+ return ice_aq_send_cmd(hw, &desc, NULL, 0, cd);
+}
+
+/**
* ice_aq_set_port_id_led
* @pi: pointer to the port information
* @is_orig_mode: is this LED set to original mode (by the net-list)
@@ -2534,8 +2575,8 @@ do_aq:
* @dest_ctx: the context to be written to
* @ce_info: a description of the struct to be filled
*/
-static void ice_write_byte(u8 *src_ctx, u8 *dest_ctx,
- const struct ice_ctx_ele *ce_info)
+static void
+ice_write_byte(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info)
{
u8 src_byte, dest_byte, mask;
u8 *from, *dest;
@@ -2573,8 +2614,8 @@ static void ice_write_byte(u8 *src_ctx, u8 *dest_ctx,
* @dest_ctx: the context to be written to
* @ce_info: a description of the struct to be filled
*/
-static void ice_write_word(u8 *src_ctx, u8 *dest_ctx,
- const struct ice_ctx_ele *ce_info)
+static void
+ice_write_word(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info)
{
u16 src_word, mask;
__le16 dest_word;
@@ -2616,8 +2657,8 @@ static void ice_write_word(u8 *src_ctx, u8 *dest_ctx,
* @dest_ctx: the context to be written to
* @ce_info: a description of the struct to be filled
*/
-static void ice_write_dword(u8 *src_ctx, u8 *dest_ctx,
- const struct ice_ctx_ele *ce_info)
+static void
+ice_write_dword(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info)
{
u32 src_dword, mask;
__le32 dest_dword;
@@ -2667,8 +2708,8 @@ static void ice_write_dword(u8 *src_ctx, u8 *dest_ctx,
* @dest_ctx: the context to be written to
* @ce_info: a description of the struct to be filled
*/
-static void ice_write_qword(u8 *src_ctx, u8 *dest_ctx,
- const struct ice_ctx_ele *ce_info)
+static void
+ice_write_qword(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info)
{
u64 src_qword, mask;
__le64 dest_qword;
@@ -2908,7 +2949,7 @@ ice_cfg_vsi_qs(struct ice_port_info *pi, u16 vsi_handle, u8 tc_bitmap,
mutex_lock(&pi->sched_lock);
- for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) {
+ ice_for_each_traffic_class(i) {
/* configuration is possible only if TC node is present */
if (!ice_sched_get_tc_node(pi, i))
continue;
@@ -3012,8 +3053,9 @@ void ice_replay_post(struct ice_hw *hw)
* @prev_stat: ptr to previous loaded stat value
* @cur_stat: ptr to current stat value
*/
-void ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg,
- bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat)
+void
+ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg,
+ bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat)
{
u64 new_data;
@@ -3043,8 +3085,9 @@ void ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg,
* @prev_stat: ptr to previous loaded stat value
* @cur_stat: ptr to current stat value
*/
-void ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
- u64 *prev_stat, u64 *cur_stat)
+void
+ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
+ u64 *prev_stat, u64 *cur_stat)
{
u32 new_data;
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index d7c7c2ed8823..fbdfdee353bc 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -9,8 +9,8 @@
#include "ice_switch.h"
#include <linux/avf/virtchnl.h>
-void ice_debug_cq(struct ice_hw *hw, u32 mask, void *desc, void *buf,
- u16 buf_len);
+void
+ice_debug_cq(struct ice_hw *hw, u32 mask, void *desc, void *buf, u16 buf_len);
enum ice_status ice_init_hw(struct ice_hw *hw);
void ice_deinit_hw(struct ice_hw *hw);
enum ice_status ice_check_reset(struct ice_hw *hw);
@@ -28,8 +28,8 @@ ice_acquire_res(struct ice_hw *hw, enum ice_aq_res_ids res,
enum ice_aq_res_access_type access, u32 timeout);
void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res);
enum ice_status ice_init_nvm(struct ice_hw *hw);
-enum ice_status ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words,
- u16 *data);
+enum ice_status
+ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data);
enum ice_status
ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq,
struct ice_aq_desc *desc, void *buf, u16 buf_size,
@@ -89,6 +89,12 @@ enum ice_status
ice_aq_set_link_restart_an(struct ice_port_info *pi, bool ena_link,
struct ice_sq_cd *cd);
enum ice_status
+ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse,
+ struct ice_link_status *link, struct ice_sq_cd *cd);
+enum ice_status
+ice_aq_set_event_mask(struct ice_hw *hw, u8 port_num, u16 mask,
+ struct ice_sq_cd *cd);
+enum ice_status
ice_aq_set_port_id_led(struct ice_port_info *pi, bool is_orig_mode,
struct ice_sq_cd *cd);
@@ -106,8 +112,10 @@ ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_qgrps,
enum ice_status ice_replay_vsi(struct ice_hw *hw, u16 vsi_handle);
void ice_replay_post(struct ice_hw *hw);
void ice_output_fw_log(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf);
-void ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg,
- bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat);
-void ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
- u64 *prev_stat, u64 *cur_stat);
+void
+ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg,
+ bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat);
+void
+ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
+ u64 *prev_stat, u64 *cur_stat);
#endif /* _ICE_COMMON_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index eb8d149e317c..4a1920e8f168 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -1156,8 +1156,9 @@ ice_get_settings_link_down(struct ethtool_link_ksettings *ks,
*
* Reports speed/duplex settings based on media_type
*/
-static int ice_get_link_ksettings(struct net_device *netdev,
- struct ethtool_link_ksettings *ks)
+static int
+ice_get_link_ksettings(struct net_device *netdev,
+ struct ethtool_link_ksettings *ks)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_link_status *hw_link_info;
@@ -1400,13 +1401,12 @@ ice_set_link_ksettings(struct net_device *netdev,
return -EOPNOTSUPP;
/* Check if this is lan vsi */
- for (idx = 0 ; idx < pf->num_alloc_vsi ; idx++) {
+ ice_for_each_vsi(pf, idx)
if (pf->vsi[idx]->type == ICE_VSI_PF) {
if (np->vsi != pf->vsi[idx])
return -EOPNOTSUPP;
break;
}
- }
if (p->phy.media_type != ICE_MEDIA_BASET &&
p->phy.media_type != ICE_MEDIA_FIBER &&
@@ -1566,8 +1566,9 @@ done:
*
* Returns Success if the command is supported.
*/
-static int ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
- u32 __always_unused *rule_locs)
+static int
+ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
+ u32 __always_unused *rule_locs)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
@@ -2024,8 +2025,9 @@ out:
* Returns -EINVAL if the table specifies an invalid queue id, otherwise
* returns 0 after programming the table.
*/
-static int ice_set_rxfh(struct net_device *netdev, const u32 *indir,
- const u8 *key, const u8 hfunc)
+static int
+ice_set_rxfh(struct net_device *netdev, const u32 *indir, const u8 *key,
+ const u8 hfunc)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
@@ -2180,8 +2182,9 @@ ice_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec)
return __ice_get_coalesce(netdev, ec, -1);
}
-static int ice_get_per_q_coalesce(struct net_device *netdev, u32 q_num,
- struct ethtool_coalesce *ec)
+static int
+ice_get_per_q_coalesce(struct net_device *netdev, u32 q_num,
+ struct ethtool_coalesce *ec)
{
return __ice_get_coalesce(netdev, ec, q_num);
}
@@ -2325,8 +2328,9 @@ ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec)
return __ice_set_coalesce(netdev, ec, -1);
}
-static int ice_set_per_q_coalesce(struct net_device *netdev, u32 q_num,
- struct ethtool_coalesce *ec)
+static int
+ice_set_per_q_coalesce(struct net_device *netdev, u32 q_num,
+ struct ethtool_coalesce *ec)
{
return __ice_set_coalesce(netdev, ec, q_num);
}
diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
index 6bf5cc064270..af6f32358363 100644
--- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
+++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
@@ -106,6 +106,16 @@
#define VPGEN_VFRTRIG_VFSWR_M BIT(0)
#define PFHMC_ERRORDATA 0x00520500
#define PFHMC_ERRORINFO 0x00520400
+#define GLINT_CTL 0x0016CC54
+#define GLINT_CTL_DIS_AUTOMASK_M BIT(0)
+#define GLINT_CTL_ITR_GRAN_200_S 16
+#define GLINT_CTL_ITR_GRAN_200_M ICE_M(0xF, 16)
+#define GLINT_CTL_ITR_GRAN_100_S 20
+#define GLINT_CTL_ITR_GRAN_100_M ICE_M(0xF, 20)
+#define GLINT_CTL_ITR_GRAN_50_S 24
+#define GLINT_CTL_ITR_GRAN_50_M ICE_M(0xF, 24)
+#define GLINT_CTL_ITR_GRAN_25_S 28
+#define GLINT_CTL_ITR_GRAN_25_M ICE_M(0xF, 28)
#define GLINT_DYN_CTL(_INT) (0x00160000 + ((_INT) * 4))
#define GLINT_DYN_CTL_INTENA_M BIT(0)
#define GLINT_DYN_CTL_CLEARPBA_M BIT(1)
@@ -168,6 +178,8 @@
#define VPINT_ALLOC_PCI_LAST_S 12
#define VPINT_ALLOC_PCI_LAST_M ICE_M(0x7FF, 12)
#define VPINT_ALLOC_PCI_VALID_M BIT(31)
+#define VPINT_MBX_CTL(_VSI) (0x0016A000 + ((_VSI) * 4))
+#define VPINT_MBX_CTL_CAUSE_ENA_M BIT(30)
#define GLLAN_RCTL_0 0x002941F8
#define QRX_CONTEXT(_i, _QRX) (0x00280000 + ((_i) * 8192 + (_QRX) * 4))
#define QRX_CTRL(_QRX) (0x00120000 + ((_QRX) * 4))
diff --git a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h
index ef4c79b5aa32..a8c3fe87d7aa 100644
--- a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h
+++ b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h
@@ -208,23 +208,23 @@ enum ice_flex_rx_mdid {
ICE_RX_MDID_HASH_HIGH,
};
-/* Rx Flag64 packet flag bits */
-enum ice_rx_flg64_bits {
- ICE_RXFLG_PKT_DSI = 0,
- ICE_RXFLG_EVLAN_x8100 = 15,
- ICE_RXFLG_EVLAN_x9100,
- ICE_RXFLG_VLAN_x8100,
- ICE_RXFLG_TNL_MAC = 22,
- ICE_RXFLG_TNL_VLAN,
- ICE_RXFLG_PKT_FRG,
- ICE_RXFLG_FIN = 32,
- ICE_RXFLG_SYN,
- ICE_RXFLG_RST,
- ICE_RXFLG_TNL0 = 38,
- ICE_RXFLG_TNL1,
- ICE_RXFLG_TNL2,
- ICE_RXFLG_UDP_GRE,
- ICE_RXFLG_RSVD = 63
+/* RX/TX Flag64 packet flag bits */
+enum ice_flg64_bits {
+ ICE_FLG_PKT_DSI = 0,
+ ICE_FLG_EVLAN_x8100 = 15,
+ ICE_FLG_EVLAN_x9100,
+ ICE_FLG_VLAN_x8100,
+ ICE_FLG_TNL_MAC = 22,
+ ICE_FLG_TNL_VLAN,
+ ICE_FLG_PKT_FRG,
+ ICE_FLG_FIN = 32,
+ ICE_FLG_SYN,
+ ICE_FLG_RST,
+ ICE_FLG_TNL0 = 38,
+ ICE_FLG_TNL1,
+ ICE_FLG_TNL2,
+ ICE_FLG_UDP_GRE,
+ ICE_FLG_RSVD = 63
};
/* for ice_32byte_rx_flex_desc.ptype_flexi_flags0 member */
@@ -342,12 +342,12 @@ enum ice_tx_desc_cmd_bits {
ICE_TX_DESC_CMD_EOP = 0x0001,
ICE_TX_DESC_CMD_RS = 0x0002,
ICE_TX_DESC_CMD_IL2TAG1 = 0x0008,
- ICE_TX_DESC_CMD_IIPT_IPV6 = 0x0020, /* 2 BITS */
- ICE_TX_DESC_CMD_IIPT_IPV4 = 0x0040, /* 2 BITS */
- ICE_TX_DESC_CMD_IIPT_IPV4_CSUM = 0x0060, /* 2 BITS */
- ICE_TX_DESC_CMD_L4T_EOFT_TCP = 0x0100, /* 2 BITS */
- ICE_TX_DESC_CMD_L4T_EOFT_SCTP = 0x0200, /* 2 BITS */
- ICE_TX_DESC_CMD_L4T_EOFT_UDP = 0x0300, /* 2 BITS */
+ ICE_TX_DESC_CMD_IIPT_IPV6 = 0x0020,
+ ICE_TX_DESC_CMD_IIPT_IPV4 = 0x0040,
+ ICE_TX_DESC_CMD_IIPT_IPV4_CSUM = 0x0060,
+ ICE_TX_DESC_CMD_L4T_EOFT_TCP = 0x0100,
+ ICE_TX_DESC_CMD_L4T_EOFT_SCTP = 0x0200,
+ ICE_TX_DESC_CMD_L4T_EOFT_UDP = 0x0300,
};
#define ICE_TXD_QW1_OFFSET_S 16
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index fa61203bee26..45e361f72057 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -175,17 +175,14 @@ static int ice_pf_rxq_wait(struct ice_pf *pf, int pf_q, bool ena)
int i;
for (i = 0; i < ICE_Q_WAIT_MAX_RETRY; i++) {
- u32 rx_reg = rd32(&pf->hw, QRX_CTRL(pf_q));
-
- if (ena == !!(rx_reg & QRX_CTRL_QENA_STAT_M))
- break;
+ if (ena == !!(rd32(&pf->hw, QRX_CTRL(pf_q)) &
+ QRX_CTRL_QENA_STAT_M))
+ return 0;
usleep_range(20, 40);
}
- if (i >= ICE_Q_WAIT_MAX_RETRY)
- return -ETIMEDOUT;
- return 0;
+ return -ETIMEDOUT;
}
/**
@@ -279,25 +276,50 @@ err_txrings:
}
/**
- * ice_vsi_set_num_qs - Set num queues, descriptors and vectors for a VSI
+ * ice_vsi_set_num_desc - Set number of descriptors for queues on this VSI
* @vsi: the VSI being configured
+ */
+static void ice_vsi_set_num_desc(struct ice_vsi *vsi)
+{
+ switch (vsi->type) {
+ case ICE_VSI_PF:
+ vsi->num_rx_desc = ICE_DFLT_NUM_RX_DESC;
+ vsi->num_tx_desc = ICE_DFLT_NUM_TX_DESC;
+ break;
+ default:
+ dev_dbg(&vsi->back->pdev->dev,
+ "Not setting number of Tx/Rx descriptors for VSI type %d\n",
+ vsi->type);
+ break;
+ }
+}
+
+/**
+ * ice_vsi_set_num_qs - Set number of queues, descriptors and vectors for a VSI
+ * @vsi: the VSI being configured
+ * @vf_id: Id of the VF being configured
*
* Return 0 on success and a negative value on error
*/
-static void ice_vsi_set_num_qs(struct ice_vsi *vsi)
+static void ice_vsi_set_num_qs(struct ice_vsi *vsi, u16 vf_id)
{
struct ice_pf *pf = vsi->back;
+ struct ice_vf *vf = NULL;
+
+ if (vsi->type == ICE_VSI_VF)
+ vsi->vf_id = vf_id;
+
switch (vsi->type) {
case ICE_VSI_PF:
vsi->alloc_txq = pf->num_lan_tx;
vsi->alloc_rxq = pf->num_lan_rx;
- vsi->num_desc = ALIGN(ICE_DFLT_NUM_DESC, ICE_REQ_DESC_MULTIPLE);
vsi->num_q_vectors = max_t(int, pf->num_lan_rx, pf->num_lan_tx);
break;
case ICE_VSI_VF:
- vsi->alloc_txq = pf->num_vf_qps;
- vsi->alloc_rxq = pf->num_vf_qps;
+ vf = &pf->vf[vsi->vf_id];
+ vsi->alloc_txq = vf->num_vf_qs;
+ vsi->alloc_rxq = vf->num_vf_qs;
/* pf->num_vf_msix includes (VF miscellaneous vector +
* data queue interrupts). Since vsi->num_q_vectors is number
* of queues vectors, subtract 1 from the original vector
@@ -310,6 +332,8 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi)
vsi->type);
break;
}
+
+ ice_vsi_set_num_desc(vsi);
}
/**
@@ -455,10 +479,12 @@ static irqreturn_t ice_msix_clean_rings(int __always_unused irq, void *data)
* ice_vsi_alloc - Allocates the next available struct VSI in the PF
* @pf: board private structure
* @type: type of VSI
+ * @vf_id: Id of the VF being configured
*
* returns a pointer to a VSI on success, NULL on failure.
*/
-static struct ice_vsi *ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type)
+static struct ice_vsi *
+ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type, u16 vf_id)
{
struct ice_vsi *vsi = NULL;
@@ -484,7 +510,10 @@ static struct ice_vsi *ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type)
vsi->idx = pf->next_vsi;
vsi->work_lmt = ICE_DFLT_IRQ_WORK;
- ice_vsi_set_num_qs(vsi);
+ if (type == ICE_VSI_VF)
+ ice_vsi_set_num_qs(vsi, vf_id);
+ else
+ ice_vsi_set_num_qs(vsi, ICE_INVAL_VFID);
switch (vsi->type) {
case ICE_VSI_PF:
@@ -579,11 +608,10 @@ err_scatter:
/**
* __ice_vsi_get_qs - helper function for assigning queues from PF to VSI
- * @qs_cfg: gathered variables needed for PF->VSI queues assignment
+ * @qs_cfg: gathered variables needed for pf->vsi queues assignment
*
- * This is an internal function for assigning queues from the PF to VSI and
- * initially tries to find contiguous space. If it is not successful to find
- * contiguous space, then it tries with the scatter approach.
+ * This function first tries to find contiguous space. If it is not successful,
+ * it tries with the scatter approach.
*
* Return 0 on success and -ENOMEM in case of no left space in PF queue bitmap
*/
@@ -827,7 +855,7 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt)
/* find the (rounded up) power-of-2 of qcount */
pow = order_base_2(qcount_rx);
- for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) {
+ ice_for_each_traffic_class(i) {
if (!(vsi->tc_cfg.ena_tc & BIT(i))) {
/* TC is not enabled */
vsi->tc_cfg.tc_info[i].qoffset = 0;
@@ -852,7 +880,18 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt)
tx_count += tx_numq_tc;
ctxt->info.tc_mapping[i] = cpu_to_le16(qmap);
}
- vsi->num_rxq = offset;
+
+ /* if offset is non-zero, means it is calculated correctly based on
+ * enabled TCs for a given VSI otherwise qcount_rx will always
+ * be correct and non-zero because it is based off - VSI's
+ * allocated Rx queues which is at least 1 (hence qcount_tx will be
+ * at least 1)
+ */
+ if (offset)
+ vsi->num_rxq = offset;
+ else
+ vsi->num_rxq = qcount_rx;
+
vsi->num_txq = tx_count;
if (vsi->type == ICE_VSI_VF && vsi->num_txq != vsi->num_rxq) {
@@ -923,6 +962,7 @@ static int ice_vsi_init(struct ice_vsi *vsi)
if (!ctxt)
return -ENOMEM;
+ ctxt->info = vsi->info;
switch (vsi->type) {
case ICE_VSI_PF:
ctxt->flags = ICE_AQ_VSI_TYPE_PF;
@@ -948,6 +988,14 @@ static int ice_vsi_init(struct ice_vsi *vsi)
ctxt->info.sw_id = vsi->port_info->sw_id;
ice_vsi_setup_q_map(vsi, ctxt);
+ /* Enable MAC Antispoof with new VSI being initialized or updated */
+ if (vsi->type == ICE_VSI_VF && pf->vf[vsi->vf_id].spoofchk) {
+ ctxt->info.valid_sections |=
+ cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID);
+ ctxt->info.sec_flags |=
+ ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF;
+ }
+
ret = ice_add_vsi(hw, vsi->idx, ctxt, NULL);
if (ret) {
dev_err(&pf->pdev->dev,
@@ -1215,7 +1263,7 @@ static int ice_vsi_alloc_rings(struct ice_vsi *vsi)
ring->ring_active = false;
ring->vsi = vsi;
ring->dev = &pf->pdev->dev;
- ring->count = vsi->num_desc;
+ ring->count = vsi->num_tx_desc;
vsi->tx_rings[i] = ring;
}
@@ -1234,7 +1282,7 @@ static int ice_vsi_alloc_rings(struct ice_vsi *vsi)
ring->vsi = vsi;
ring->netdev = vsi->netdev;
ring->dev = &pf->pdev->dev;
- ring->count = vsi->num_desc;
+ ring->count = vsi->num_rx_desc;
vsi->rx_rings[i] = ring;
}
@@ -1640,7 +1688,7 @@ ice_vsi_cfg_txqs(struct ice_vsi *vsi, struct ice_ring **rings, int offset)
num_q_grps = 1;
/* set up and configure the Tx queues for each enabled TC */
- for (tc = 0; tc < ICE_MAX_TRAFFIC_CLASS; tc++) {
+ ice_for_each_traffic_class(tc) {
if (!(vsi->tc_cfg.ena_tc & BIT(tc)))
break;
@@ -1717,6 +1765,37 @@ static u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran)
}
/**
+ * ice_cfg_itr_gran - set the ITR granularity to 2 usecs if not already set
+ * @hw: board specific structure
+ */
+static void ice_cfg_itr_gran(struct ice_hw *hw)
+{
+ u32 regval = rd32(hw, GLINT_CTL);
+
+ /* no need to update global register if ITR gran is already set */
+ if (!(regval & GLINT_CTL_DIS_AUTOMASK_M) &&
+ (((regval & GLINT_CTL_ITR_GRAN_200_M) >>
+ GLINT_CTL_ITR_GRAN_200_S) == ICE_ITR_GRAN_US) &&
+ (((regval & GLINT_CTL_ITR_GRAN_100_M) >>
+ GLINT_CTL_ITR_GRAN_100_S) == ICE_ITR_GRAN_US) &&
+ (((regval & GLINT_CTL_ITR_GRAN_50_M) >>
+ GLINT_CTL_ITR_GRAN_50_S) == ICE_ITR_GRAN_US) &&
+ (((regval & GLINT_CTL_ITR_GRAN_25_M) >>
+ GLINT_CTL_ITR_GRAN_25_S) == ICE_ITR_GRAN_US))
+ return;
+
+ regval = ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_200_S) &
+ GLINT_CTL_ITR_GRAN_200_M) |
+ ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_100_S) &
+ GLINT_CTL_ITR_GRAN_100_M) |
+ ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_50_S) &
+ GLINT_CTL_ITR_GRAN_50_M) |
+ ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_25_S) &
+ GLINT_CTL_ITR_GRAN_25_M);
+ wr32(hw, GLINT_CTL, regval);
+}
+
+/**
* ice_cfg_itr - configure the initial interrupt throttle values
* @hw: pointer to the HW structure
* @q_vector: interrupt vector that's being configured
@@ -1728,6 +1807,8 @@ static u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran)
static void
ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector, u16 vector)
{
+ ice_cfg_itr_gran(hw);
+
if (q_vector->num_ring_rx) {
struct ice_ring_container *rc = &q_vector->rx;
@@ -1738,7 +1819,6 @@ ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector, u16 vector)
rc->target_itr = ITR_TO_REG(rc->itr_setting);
rc->next_update = jiffies + 1;
rc->current_itr = rc->target_itr;
- rc->latency_range = ICE_LOW_LATENCY;
wr32(hw, GLINT_ITR(rc->itr_idx, vector),
ITR_REG_ALIGN(rc->current_itr) >> ICE_ITR_GRAN_S);
}
@@ -1753,7 +1833,6 @@ ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector, u16 vector)
rc->target_itr = ITR_TO_REG(rc->itr_setting);
rc->next_update = jiffies + 1;
rc->current_itr = rc->target_itr;
- rc->latency_range = ICE_LOW_LATENCY;
wr32(hw, GLINT_ITR(rc->itr_idx, vector),
ITR_REG_ALIGN(rc->current_itr) >> ICE_ITR_GRAN_S);
}
@@ -2025,8 +2104,9 @@ err_alloc_q_ids:
* @rst_src: reset source
* @rel_vmvf_num: Relative id of VF/VM
*/
-int ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi,
- enum ice_disq_rst_src rst_src, u16 rel_vmvf_num)
+int
+ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src,
+ u16 rel_vmvf_num)
{
return ice_vsi_stop_tx_rings(vsi, rst_src, rel_vmvf_num, vsi->tx_rings,
0);
@@ -2036,10 +2116,11 @@ int ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi,
* ice_cfg_vlan_pruning - enable or disable VLAN pruning on the VSI
* @vsi: VSI to enable or disable VLAN pruning on
* @ena: set to true to enable VLAN pruning and false to disable it
+ * @vlan_promisc: enable valid security flags if not in VLAN promiscuous mode
*
* returns 0 if VSI is updated, negative otherwise
*/
-int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena)
+int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc)
{
struct ice_vsi_ctx *ctxt;
struct device *dev;
@@ -2067,8 +2148,10 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena)
ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
}
- ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID |
- ICE_AQ_VSI_PROP_SW_VALID);
+ if (!vlan_promisc)
+ ctxt->info.valid_sections =
+ cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID |
+ ICE_AQ_VSI_PROP_SW_VALID);
status = ice_update_vsi(&vsi->back->hw, vsi->idx, ctxt, NULL);
if (status) {
@@ -2112,7 +2195,11 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi,
struct ice_vsi *vsi;
int ret, i;
- vsi = ice_vsi_alloc(pf, type);
+ if (type == ICE_VSI_VF)
+ vsi = ice_vsi_alloc(pf, type, vf_id);
+ else
+ vsi = ice_vsi_alloc(pf, type, ICE_INVAL_VFID);
+
if (!vsi) {
dev_err(dev, "could not allocate VSI\n");
return NULL;
@@ -2596,6 +2683,7 @@ int ice_vsi_release(struct ice_vsi *vsi)
int ice_vsi_rebuild(struct ice_vsi *vsi)
{
u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 };
+ struct ice_vf *vf = NULL;
struct ice_pf *pf;
int ret, i;
@@ -2603,16 +2691,38 @@ int ice_vsi_rebuild(struct ice_vsi *vsi)
return -EINVAL;
pf = vsi->back;
+ if (vsi->type == ICE_VSI_VF)
+ vf = &pf->vf[vsi->vf_id];
+
ice_rm_vsi_lan_cfg(vsi->port_info, vsi->idx);
ice_vsi_free_q_vectors(vsi);
- ice_free_res(vsi->back->sw_irq_tracker, vsi->sw_base_vector, vsi->idx);
- ice_free_res(vsi->back->hw_irq_tracker, vsi->hw_base_vector, vsi->idx);
- vsi->sw_base_vector = 0;
+
+ if (vsi->type != ICE_VSI_VF) {
+ /* reclaim SW interrupts back to the common pool */
+ ice_free_res(pf->sw_irq_tracker, vsi->sw_base_vector, vsi->idx);
+ pf->num_avail_sw_msix += vsi->num_q_vectors;
+ vsi->sw_base_vector = 0;
+ /* reclaim HW interrupts back to the common pool */
+ ice_free_res(pf->hw_irq_tracker, vsi->hw_base_vector,
+ vsi->idx);
+ pf->num_avail_hw_msix += vsi->num_q_vectors;
+ } else {
+ /* Reclaim VF resources back to the common pool for reset and
+ * and rebuild, with vector reassignment
+ */
+ ice_free_res(pf->hw_irq_tracker, vf->first_vector_idx,
+ vsi->idx);
+ pf->num_avail_hw_msix += pf->num_vf_msix;
+ }
vsi->hw_base_vector = 0;
+
ice_vsi_clear_rings(vsi);
ice_vsi_free_arrays(vsi, false);
ice_dev_onetime_setup(&vsi->back->hw);
- ice_vsi_set_num_qs(vsi);
+ if (vsi->type == ICE_VSI_VF)
+ ice_vsi_set_num_qs(vsi, vf->vf_id);
+ else
+ ice_vsi_set_num_qs(vsi, ICE_INVAL_VFID);
ice_vsi_set_tc_cfg(vsi);
/* Initialize VSI struct elements and create VSI in FW */
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h
index 7988a53729a9..519ef59e9e43 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_lib.h
@@ -35,7 +35,7 @@ int
ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src,
u16 rel_vmvf_num);
-int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena);
+int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc);
void ice_vsi_delete(struct ice_vsi *vsi);
@@ -70,8 +70,6 @@ void ice_vsi_free_rx_rings(struct ice_vsi *vsi);
void ice_vsi_free_tx_rings(struct ice_vsi *vsi);
-int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc);
-
int ice_vsi_manage_rss_lut(struct ice_vsi *vsi, bool ena);
#endif /* !_ICE_LIB_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 47cc3f905b7f..f7073e046979 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -168,6 +168,39 @@ static bool ice_vsi_fltr_changed(struct ice_vsi *vsi)
}
/**
+ * ice_cfg_promisc - Enable or disable promiscuous mode for a given PF
+ * @vsi: the VSI being configured
+ * @promisc_m: mask of promiscuous config bits
+ * @set_promisc: enable or disable promisc flag request
+ *
+ */
+static int ice_cfg_promisc(struct ice_vsi *vsi, u8 promisc_m, bool set_promisc)
+{
+ struct ice_hw *hw = &vsi->back->hw;
+ enum ice_status status = 0;
+
+ if (vsi->type != ICE_VSI_PF)
+ return 0;
+
+ if (vsi->vlan_ena) {
+ status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m,
+ set_promisc);
+ } else {
+ if (set_promisc)
+ status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m,
+ 0);
+ else
+ status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m,
+ 0);
+ }
+
+ if (status)
+ return -EIO;
+
+ return 0;
+}
+
+/**
* ice_vsi_sync_fltr - Update the VSI filter list to the HW
* @vsi: ptr to the VSI
*
@@ -182,6 +215,7 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi)
struct ice_hw *hw = &pf->hw;
enum ice_status status = 0;
u32 changed_flags = 0;
+ u8 promisc_m;
int err = 0;
if (!vsi->netdev)
@@ -226,7 +260,11 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi)
/* Add mac addresses in the sync list */
status = ice_add_mac(hw, &vsi->tmp_sync_list);
ice_free_fltr_list(dev, &vsi->tmp_sync_list);
- if (status) {
+ /* If filter is added successfully or already exists, do not go into
+ * 'if' condition and report it as error. Instead continue processing
+ * rest of the function.
+ */
+ if (status && status != ICE_ERR_ALREADY_EXISTS) {
netdev_err(netdev, "Failed to add MAC filters\n");
/* If there is no more space for new umac filters, vsi
* should go into promiscuous mode. There should be some
@@ -245,8 +283,35 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi)
}
}
/* check for changes in promiscuous modes */
- if (changed_flags & IFF_ALLMULTI)
- netdev_warn(netdev, "Unsupported configuration\n");
+ if (changed_flags & IFF_ALLMULTI) {
+ if (vsi->current_netdev_flags & IFF_ALLMULTI) {
+ if (vsi->vlan_ena)
+ promisc_m = ICE_MCAST_VLAN_PROMISC_BITS;
+ else
+ promisc_m = ICE_MCAST_PROMISC_BITS;
+
+ err = ice_cfg_promisc(vsi, promisc_m, true);
+ if (err) {
+ netdev_err(netdev, "Error setting Multicast promiscuous mode on VSI %i\n",
+ vsi->vsi_num);
+ vsi->current_netdev_flags &= ~IFF_ALLMULTI;
+ goto out_promisc;
+ }
+ } else if (!(vsi->current_netdev_flags & IFF_ALLMULTI)) {
+ if (vsi->vlan_ena)
+ promisc_m = ICE_MCAST_VLAN_PROMISC_BITS;
+ else
+ promisc_m = ICE_MCAST_PROMISC_BITS;
+
+ err = ice_cfg_promisc(vsi, promisc_m, false);
+ if (err) {
+ netdev_err(netdev, "Error clearing Multicast promiscuous mode on VSI %i\n",
+ vsi->vsi_num);
+ vsi->current_netdev_flags |= IFF_ALLMULTI;
+ goto out_promisc;
+ }
+ }
+ }
if (((changed_flags & IFF_PROMISC) || promisc_forced_on) ||
test_bit(ICE_VSI_FLAG_PROMISC_CHANGED, vsi->flags)) {
@@ -322,7 +387,7 @@ static void ice_sync_fltr_subtask(struct ice_pf *pf)
clear_bit(ICE_FLAG_FLTR_SYNC, pf->flags);
- for (v = 0; v < pf->num_alloc_vsi; v++)
+ ice_for_each_vsi(pf, v)
if (pf->vsi[v] && ice_vsi_fltr_changed(pf->vsi[v]) &&
ice_vsi_sync_fltr(pf->vsi[v])) {
/* come back and try again later */
@@ -342,6 +407,10 @@ ice_prepare_for_reset(struct ice_pf *pf)
{
struct ice_hw *hw = &pf->hw;
+ /* already prepared for reset */
+ if (test_bit(__ICE_PREPARED_FOR_RESET, pf->state))
+ return;
+
/* Notify VFs of impending reset */
if (ice_check_sq_alive(hw, &hw->mailboxq))
ice_vc_notify_reset(pf);
@@ -394,6 +463,7 @@ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type)
ice_rebuild(pf);
clear_bit(__ICE_PREPARED_FOR_RESET, pf->state);
clear_bit(__ICE_PFR_REQ, pf->state);
+ ice_reset_all_vfs(pf, true);
}
}
@@ -416,10 +486,15 @@ static void ice_reset_subtask(struct ice_pf *pf)
* for the reset now), poll for reset done, rebuild and return.
*/
if (test_bit(__ICE_RESET_OICR_RECV, pf->state)) {
- clear_bit(__ICE_GLOBR_RECV, pf->state);
- clear_bit(__ICE_CORER_RECV, pf->state);
- if (!test_bit(__ICE_PREPARED_FOR_RESET, pf->state))
- ice_prepare_for_reset(pf);
+ /* Perform the largest reset requested */
+ if (test_and_clear_bit(__ICE_CORER_RECV, pf->state))
+ reset_type = ICE_RESET_CORER;
+ if (test_and_clear_bit(__ICE_GLOBR_RECV, pf->state))
+ reset_type = ICE_RESET_GLOBR;
+ /* return if no valid reset type requested */
+ if (reset_type == ICE_RESET_INVAL)
+ return;
+ ice_prepare_for_reset(pf);
/* make sure we are ready to rebuild */
if (ice_check_reset(&pf->hw)) {
@@ -436,6 +511,7 @@ static void ice_reset_subtask(struct ice_pf *pf)
clear_bit(__ICE_PFR_REQ, pf->state);
clear_bit(__ICE_CORER_REQ, pf->state);
clear_bit(__ICE_GLOBR_REQ, pf->state);
+ ice_reset_all_vfs(pf, true);
}
return;
@@ -519,6 +595,9 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup)
case ICE_FC_RX_PAUSE:
fc = "RX";
break;
+ case ICE_FC_NONE:
+ fc = "None";
+ break;
default:
fc = "Unknown";
break;
@@ -635,19 +714,70 @@ static void ice_watchdog_subtask(struct ice_pf *pf)
pf->serv_tmr_prev = jiffies;
- if (ice_link_event(pf, pf->hw.port_info))
- dev_dbg(&pf->pdev->dev, "ice_link_event failed\n");
-
/* Update the stats for active netdevs so the network stack
* can look at updated numbers whenever it cares to
*/
ice_update_pf_stats(pf);
- for (i = 0; i < pf->num_alloc_vsi; i++)
+ ice_for_each_vsi(pf, i)
if (pf->vsi[i] && pf->vsi[i]->netdev)
ice_update_vsi_stats(pf->vsi[i]);
}
/**
+ * ice_init_link_events - enable/initialize link events
+ * @pi: pointer to the port_info instance
+ *
+ * Returns -EIO on failure, 0 on success
+ */
+static int ice_init_link_events(struct ice_port_info *pi)
+{
+ u16 mask;
+
+ mask = ~((u16)(ICE_AQ_LINK_EVENT_UPDOWN | ICE_AQ_LINK_EVENT_MEDIA_NA |
+ ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL));
+
+ if (ice_aq_set_event_mask(pi->hw, pi->lport, mask, NULL)) {
+ dev_dbg(ice_hw_to_dev(pi->hw),
+ "Failed to set link event mask for port %d\n",
+ pi->lport);
+ return -EIO;
+ }
+
+ if (ice_aq_get_link_info(pi, true, NULL, NULL)) {
+ dev_dbg(ice_hw_to_dev(pi->hw),
+ "Failed to enable link events for port %d\n",
+ pi->lport);
+ return -EIO;
+ }
+
+ return 0;
+}
+
+/**
+ * ice_handle_link_event - handle link event via ARQ
+ * @pf: pf that the link event is associated with
+ *
+ * Return -EINVAL if port_info is null
+ * Return status on success
+ */
+static int ice_handle_link_event(struct ice_pf *pf)
+{
+ struct ice_port_info *port_info;
+ int status;
+
+ port_info = pf->hw.port_info;
+ if (!port_info)
+ return -EINVAL;
+
+ status = ice_link_event(pf, port_info);
+ if (status)
+ dev_dbg(&pf->pdev->dev,
+ "Could not process link event, error %d\n", status);
+
+ return status;
+}
+
+/**
* __ice_clean_ctrlq - helper function to clean controlq rings
* @pf: ptr to struct ice_pf
* @q_type: specific Control queue type
@@ -750,6 +880,11 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type)
opcode = le16_to_cpu(event.desc.opcode);
switch (opcode) {
+ case ice_aqc_opc_get_link_status:
+ if (ice_handle_link_event(pf))
+ dev_err(&pf->pdev->dev,
+ "Could not handle link event\n");
+ break;
case ice_mbx_opc_send_msg_to_pf:
ice_vc_process_vf_msg(pf, &event);
break;
@@ -877,6 +1012,18 @@ static void ice_service_task_stop(struct ice_pf *pf)
}
/**
+ * ice_service_task_restart - restart service task and schedule works
+ * @pf: board private structure
+ *
+ * This function is needed for suspend and resume works (e.g WoL scenario)
+ */
+static void ice_service_task_restart(struct ice_pf *pf)
+{
+ clear_bit(__ICE_SERVICE_DIS, pf->state);
+ ice_service_task_schedule(pf);
+}
+
+/**
* ice_service_timer - timer callback to schedule service task
* @t: pointer to timer_list
*/
@@ -1111,8 +1258,9 @@ static void ice_set_ctrlq_len(struct ice_hw *hw)
* This is a callback function used by the irq_set_affinity_notifier function
* so that we may register to receive changes to the irq affinity masks.
*/
-static void ice_irq_affinity_notify(struct irq_affinity_notify *notify,
- const cpumask_t *mask)
+static void
+ice_irq_affinity_notify(struct irq_affinity_notify *notify,
+ const cpumask_t *mask)
{
struct ice_q_vector *q_vector =
container_of(notify, struct ice_q_vector, affinity_notify);
@@ -1184,10 +1332,9 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename)
/* skip this unused q_vector */
continue;
}
- err = devm_request_irq(&pf->pdev->dev,
- pf->msix_entries[base + vector].vector,
- vsi->irq_handler, 0, q_vector->name,
- q_vector);
+ err = devm_request_irq(&pf->pdev->dev, irq_num,
+ vsi->irq_handler, 0,
+ q_vector->name, q_vector);
if (err) {
netdev_err(vsi->netdev,
"MSIX request_irq failed, error: %d\n", err);
@@ -1656,11 +1803,13 @@ ice_pf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi)
*
* net_device_ops implementation for adding vlan ids
*/
-static int ice_vlan_rx_add_vid(struct net_device *netdev,
- __always_unused __be16 proto, u16 vid)
+static int
+ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto,
+ u16 vid)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
+ int ret;
if (vid >= VLAN_N_VID) {
netdev_err(netdev, "VLAN id requested %d is out of range %d\n",
@@ -1673,8 +1822,7 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
/* Enable VLAN pruning when VLAN 0 is added */
if (unlikely(!vid)) {
- int ret = ice_cfg_vlan_pruning(vsi, true);
-
+ ret = ice_cfg_vlan_pruning(vsi, true, false);
if (ret)
return ret;
}
@@ -1683,7 +1831,13 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
* needed to continue allowing all untagged packets since VLAN prune
* list is applied to all packets by the switch
*/
- return ice_vsi_add_vlan(vsi, vid);
+ ret = ice_vsi_add_vlan(vsi, vid);
+ if (!ret) {
+ vsi->vlan_ena = true;
+ set_bit(ICE_VSI_FLAG_VLAN_FLTR_CHANGED, vsi->flags);
+ }
+
+ return ret;
}
/**
@@ -1694,12 +1848,13 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
*
* net_device_ops implementation for removing vlan ids
*/
-static int ice_vlan_rx_kill_vid(struct net_device *netdev,
- __always_unused __be16 proto, u16 vid)
+static int
+ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto,
+ u16 vid)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
- int status;
+ int ret;
if (vsi->info.pvid)
return -EINVAL;
@@ -1707,15 +1862,17 @@ static int ice_vlan_rx_kill_vid(struct net_device *netdev,
/* Make sure ice_vsi_kill_vlan is successful before updating VLAN
* information
*/
- status = ice_vsi_kill_vlan(vsi, vid);
- if (status)
- return status;
+ ret = ice_vsi_kill_vlan(vsi, vid);
+ if (ret)
+ return ret;
/* Disable VLAN pruning when VLAN 0 is removed */
if (unlikely(!vid))
- status = ice_cfg_vlan_pruning(vsi, false);
+ ret = ice_cfg_vlan_pruning(vsi, false, false);
- return status;
+ vsi->vlan_ena = false;
+ set_bit(ICE_VSI_FLAG_VLAN_FLTR_CHANGED, vsi->flags);
+ return ret;
}
/**
@@ -2033,23 +2190,6 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf)
}
/**
- * ice_verify_itr_gran - verify driver's assumption of ITR granularity
- * @pf: pointer to the PF structure
- *
- * There is no error returned here because the driver will be able to handle a
- * different ITR granularity, but interrupt moderation will not be accurate if
- * the driver's assumptions are not verified. This assumption is made so we can
- * use constants in the hot path instead of accessing structure members.
- */
-static void ice_verify_itr_gran(struct ice_pf *pf)
-{
- if (pf->hw.itr_gran != (ICE_ITR_GRAN_S << 1))
- dev_warn(&pf->pdev->dev,
- "%d ITR granularity assumption is invalid, actual ITR granularity is %d. Interrupt moderation will be inaccurate!\n",
- (ICE_ITR_GRAN_S << 1), pf->hw.itr_gran);
-}
-
-/**
* ice_verify_cacheline_size - verify driver's assumption of 64 Byte cache lines
* @pf: pointer to the PF structure
*
@@ -2072,9 +2212,10 @@ static void ice_verify_cacheline_size(struct ice_pf *pf)
*
* Returns 0 on success, negative on failure
*/
-static int ice_probe(struct pci_dev *pdev,
- const struct pci_device_id __always_unused *ent)
+static int
+ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent)
{
+ struct device *dev = &pdev->dev;
struct ice_pf *pf;
struct ice_hw *hw;
int err;
@@ -2086,20 +2227,20 @@ static int ice_probe(struct pci_dev *pdev,
err = pcim_iomap_regions(pdev, BIT(ICE_BAR0), pci_name(pdev));
if (err) {
- dev_err(&pdev->dev, "BAR0 I/O map error %d\n", err);
+ dev_err(dev, "BAR0 I/O map error %d\n", err);
return err;
}
- pf = devm_kzalloc(&pdev->dev, sizeof(*pf), GFP_KERNEL);
+ pf = devm_kzalloc(dev, sizeof(*pf), GFP_KERNEL);
if (!pf)
return -ENOMEM;
/* set up for high or low dma */
- err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+ err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64));
if (err)
- err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+ err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));
if (err) {
- dev_err(&pdev->dev, "DMA configuration failed: 0x%x\n", err);
+ dev_err(dev, "DMA configuration failed: 0x%x\n", err);
return err;
}
@@ -2133,12 +2274,12 @@ static int ice_probe(struct pci_dev *pdev,
err = ice_init_hw(hw);
if (err) {
- dev_err(&pdev->dev, "ice_init_hw failed: %d\n", err);
+ dev_err(dev, "ice_init_hw failed: %d\n", err);
err = -EIO;
goto err_exit_unroll;
}
- dev_info(&pdev->dev, "firmware %d.%d.%05d api %d.%d\n",
+ dev_info(dev, "firmware %d.%d.%05d api %d.%d\n",
hw->fw_maj_ver, hw->fw_min_ver, hw->fw_build,
hw->api_maj_ver, hw->api_min_ver);
@@ -2152,8 +2293,8 @@ static int ice_probe(struct pci_dev *pdev,
goto err_init_pf_unroll;
}
- pf->vsi = devm_kcalloc(&pdev->dev, pf->num_alloc_vsi,
- sizeof(*pf->vsi), GFP_KERNEL);
+ pf->vsi = devm_kcalloc(dev, pf->num_alloc_vsi, sizeof(*pf->vsi),
+ GFP_KERNEL);
if (!pf->vsi) {
err = -ENOMEM;
goto err_init_pf_unroll;
@@ -2161,8 +2302,7 @@ static int ice_probe(struct pci_dev *pdev,
err = ice_init_interrupt_scheme(pf);
if (err) {
- dev_err(&pdev->dev,
- "ice_init_interrupt_scheme failed: %d\n", err);
+ dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err);
err = -EIO;
goto err_init_interrupt_unroll;
}
@@ -2178,15 +2318,13 @@ static int ice_probe(struct pci_dev *pdev,
if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) {
err = ice_req_irq_msix_misc(pf);
if (err) {
- dev_err(&pdev->dev,
- "setup of misc vector failed: %d\n", err);
+ dev_err(dev, "setup of misc vector failed: %d\n", err);
goto err_init_interrupt_unroll;
}
}
/* create switch struct for the switch element created by FW on boot */
- pf->first_sw = devm_kzalloc(&pdev->dev, sizeof(*pf->first_sw),
- GFP_KERNEL);
+ pf->first_sw = devm_kzalloc(dev, sizeof(*pf->first_sw), GFP_KERNEL);
if (!pf->first_sw) {
err = -ENOMEM;
goto err_msix_misc_unroll;
@@ -2204,8 +2342,7 @@ static int ice_probe(struct pci_dev *pdev,
err = ice_setup_pf_sw(pf);
if (err) {
- dev_err(&pdev->dev,
- "probe failed due to setup pf switch:%d\n", err);
+ dev_err(dev, "probe failed due to setup pf switch:%d\n", err);
goto err_alloc_sw_unroll;
}
@@ -2214,8 +2351,13 @@ static int ice_probe(struct pci_dev *pdev,
/* since everything is good, start the service timer */
mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period));
+ err = ice_init_link_events(pf->hw.port_info);
+ if (err) {
+ dev_err(dev, "ice_init_link_events failed: %d\n", err);
+ goto err_alloc_sw_unroll;
+ }
+
ice_verify_cacheline_size(pf);
- ice_verify_itr_gran(pf);
return 0;
@@ -2227,7 +2369,7 @@ err_msix_misc_unroll:
ice_free_irq_msix_misc(pf);
err_init_interrupt_unroll:
ice_clear_interrupt_scheme(pf);
- devm_kfree(&pdev->dev, pf->vsi);
+ devm_kfree(dev, pf->vsi);
err_init_pf_unroll:
ice_deinit_pf(pf);
ice_deinit_hw(hw);
@@ -2272,6 +2414,136 @@ static void ice_remove(struct pci_dev *pdev)
pci_disable_pcie_error_reporting(pdev);
}
+/**
+ * ice_pci_err_detected - warning that PCI error has been detected
+ * @pdev: PCI device information struct
+ * @err: the type of PCI error
+ *
+ * Called to warn that something happened on the PCI bus and the error handling
+ * is in progress. Allows the driver to gracefully prepare/handle PCI errors.
+ */
+static pci_ers_result_t
+ice_pci_err_detected(struct pci_dev *pdev, enum pci_channel_state err)
+{
+ struct ice_pf *pf = pci_get_drvdata(pdev);
+
+ if (!pf) {
+ dev_err(&pdev->dev, "%s: unrecoverable device error %d\n",
+ __func__, err);
+ return PCI_ERS_RESULT_DISCONNECT;
+ }
+
+ if (!test_bit(__ICE_SUSPENDED, pf->state)) {
+ ice_service_task_stop(pf);
+
+ if (!test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) {
+ set_bit(__ICE_PFR_REQ, pf->state);
+ ice_prepare_for_reset(pf);
+ }
+ }
+
+ return PCI_ERS_RESULT_NEED_RESET;
+}
+
+/**
+ * ice_pci_err_slot_reset - a PCI slot reset has just happened
+ * @pdev: PCI device information struct
+ *
+ * Called to determine if the driver can recover from the PCI slot reset by
+ * using a register read to determine if the device is recoverable.
+ */
+static pci_ers_result_t ice_pci_err_slot_reset(struct pci_dev *pdev)
+{
+ struct ice_pf *pf = pci_get_drvdata(pdev);
+ pci_ers_result_t result;
+ int err;
+ u32 reg;
+
+ err = pci_enable_device_mem(pdev);
+ if (err) {
+ dev_err(&pdev->dev,
+ "Cannot re-enable PCI device after reset, error %d\n",
+ err);
+ result = PCI_ERS_RESULT_DISCONNECT;
+ } else {
+ pci_set_master(pdev);
+ pci_restore_state(pdev);
+ pci_save_state(pdev);
+ pci_wake_from_d3(pdev, false);
+
+ /* Check for life */
+ reg = rd32(&pf->hw, GLGEN_RTRIG);
+ if (!reg)
+ result = PCI_ERS_RESULT_RECOVERED;
+ else
+ result = PCI_ERS_RESULT_DISCONNECT;
+ }
+
+ err = pci_cleanup_aer_uncorrect_error_status(pdev);
+ if (err)
+ dev_dbg(&pdev->dev,
+ "pci_cleanup_aer_uncorrect_error_status failed, error %d\n",
+ err);
+ /* non-fatal, continue */
+
+ return result;
+}
+
+/**
+ * ice_pci_err_resume - restart operations after PCI error recovery
+ * @pdev: PCI device information struct
+ *
+ * Called to allow the driver to bring things back up after PCI error and/or
+ * reset recovery have finished
+ */
+static void ice_pci_err_resume(struct pci_dev *pdev)
+{
+ struct ice_pf *pf = pci_get_drvdata(pdev);
+
+ if (!pf) {
+ dev_err(&pdev->dev,
+ "%s failed, device is unrecoverable\n", __func__);
+ return;
+ }
+
+ if (test_bit(__ICE_SUSPENDED, pf->state)) {
+ dev_dbg(&pdev->dev, "%s failed to resume normal operations!\n",
+ __func__);
+ return;
+ }
+
+ ice_do_reset(pf, ICE_RESET_PFR);
+ ice_service_task_restart(pf);
+ mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period));
+}
+
+/**
+ * ice_pci_err_reset_prepare - prepare device driver for PCI reset
+ * @pdev: PCI device information struct
+ */
+static void ice_pci_err_reset_prepare(struct pci_dev *pdev)
+{
+ struct ice_pf *pf = pci_get_drvdata(pdev);
+
+ if (!test_bit(__ICE_SUSPENDED, pf->state)) {
+ ice_service_task_stop(pf);
+
+ if (!test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) {
+ set_bit(__ICE_PFR_REQ, pf->state);
+ ice_prepare_for_reset(pf);
+ }
+ }
+}
+
+/**
+ * ice_pci_err_reset_done - PCI reset done, device driver reset can begin
+ * @pdev: PCI device information struct
+ */
+static void ice_pci_err_reset_done(struct pci_dev *pdev)
+{
+ ice_pci_err_resume(pdev);
+}
+
/* ice_pci_tbl - PCI Device ID Table
*
* Wildcard entries (PCI_ANY_ID) should come last
@@ -2289,12 +2561,21 @@ static const struct pci_device_id ice_pci_tbl[] = {
};
MODULE_DEVICE_TABLE(pci, ice_pci_tbl);
+static const struct pci_error_handlers ice_pci_err_handler = {
+ .error_detected = ice_pci_err_detected,
+ .slot_reset = ice_pci_err_slot_reset,
+ .reset_prepare = ice_pci_err_reset_prepare,
+ .reset_done = ice_pci_err_reset_done,
+ .resume = ice_pci_err_resume
+};
+
static struct pci_driver ice_driver = {
.name = KBUILD_MODNAME,
.id_table = ice_pci_tbl,
.probe = ice_probe,
.remove = ice_remove,
.sriov_configure = ice_sriov_configure,
+ .err_handler = &ice_pci_err_handler
};
/**
@@ -2512,9 +2793,10 @@ ice_fdb_add(struct ndmsg *ndm, struct nlattr __always_unused *tb[],
* @addr: the MAC address entry being added
* @vid: VLAN id
*/
-static int ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[],
- struct net_device *dev, const unsigned char *addr,
- __always_unused u16 vid)
+static int
+ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[],
+ struct net_device *dev, const unsigned char *addr,
+ __always_unused u16 vid)
{
int err;
@@ -2538,8 +2820,8 @@ static int ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[],
* @netdev: ptr to the netdev being adjusted
* @features: the feature set that the stack is suggesting
*/
-static int ice_set_features(struct net_device *netdev,
- netdev_features_t features)
+static int
+ice_set_features(struct net_device *netdev, netdev_features_t features)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
@@ -2666,7 +2948,7 @@ static int ice_up_complete(struct ice_vsi *vsi)
ice_service_task_schedule(pf);
- return err;
+ return 0;
}
/**
@@ -2693,8 +2975,8 @@ int ice_up(struct ice_vsi *vsi)
* This function fetches stats from the ring considering the atomic operations
* that needs to be performed to read u64 values in 32 bit machine.
*/
-static void ice_fetch_u64_stats_per_ring(struct ice_ring *ring, u64 *pkts,
- u64 *bytes)
+static void
+ice_fetch_u64_stats_per_ring(struct ice_ring *ring, u64 *pkts, u64 *bytes)
{
unsigned int start;
*pkts = 0;
@@ -3276,7 +3558,7 @@ static void ice_vsi_release_all(struct ice_pf *pf)
if (!pf->vsi)
return;
- for (i = 0; i < pf->num_alloc_vsi; i++) {
+ ice_for_each_vsi(pf, i) {
if (!pf->vsi[i])
continue;
@@ -3375,16 +3657,12 @@ static int ice_vsi_rebuild_all(struct ice_pf *pf)
int i;
/* loop through pf->vsi array and reinit the VSI if found */
- for (i = 0; i < pf->num_alloc_vsi; i++) {
+ ice_for_each_vsi(pf, i) {
int err;
if (!pf->vsi[i])
continue;
- /* VF VSI rebuild isn't supported yet */
- if (pf->vsi[i]->type == ICE_VSI_VF)
- continue;
-
err = ice_vsi_rebuild(pf->vsi[i]);
if (err) {
dev_err(&pf->pdev->dev,
@@ -3412,7 +3690,7 @@ static int ice_vsi_replay_all(struct ice_pf *pf)
int i;
/* loop through pf->vsi array and replay the VSI if found */
- for (i = 0; i < pf->num_alloc_vsi; i++) {
+ ice_for_each_vsi(pf, i) {
if (!pf->vsi[i])
continue;
@@ -3521,9 +3799,7 @@ static void ice_rebuild(struct ice_pf *pf)
goto err_vsi_rebuild;
}
- ice_reset_all_vfs(pf, true);
-
- for (i = 0; i < pf->num_alloc_vsi; i++) {
+ ice_for_each_vsi(pf, i) {
bool link_up;
if (!pf->vsi[i] || pf->vsi[i]->type != ICE_VSI_PF)
diff --git a/drivers/net/ethernet/intel/ice/ice_sched.c b/drivers/net/ethernet/intel/ice/ice_sched.c
index 56049739a250..e0218f4c8f0b 100644
--- a/drivers/net/ethernet/intel/ice/ice_sched.c
+++ b/drivers/net/ethernet/intel/ice/ice_sched.c
@@ -276,7 +276,8 @@ ice_sched_remove_elems(struct ice_hw *hw, struct ice_sched_node *parent,
status = ice_aq_delete_sched_elems(hw, 1, buf, buf_size,
&num_groups_removed, NULL);
if (status || num_groups_removed != 1)
- ice_debug(hw, ICE_DBG_SCHED, "remove elements failed\n");
+ ice_debug(hw, ICE_DBG_SCHED, "remove node failed FW error %d\n",
+ hw->adminq.sq_last_status);
devm_kfree(ice_hw_to_dev(hw), buf);
return status;
@@ -360,12 +361,8 @@ void ice_free_sched_node(struct ice_port_info *pi, struct ice_sched_node *node)
node->info.data.elem_type != ICE_AQC_ELEM_TYPE_ROOT_PORT &&
node->info.data.elem_type != ICE_AQC_ELEM_TYPE_LEAF) {
u32 teid = le32_to_cpu(node->info.node_teid);
- enum ice_status status;
- status = ice_sched_remove_elems(hw, node->parent, 1, &teid);
- if (status)
- ice_debug(hw, ICE_DBG_SCHED,
- "remove element failed %d\n", status);
+ ice_sched_remove_elems(hw, node->parent, 1, &teid);
}
parent = node->parent;
/* root has no parent */
@@ -697,7 +694,8 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node,
status = ice_aq_add_sched_elems(hw, 1, buf, buf_size,
&num_groups_added, NULL);
if (status || num_groups_added != 1) {
- ice_debug(hw, ICE_DBG_SCHED, "add elements failed\n");
+ ice_debug(hw, ICE_DBG_SCHED, "add node failed FW Error %d\n",
+ hw->adminq.sq_last_status);
devm_kfree(ice_hw_to_dev(hw), buf);
return ICE_ERR_CFG;
}
@@ -1271,42 +1269,6 @@ ice_sched_add_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle,
}
/**
- * ice_sched_rm_vsi_child_nodes - remove VSI child nodes from the tree
- * @pi: port information structure
- * @vsi_node: pointer to the VSI node
- * @num_nodes: pointer to the num nodes that needs to be removed per layer
- * @owner: node owner (lan or rdma)
- *
- * This function removes the VSI child nodes from the tree. It gets called for
- * lan and rdma separately.
- */
-static void
-ice_sched_rm_vsi_child_nodes(struct ice_port_info *pi,
- struct ice_sched_node *vsi_node, u16 *num_nodes,
- u8 owner)
-{
- struct ice_sched_node *node, *next;
- u8 i, qgl, vsil;
- u16 num;
-
- qgl = ice_sched_get_qgrp_layer(pi->hw);
- vsil = ice_sched_get_vsi_layer(pi->hw);
-
- for (i = qgl; i > vsil; i--) {
- num = num_nodes[i];
- node = ice_sched_get_first_node(pi->hw, vsi_node, i);
- while (node && num) {
- next = node->sibling;
- if (node->owner == owner && !node->num_children) {
- ice_free_sched_node(pi, node);
- num--;
- }
- node = next;
- }
- }
-}
-
-/**
* ice_sched_calc_vsi_support_nodes - calculate number of VSI support nodes
* @hw: pointer to the hw struct
* @tc_node: pointer to TC node
@@ -1446,7 +1408,6 @@ static enum ice_status
ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle,
u8 tc, u16 new_numqs, u8 owner)
{
- u16 prev_num_nodes[ICE_AQC_TOPO_MAX_LEVEL_NUM] = { 0 };
u16 new_num_nodes[ICE_AQC_TOPO_MAX_LEVEL_NUM] = { 0 };
struct ice_sched_node *vsi_node;
struct ice_sched_node *tc_node;
@@ -1454,7 +1415,6 @@ ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle,
enum ice_status status = 0;
struct ice_hw *hw = pi->hw;
u16 prev_numqs;
- u8 i;
tc_node = ice_sched_get_tc_node(pi, tc);
if (!tc_node)
@@ -1473,36 +1433,25 @@ ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle,
else
return ICE_ERR_PARAM;
- /* num queues are not changed */
- if (prev_numqs == new_numqs)
+ /* num queues are not changed or less than the previous number */
+ if (new_numqs <= prev_numqs)
return status;
-
- /* calculate number of nodes based on prev/new number of qs */
- if (prev_numqs)
- ice_sched_calc_vsi_child_nodes(hw, prev_numqs, prev_num_nodes);
-
if (new_numqs)
ice_sched_calc_vsi_child_nodes(hw, new_numqs, new_num_nodes);
-
- if (prev_numqs > new_numqs) {
- for (i = 0; i < ICE_AQC_TOPO_MAX_LEVEL_NUM; i++)
- new_num_nodes[i] = prev_num_nodes[i] - new_num_nodes[i];
-
- ice_sched_rm_vsi_child_nodes(pi, vsi_node, new_num_nodes,
- owner);
- } else {
- for (i = 0; i < ICE_AQC_TOPO_MAX_LEVEL_NUM; i++)
- new_num_nodes[i] -= prev_num_nodes[i];
-
- status = ice_sched_add_vsi_child_nodes(pi, vsi_handle, tc_node,
- new_num_nodes, owner);
- if (status)
- return status;
- }
-
+ /* Keep the max number of queue configuration all the time. Update the
+ * tree only if number of queues > previous number of queues. This may
+ * leave some extra nodes in the tree if number of queues < previous
+ * number but that wouldn't harm anything. Removing those extra nodes
+ * may complicate the code if those nodes are part of SRL or
+ * individually rate limited.
+ */
+ status = ice_sched_add_vsi_child_nodes(pi, vsi_handle, tc_node,
+ new_num_nodes, owner);
+ if (status)
+ return status;
vsi_ctx->sched.max_lanq[tc] = new_numqs;
- return status;
+ return 0;
}
/**
@@ -1527,6 +1476,7 @@ ice_sched_cfg_vsi(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 maxqs,
enum ice_status status = 0;
struct ice_hw *hw = pi->hw;
+ ice_debug(pi->hw, ICE_DBG_SCHED, "add/config VSI %d\n", vsi_handle);
tc_node = ice_sched_get_tc_node(pi, tc);
if (!tc_node)
return ICE_ERR_PARAM;
@@ -1646,8 +1596,9 @@ ice_sched_rm_vsi_cfg(struct ice_port_info *pi, u16 vsi_handle, u8 owner)
{
enum ice_status status = ICE_ERR_PARAM;
struct ice_vsi_ctx *vsi_ctx;
- u8 i, j = 0;
+ u8 i;
+ ice_debug(pi->hw, ICE_DBG_SCHED, "removing VSI %d\n", vsi_handle);
if (!ice_is_vsi_valid(pi->hw, vsi_handle))
return status;
mutex_lock(&pi->sched_lock);
@@ -1655,8 +1606,9 @@ ice_sched_rm_vsi_cfg(struct ice_port_info *pi, u16 vsi_handle, u8 owner)
if (!vsi_ctx)
goto exit_sched_rm_vsi_cfg;
- for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) {
+ ice_for_each_traffic_class(i) {
struct ice_sched_node *vsi_node, *tc_node;
+ u8 j = 0;
tc_node = ice_sched_get_tc_node(pi, i);
if (!tc_node)
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 09d1c314b68f..7dcd9ddf54f7 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -322,8 +322,8 @@ struct ice_vsi_ctx *ice_get_vsi_ctx(struct ice_hw *hw, u16 vsi_handle)
*
* save the VSI context entry for a given VSI handle
*/
-static void ice_save_vsi_ctx(struct ice_hw *hw, u16 vsi_handle,
- struct ice_vsi_ctx *vsi)
+static void
+ice_save_vsi_ctx(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi)
{
hw->vsi_ctx[vsi_handle] = vsi;
}
@@ -398,7 +398,7 @@ ice_add_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx,
tmp_vsi_ctx->vsi_num = vsi_ctx->vsi_num;
}
- return status;
+ return 0;
}
/**
@@ -643,21 +643,43 @@ static void ice_fill_sw_info(struct ice_hw *hw, struct ice_fltr_info *fi)
fi->fltr_act == ICE_FWD_TO_VSI_LIST ||
fi->fltr_act == ICE_FWD_TO_Q ||
fi->fltr_act == ICE_FWD_TO_QGRP)) {
- fi->lb_en = true;
- /* Do not set lan_en to TRUE if
+ /* Setting LB for prune actions will result in replicated
+ * packets to the internal switch that will be dropped.
+ */
+ if (fi->lkup_type != ICE_SW_LKUP_VLAN)
+ fi->lb_en = true;
+
+ /* Set lan_en to TRUE if
* 1. The switch is a VEB AND
* 2
- * 2.1 The lookup is MAC with unicast addr for MAC, OR
- * 2.2 The lookup is MAC_VLAN with unicast addr for MAC
+ * 2.1 The lookup is a directional lookup like ethertype,
+ * promiscuous, ethertype-mac, promiscuous-vlan
+ * and default-port OR
+ * 2.2 The lookup is VLAN, OR
+ * 2.3 The lookup is MAC with mcast or bcast addr for MAC, OR
+ * 2.4 The lookup is MAC_VLAN with mcast or bcast addr for MAC.
*
- * In all other cases, the LAN enable has to be set to true.
+ * OR
+ *
+ * The switch is a VEPA.
+ *
+ * In all other cases, the LAN enable has to be set to false.
*/
- if (!(hw->evb_veb &&
- ((fi->lkup_type == ICE_SW_LKUP_MAC &&
- is_unicast_ether_addr(fi->l_data.mac.mac_addr)) ||
- (fi->lkup_type == ICE_SW_LKUP_MAC_VLAN &&
- is_unicast_ether_addr(fi->l_data.mac_vlan.mac_addr)))))
+ if (hw->evb_veb) {
+ if (fi->lkup_type == ICE_SW_LKUP_ETHERTYPE ||
+ fi->lkup_type == ICE_SW_LKUP_PROMISC ||
+ fi->lkup_type == ICE_SW_LKUP_ETHERTYPE_MAC ||
+ fi->lkup_type == ICE_SW_LKUP_PROMISC_VLAN ||
+ fi->lkup_type == ICE_SW_LKUP_DFLT ||
+ fi->lkup_type == ICE_SW_LKUP_VLAN ||
+ (fi->lkup_type == ICE_SW_LKUP_MAC &&
+ !is_unicast_ether_addr(fi->l_data.mac.mac_addr)) ||
+ (fi->lkup_type == ICE_SW_LKUP_MAC_VLAN &&
+ !is_unicast_ether_addr(fi->l_data.mac.mac_addr)))
+ fi->lan_en = true;
+ } else {
fi->lan_en = true;
+ }
}
}
@@ -2190,6 +2212,291 @@ ice_add_to_vsi_fltr_list(struct ice_hw *hw, u16 vsi_handle,
}
/**
+ * ice_determine_promisc_mask
+ * @fi: filter info to parse
+ *
+ * Helper function to determine which ICE_PROMISC_ mask corresponds
+ * to given filter into.
+ */
+static u8 ice_determine_promisc_mask(struct ice_fltr_info *fi)
+{
+ u16 vid = fi->l_data.mac_vlan.vlan_id;
+ u8 *macaddr = fi->l_data.mac.mac_addr;
+ bool is_tx_fltr = false;
+ u8 promisc_mask = 0;
+
+ if (fi->flag == ICE_FLTR_TX)
+ is_tx_fltr = true;
+
+ if (is_broadcast_ether_addr(macaddr))
+ promisc_mask |= is_tx_fltr ?
+ ICE_PROMISC_BCAST_TX : ICE_PROMISC_BCAST_RX;
+ else if (is_multicast_ether_addr(macaddr))
+ promisc_mask |= is_tx_fltr ?
+ ICE_PROMISC_MCAST_TX : ICE_PROMISC_MCAST_RX;
+ else if (is_unicast_ether_addr(macaddr))
+ promisc_mask |= is_tx_fltr ?
+ ICE_PROMISC_UCAST_TX : ICE_PROMISC_UCAST_RX;
+ if (vid)
+ promisc_mask |= is_tx_fltr ?
+ ICE_PROMISC_VLAN_TX : ICE_PROMISC_VLAN_RX;
+
+ return promisc_mask;
+}
+
+/**
+ * ice_remove_promisc - Remove promisc based filter rules
+ * @hw: pointer to the hardware structure
+ * @recp_id: recipe id for which the rule needs to removed
+ * @v_list: list of promisc entries
+ */
+static enum ice_status
+ice_remove_promisc(struct ice_hw *hw, u8 recp_id,
+ struct list_head *v_list)
+{
+ struct ice_fltr_list_entry *v_list_itr, *tmp;
+
+ list_for_each_entry_safe(v_list_itr, tmp, v_list, list_entry) {
+ v_list_itr->status =
+ ice_remove_rule_internal(hw, recp_id, v_list_itr);
+ if (v_list_itr->status)
+ return v_list_itr->status;
+ }
+ return 0;
+}
+
+/**
+ * ice_clear_vsi_promisc - clear specified promiscuous mode(s) for given VSI
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: VSI handle to clear mode
+ * @promisc_mask: mask of promiscuous config bits to clear
+ * @vid: VLAN ID to clear VLAN promiscuous
+ */
+enum ice_status
+ice_clear_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask,
+ u16 vid)
+{
+ struct ice_switch_info *sw = hw->switch_info;
+ struct ice_fltr_list_entry *fm_entry, *tmp;
+ struct list_head remove_list_head;
+ struct ice_fltr_mgmt_list_entry *itr;
+ struct list_head *rule_head;
+ struct mutex *rule_lock; /* Lock to protect filter rule list */
+ enum ice_status status = 0;
+ u8 recipe_id;
+
+ if (!ice_is_vsi_valid(hw, vsi_handle))
+ return ICE_ERR_PARAM;
+
+ if (vid)
+ recipe_id = ICE_SW_LKUP_PROMISC_VLAN;
+ else
+ recipe_id = ICE_SW_LKUP_PROMISC;
+
+ rule_head = &sw->recp_list[recipe_id].filt_rules;
+ rule_lock = &sw->recp_list[recipe_id].filt_rule_lock;
+
+ INIT_LIST_HEAD(&remove_list_head);
+
+ mutex_lock(rule_lock);
+ list_for_each_entry(itr, rule_head, list_entry) {
+ u8 fltr_promisc_mask = 0;
+
+ if (!ice_vsi_uses_fltr(itr, vsi_handle))
+ continue;
+
+ fltr_promisc_mask |=
+ ice_determine_promisc_mask(&itr->fltr_info);
+
+ /* Skip if filter is not completely specified by given mask */
+ if (fltr_promisc_mask & ~promisc_mask)
+ continue;
+
+ status = ice_add_entry_to_vsi_fltr_list(hw, vsi_handle,
+ &remove_list_head,
+ &itr->fltr_info);
+ if (status) {
+ mutex_unlock(rule_lock);
+ goto free_fltr_list;
+ }
+ }
+ mutex_unlock(rule_lock);
+
+ status = ice_remove_promisc(hw, recipe_id, &remove_list_head);
+
+free_fltr_list:
+ list_for_each_entry_safe(fm_entry, tmp, &remove_list_head, list_entry) {
+ list_del(&fm_entry->list_entry);
+ devm_kfree(ice_hw_to_dev(hw), fm_entry);
+ }
+
+ return status;
+}
+
+/**
+ * ice_set_vsi_promisc - set given VSI to given promiscuous mode(s)
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: VSI handle to configure
+ * @promisc_mask: mask of promiscuous config bits
+ * @vid: VLAN ID to set VLAN promiscuous
+ */
+enum ice_status
+ice_set_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, u16 vid)
+{
+ enum { UCAST_FLTR = 1, MCAST_FLTR, BCAST_FLTR };
+ struct ice_fltr_list_entry f_list_entry;
+ struct ice_fltr_info new_fltr;
+ enum ice_status status = 0;
+ bool is_tx_fltr;
+ u16 hw_vsi_id;
+ int pkt_type;
+ u8 recipe_id;
+
+ if (!ice_is_vsi_valid(hw, vsi_handle))
+ return ICE_ERR_PARAM;
+ hw_vsi_id = ice_get_hw_vsi_num(hw, vsi_handle);
+
+ memset(&new_fltr, 0, sizeof(new_fltr));
+
+ if (promisc_mask & (ICE_PROMISC_VLAN_RX | ICE_PROMISC_VLAN_TX)) {
+ new_fltr.lkup_type = ICE_SW_LKUP_PROMISC_VLAN;
+ new_fltr.l_data.mac_vlan.vlan_id = vid;
+ recipe_id = ICE_SW_LKUP_PROMISC_VLAN;
+ } else {
+ new_fltr.lkup_type = ICE_SW_LKUP_PROMISC;
+ recipe_id = ICE_SW_LKUP_PROMISC;
+ }
+
+ /* Separate filters must be set for each direction/packet type
+ * combination, so we will loop over the mask value, store the
+ * individual type, and clear it out in the input mask as it
+ * is found.
+ */
+ while (promisc_mask) {
+ u8 *mac_addr;
+
+ pkt_type = 0;
+ is_tx_fltr = false;
+
+ if (promisc_mask & ICE_PROMISC_UCAST_RX) {
+ promisc_mask &= ~ICE_PROMISC_UCAST_RX;
+ pkt_type = UCAST_FLTR;
+ } else if (promisc_mask & ICE_PROMISC_UCAST_TX) {
+ promisc_mask &= ~ICE_PROMISC_UCAST_TX;
+ pkt_type = UCAST_FLTR;
+ is_tx_fltr = true;
+ } else if (promisc_mask & ICE_PROMISC_MCAST_RX) {
+ promisc_mask &= ~ICE_PROMISC_MCAST_RX;
+ pkt_type = MCAST_FLTR;
+ } else if (promisc_mask & ICE_PROMISC_MCAST_TX) {
+ promisc_mask &= ~ICE_PROMISC_MCAST_TX;
+ pkt_type = MCAST_FLTR;
+ is_tx_fltr = true;
+ } else if (promisc_mask & ICE_PROMISC_BCAST_RX) {
+ promisc_mask &= ~ICE_PROMISC_BCAST_RX;
+ pkt_type = BCAST_FLTR;
+ } else if (promisc_mask & ICE_PROMISC_BCAST_TX) {
+ promisc_mask &= ~ICE_PROMISC_BCAST_TX;
+ pkt_type = BCAST_FLTR;
+ is_tx_fltr = true;
+ }
+
+ /* Check for VLAN promiscuous flag */
+ if (promisc_mask & ICE_PROMISC_VLAN_RX) {
+ promisc_mask &= ~ICE_PROMISC_VLAN_RX;
+ } else if (promisc_mask & ICE_PROMISC_VLAN_TX) {
+ promisc_mask &= ~ICE_PROMISC_VLAN_TX;
+ is_tx_fltr = true;
+ }
+
+ /* Set filter DA based on packet type */
+ mac_addr = new_fltr.l_data.mac.mac_addr;
+ if (pkt_type == BCAST_FLTR) {
+ eth_broadcast_addr(mac_addr);
+ } else if (pkt_type == MCAST_FLTR ||
+ pkt_type == UCAST_FLTR) {
+ /* Use the dummy ether header DA */
+ ether_addr_copy(mac_addr, dummy_eth_header);
+ if (pkt_type == MCAST_FLTR)
+ mac_addr[0] |= 0x1; /* Set multicast bit */
+ }
+
+ /* Need to reset this to zero for all iterations */
+ new_fltr.flag = 0;
+ if (is_tx_fltr) {
+ new_fltr.flag |= ICE_FLTR_TX;
+ new_fltr.src = hw_vsi_id;
+ } else {
+ new_fltr.flag |= ICE_FLTR_RX;
+ new_fltr.src = hw->port_info->lport;
+ }
+
+ new_fltr.fltr_act = ICE_FWD_TO_VSI;
+ new_fltr.vsi_handle = vsi_handle;
+ new_fltr.fwd_id.hw_vsi_id = hw_vsi_id;
+ f_list_entry.fltr_info = new_fltr;
+
+ status = ice_add_rule_internal(hw, recipe_id, &f_list_entry);
+ if (status)
+ goto set_promisc_exit;
+ }
+
+set_promisc_exit:
+ return status;
+}
+
+/**
+ * ice_set_vlan_vsi_promisc
+ * @hw: pointer to the hardware structure
+ * @vsi_handle: VSI handle to configure
+ * @promisc_mask: mask of promiscuous config bits
+ * @rm_vlan_promisc: Clear VLANs VSI promisc mode
+ *
+ * Configure VSI with all associated VLANs to given promiscuous mode(s)
+ */
+enum ice_status
+ice_set_vlan_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask,
+ bool rm_vlan_promisc)
+{
+ struct ice_switch_info *sw = hw->switch_info;
+ struct ice_fltr_list_entry *list_itr, *tmp;
+ struct list_head vsi_list_head;
+ struct list_head *vlan_head;
+ struct mutex *vlan_lock; /* Lock to protect filter rule list */
+ enum ice_status status;
+ u16 vlan_id;
+
+ INIT_LIST_HEAD(&vsi_list_head);
+ vlan_lock = &sw->recp_list[ICE_SW_LKUP_VLAN].filt_rule_lock;
+ vlan_head = &sw->recp_list[ICE_SW_LKUP_VLAN].filt_rules;
+ mutex_lock(vlan_lock);
+ status = ice_add_to_vsi_fltr_list(hw, vsi_handle, vlan_head,
+ &vsi_list_head);
+ mutex_unlock(vlan_lock);
+ if (status)
+ goto free_fltr_list;
+
+ list_for_each_entry(list_itr, &vsi_list_head, list_entry) {
+ vlan_id = list_itr->fltr_info.l_data.vlan.vlan_id;
+ if (rm_vlan_promisc)
+ status = ice_clear_vsi_promisc(hw, vsi_handle,
+ promisc_mask, vlan_id);
+ else
+ status = ice_set_vsi_promisc(hw, vsi_handle,
+ promisc_mask, vlan_id);
+ if (status)
+ break;
+ }
+
+free_fltr_list:
+ list_for_each_entry_safe(list_itr, tmp, &vsi_list_head, list_entry) {
+ list_del(&list_itr->list_entry);
+ devm_kfree(ice_hw_to_dev(hw), list_itr);
+ }
+ return status;
+}
+
+/**
* ice_remove_vsi_lkup_fltr - Remove lookup type filters for a VSI
* @hw: pointer to the hardware structure
* @vsi_handle: VSI handle to remove filters from
@@ -2224,12 +2531,14 @@ ice_remove_vsi_lkup_fltr(struct ice_hw *hw, u16 vsi_handle,
case ICE_SW_LKUP_VLAN:
ice_remove_vlan(hw, &remove_list_head);
break;
+ case ICE_SW_LKUP_PROMISC:
+ case ICE_SW_LKUP_PROMISC_VLAN:
+ ice_remove_promisc(hw, lkup, &remove_list_head);
+ break;
case ICE_SW_LKUP_MAC_VLAN:
case ICE_SW_LKUP_ETHERTYPE:
case ICE_SW_LKUP_ETHERTYPE_MAC:
- case ICE_SW_LKUP_PROMISC:
case ICE_SW_LKUP_DFLT:
- case ICE_SW_LKUP_PROMISC_VLAN:
case ICE_SW_LKUP_LAST:
default:
ice_debug(hw, ICE_DBG_SW, "Unsupported lookup type %d\n", lkup);
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index d5ef0bd58bf9..e4ce0720b871 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -178,6 +178,17 @@ struct ice_fltr_mgmt_list_entry {
u8 counter_index;
};
+enum ice_promisc_flags {
+ ICE_PROMISC_UCAST_RX = 0x1,
+ ICE_PROMISC_UCAST_TX = 0x2,
+ ICE_PROMISC_MCAST_RX = 0x4,
+ ICE_PROMISC_MCAST_TX = 0x8,
+ ICE_PROMISC_BCAST_RX = 0x10,
+ ICE_PROMISC_BCAST_TX = 0x20,
+ ICE_PROMISC_VLAN_RX = 0x40,
+ ICE_PROMISC_VLAN_TX = 0x80,
+};
+
/* VSI related commands */
enum ice_status
ice_add_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx,
@@ -199,10 +210,22 @@ enum ice_status ice_update_sw_rule_bridge_mode(struct ice_hw *hw);
enum ice_status ice_add_mac(struct ice_hw *hw, struct list_head *m_lst);
enum ice_status ice_remove_mac(struct ice_hw *hw, struct list_head *m_lst);
void ice_remove_vsi_fltr(struct ice_hw *hw, u16 vsi_handle);
-enum ice_status ice_add_vlan(struct ice_hw *hw, struct list_head *m_list);
+enum ice_status
+ice_add_vlan(struct ice_hw *hw, struct list_head *m_list);
enum ice_status ice_remove_vlan(struct ice_hw *hw, struct list_head *v_list);
+
+/* Promisc/defport setup for VSIs */
enum ice_status
ice_cfg_dflt_vsi(struct ice_hw *hw, u16 vsi_handle, bool set, u8 direction);
+enum ice_status
+ice_set_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask,
+ u16 vid);
+enum ice_status
+ice_clear_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask,
+ u16 vid);
+enum ice_status
+ice_set_vlan_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask,
+ bool rm_vlan_promisc);
enum ice_status ice_init_def_sw_recp(struct ice_hw *hw);
u16 ice_get_hw_vsi_num(struct ice_hw *hw, u16 vsi_handle);
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index c289d97f477d..f2462799154a 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -100,8 +100,8 @@ void ice_free_tx_ring(struct ice_ring *tx_ring)
*
* Returns true if there's any budget left (e.g. the clean is finished)
*/
-static bool ice_clean_tx_irq(struct ice_vsi *vsi, struct ice_ring *tx_ring,
- int napi_budget)
+static bool
+ice_clean_tx_irq(struct ice_vsi *vsi, struct ice_ring *tx_ring, int napi_budget)
{
unsigned int total_bytes = 0, total_pkts = 0;
unsigned int budget = vsi->work_lmt;
@@ -236,9 +236,9 @@ int ice_setup_tx_ring(struct ice_ring *tx_ring)
if (!tx_ring->tx_buf)
return -ENOMEM;
- /* round up to nearest 4K */
+ /* round up to nearest page */
tx_ring->size = ALIGN(tx_ring->count * sizeof(struct ice_tx_desc),
- 4096);
+ PAGE_SIZE);
tx_ring->desc = dmam_alloc_coherent(dev, tx_ring->size, &tx_ring->dma,
GFP_KERNEL);
if (!tx_ring->desc) {
@@ -282,8 +282,17 @@ void ice_clean_rx_ring(struct ice_ring *rx_ring)
if (!rx_buf->page)
continue;
- dma_unmap_page(dev, rx_buf->dma, PAGE_SIZE, DMA_FROM_DEVICE);
- __free_pages(rx_buf->page, 0);
+ /* Invalidate cache lines that may have been written to by
+ * device so that we avoid corrupting memory.
+ */
+ dma_sync_single_range_for_cpu(dev, rx_buf->dma,
+ rx_buf->page_offset,
+ ICE_RXBUF_2048, DMA_FROM_DEVICE);
+
+ /* free resources associated with mapping */
+ dma_unmap_page_attrs(dev, rx_buf->dma, PAGE_SIZE,
+ DMA_FROM_DEVICE, ICE_RX_DMA_ATTR);
+ __page_frag_cache_drain(rx_buf->page, rx_buf->pagecnt_bias);
rx_buf->page = NULL;
rx_buf->page_offset = 0;
@@ -339,9 +348,9 @@ int ice_setup_rx_ring(struct ice_ring *rx_ring)
if (!rx_ring->rx_buf)
return -ENOMEM;
- /* round up to nearest 4K */
- rx_ring->size = rx_ring->count * sizeof(union ice_32byte_rx_desc);
- rx_ring->size = ALIGN(rx_ring->size, 4096);
+ /* round up to nearest page */
+ rx_ring->size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc),
+ PAGE_SIZE);
rx_ring->desc = dmam_alloc_coherent(dev, rx_ring->size, &rx_ring->dma,
GFP_KERNEL);
if (!rx_ring->desc) {
@@ -389,8 +398,8 @@ static void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val)
* Returns true if the page was successfully allocated or
* reused.
*/
-static bool ice_alloc_mapped_page(struct ice_ring *rx_ring,
- struct ice_rx_buf *bi)
+static bool
+ice_alloc_mapped_page(struct ice_ring *rx_ring, struct ice_rx_buf *bi)
{
struct page *page = bi->page;
dma_addr_t dma;
@@ -409,7 +418,8 @@ static bool ice_alloc_mapped_page(struct ice_ring *rx_ring,
}
/* map page for use */
- dma = dma_map_page(rx_ring->dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE);
+ dma = dma_map_page_attrs(rx_ring->dev, page, 0, PAGE_SIZE,
+ DMA_FROM_DEVICE, ICE_RX_DMA_ATTR);
/* if mapping failed free memory back to system since
* there isn't much point in holding memory we can't use
@@ -423,6 +433,8 @@ static bool ice_alloc_mapped_page(struct ice_ring *rx_ring,
bi->dma = dma;
bi->page = page;
bi->page_offset = 0;
+ page_ref_add(page, USHRT_MAX - 1);
+ bi->pagecnt_bias = USHRT_MAX;
return true;
}
@@ -452,6 +464,12 @@ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count)
if (!ice_alloc_mapped_page(rx_ring, bi))
goto no_bufs;
+ /* sync the buffer for use by the device */
+ dma_sync_single_range_for_device(rx_ring->dev, bi->dma,
+ bi->page_offset,
+ ICE_RXBUF_2048,
+ DMA_FROM_DEVICE);
+
/* Refresh the desc even if buffer_addrs didn't change
* because each write-back erases this info.
*/
@@ -497,61 +515,43 @@ static bool ice_page_is_reserved(struct page *page)
}
/**
- * ice_add_rx_frag - Add contents of Rx buffer to sk_buff
- * @rx_buf: buffer containing page to add
- * @rx_desc: descriptor containing length of buffer written by hardware
- * @skb: sk_buf to place the data into
+ * ice_rx_buf_adjust_pg_offset - Prepare Rx buffer for reuse
+ * @rx_buf: Rx buffer to adjust
+ * @size: Size of adjustment
*
- * This function will add the data contained in rx_buf->page to the skb.
- * This is done either through a direct copy if the data in the buffer is
- * less than the skb header size, otherwise it will just attach the page as
- * a frag to the skb.
- *
- * The function will then update the page offset if necessary and return
- * true if the buffer can be reused by the adapter.
+ * Update the offset within page so that Rx buf will be ready to be reused.
+ * For systems with PAGE_SIZE < 8192 this function will flip the page offset
+ * so the second half of page assigned to Rx buffer will be used, otherwise
+ * the offset is moved by the @size bytes
*/
-static bool ice_add_rx_frag(struct ice_rx_buf *rx_buf,
- union ice_32b_rx_flex_desc *rx_desc,
- struct sk_buff *skb)
+static void
+ice_rx_buf_adjust_pg_offset(struct ice_rx_buf *rx_buf, unsigned int size)
{
#if (PAGE_SIZE < 8192)
- unsigned int truesize = ICE_RXBUF_2048;
+ /* flip page offset to other buffer */
+ rx_buf->page_offset ^= size;
#else
- unsigned int last_offset = PAGE_SIZE - ICE_RXBUF_2048;
- unsigned int truesize;
-#endif /* PAGE_SIZE < 8192) */
-
- struct page *page;
- unsigned int size;
-
- size = le16_to_cpu(rx_desc->wb.pkt_len) &
- ICE_RX_FLX_DESC_PKT_LEN_M;
-
- page = rx_buf->page;
+ /* move offset up to the next cache line */
+ rx_buf->page_offset += size;
+#endif
+}
+/**
+ * ice_can_reuse_rx_page - Determine if page can be reused for another Rx
+ * @rx_buf: buffer containing the page
+ *
+ * If page is reusable, we have a green light for calling ice_reuse_rx_page,
+ * which will assign the current buffer to the buffer that next_to_alloc is
+ * pointing to; otherwise, the DMA mapping needs to be destroyed and
+ * page freed
+ */
+static bool ice_can_reuse_rx_page(struct ice_rx_buf *rx_buf)
+{
#if (PAGE_SIZE >= 8192)
- truesize = ALIGN(size, L1_CACHE_BYTES);
-#endif /* PAGE_SIZE >= 8192) */
-
- /* will the data fit in the skb we allocated? if so, just
- * copy it as it is pretty small anyway
- */
- if (size <= ICE_RX_HDR_SIZE && !skb_is_nonlinear(skb)) {
- unsigned char *va = page_address(page) + rx_buf->page_offset;
-
- memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
-
- /* page is not reserved, we can reuse buffer as-is */
- if (likely(!ice_page_is_reserved(page)))
- return true;
-
- /* this page cannot be reused so discard it */
- __free_pages(page, 0);
- return false;
- }
-
- skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
- rx_buf->page_offset, size, truesize);
+ unsigned int last_offset = PAGE_SIZE - ICE_RXBUF_2048;
+#endif
+ unsigned int pagecnt_bias = rx_buf->pagecnt_bias;
+ struct page *page = rx_buf->page;
/* avoid re-using remote pages */
if (unlikely(ice_page_is_reserved(page)))
@@ -559,36 +559,61 @@ static bool ice_add_rx_frag(struct ice_rx_buf *rx_buf,
#if (PAGE_SIZE < 8192)
/* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != 1))
+ if (unlikely((page_count(page) - pagecnt_bias) > 1))
return false;
-
- /* flip page offset to other buffer */
- rx_buf->page_offset ^= truesize;
#else
- /* move offset up to the next cache line */
- rx_buf->page_offset += truesize;
-
if (rx_buf->page_offset > last_offset)
return false;
#endif /* PAGE_SIZE < 8192) */
- /* Even if we own the page, we are not allowed to use atomic_set()
- * This would break get_page_unless_zero() users.
+ /* If we have drained the page fragment pool we need to update
+ * the pagecnt_bias and page count so that we fully restock the
+ * number of references the driver holds.
*/
- get_page(rx_buf->page);
+ if (unlikely(pagecnt_bias == 1)) {
+ page_ref_add(page, USHRT_MAX - 1);
+ rx_buf->pagecnt_bias = USHRT_MAX;
+ }
return true;
}
/**
+ * ice_add_rx_frag - Add contents of Rx buffer to sk_buff as a frag
+ * @rx_buf: buffer containing page to add
+ * @skb: sk_buff to place the data into
+ * @size: packet length from rx_desc
+ *
+ * This function will add the data contained in rx_buf->page to the skb.
+ * It will just attach the page as a frag to the skb.
+ * The function will then update the page offset.
+ */
+static void
+ice_add_rx_frag(struct ice_rx_buf *rx_buf, struct sk_buff *skb,
+ unsigned int size)
+{
+#if (PAGE_SIZE >= 8192)
+ unsigned int truesize = SKB_DATA_ALIGN(size);
+#else
+ unsigned int truesize = ICE_RXBUF_2048;
+#endif
+
+ skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page,
+ rx_buf->page_offset, size, truesize);
+
+ /* page is being used so we must update the page offset */
+ ice_rx_buf_adjust_pg_offset(rx_buf, truesize);
+}
+
+/**
* ice_reuse_rx_page - page flip buffer and store it back on the ring
* @rx_ring: Rx descriptor ring to store buffers on
* @old_buf: donor buffer to have page reused
*
* Synchronizes page for reuse by the adapter
*/
-static void ice_reuse_rx_page(struct ice_ring *rx_ring,
- struct ice_rx_buf *old_buf)
+static void
+ice_reuse_rx_page(struct ice_ring *rx_ring, struct ice_rx_buf *old_buf)
{
u16 nta = rx_ring->next_to_alloc;
struct ice_rx_buf *new_buf;
@@ -599,121 +624,132 @@ static void ice_reuse_rx_page(struct ice_ring *rx_ring,
nta++;
rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
- /* transfer page from old buffer to new buffer */
- *new_buf = *old_buf;
+ /* Transfer page from old buffer to new buffer.
+ * Move each member individually to avoid possible store
+ * forwarding stalls and unnecessary copy of skb.
+ */
+ new_buf->dma = old_buf->dma;
+ new_buf->page = old_buf->page;
+ new_buf->page_offset = old_buf->page_offset;
+ new_buf->pagecnt_bias = old_buf->pagecnt_bias;
}
/**
- * ice_fetch_rx_buf - Allocate skb and populate it
+ * ice_get_rx_buf - Fetch Rx buffer and synchronize data for use
* @rx_ring: Rx descriptor ring to transact packets on
- * @rx_desc: descriptor containing info written by hardware
+ * @skb: skb to be used
+ * @size: size of buffer to add to skb
*
- * This function allocates an skb on the fly, and populates it with the page
- * data from the current receive descriptor, taking care to set up the skb
- * correctly, as well as handling calling the page recycle function if
- * necessary.
+ * This function will pull an Rx buffer from the ring and synchronize it
+ * for use by the CPU.
*/
-static struct sk_buff *ice_fetch_rx_buf(struct ice_ring *rx_ring,
- union ice_32b_rx_flex_desc *rx_desc)
+static struct ice_rx_buf *
+ice_get_rx_buf(struct ice_ring *rx_ring, struct sk_buff **skb,
+ const unsigned int size)
{
struct ice_rx_buf *rx_buf;
- struct sk_buff *skb;
- struct page *page;
rx_buf = &rx_ring->rx_buf[rx_ring->next_to_clean];
- page = rx_buf->page;
- prefetchw(page);
+ prefetchw(rx_buf->page);
+ *skb = rx_buf->skb;
+
+ /* we are reusing so sync this buffer for CPU use */
+ dma_sync_single_range_for_cpu(rx_ring->dev, rx_buf->dma,
+ rx_buf->page_offset, size,
+ DMA_FROM_DEVICE);
- skb = rx_buf->skb;
+ /* We have pulled a buffer for use, so decrement pagecnt_bias */
+ rx_buf->pagecnt_bias--;
- if (likely(!skb)) {
- u8 *page_addr = page_address(page) + rx_buf->page_offset;
+ return rx_buf;
+}
- /* prefetch first cache line of first page */
- prefetch(page_addr);
+/**
+ * ice_construct_skb - Allocate skb and populate it
+ * @rx_ring: Rx descriptor ring to transact packets on
+ * @rx_buf: Rx buffer to pull data from
+ * @size: the length of the packet
+ *
+ * This function allocates an skb. It then populates it with the page
+ * data from the current receive descriptor, taking care to set up the
+ * skb correctly.
+ */
+static struct sk_buff *
+ice_construct_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf,
+ unsigned int size)
+{
+ void *va = page_address(rx_buf->page) + rx_buf->page_offset;
+ unsigned int headlen;
+ struct sk_buff *skb;
+
+ /* prefetch first cache line of first page */
+ prefetch(va);
#if L1_CACHE_BYTES < 128
- prefetch((void *)(page_addr + L1_CACHE_BYTES));
+ prefetch((u8 *)va + L1_CACHE_BYTES);
#endif /* L1_CACHE_BYTES */
- /* allocate a skb to store the frags */
- skb = __napi_alloc_skb(&rx_ring->q_vector->napi,
- ICE_RX_HDR_SIZE,
- GFP_ATOMIC | __GFP_NOWARN);
- if (unlikely(!skb)) {
- rx_ring->rx_stats.alloc_buf_failed++;
- return NULL;
- }
-
- /* we will be copying header into skb->data in
- * pskb_may_pull so it is in our interest to prefetch
- * it now to avoid a possible cache miss
- */
- prefetchw(skb->data);
+ /* allocate a skb to store the frags */
+ skb = __napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE,
+ GFP_ATOMIC | __GFP_NOWARN);
+ if (unlikely(!skb))
+ return NULL;
- skb_record_rx_queue(skb, rx_ring->q_index);
- } else {
- /* we are reusing so sync this buffer for CPU use */
- dma_sync_single_range_for_cpu(rx_ring->dev, rx_buf->dma,
- rx_buf->page_offset,
- ICE_RXBUF_2048,
- DMA_FROM_DEVICE);
+ skb_record_rx_queue(skb, rx_ring->q_index);
+ /* Determine available headroom for copy */
+ headlen = size;
+ if (headlen > ICE_RX_HDR_SIZE)
+ headlen = eth_get_headlen(va, ICE_RX_HDR_SIZE);
- rx_buf->skb = NULL;
- }
+ /* align pull length to size of long to optimize memcpy performance */
+ memcpy(__skb_put(skb, headlen), va, ALIGN(headlen, sizeof(long)));
- /* pull page into skb */
- if (ice_add_rx_frag(rx_buf, rx_desc, skb)) {
- /* hand second half of page back to the ring */
- ice_reuse_rx_page(rx_ring, rx_buf);
- rx_ring->rx_stats.page_reuse_count++;
+ /* if we exhaust the linear part then add what is left as a frag */
+ size -= headlen;
+ if (size) {
+#if (PAGE_SIZE >= 8192)
+ unsigned int truesize = SKB_DATA_ALIGN(size);
+#else
+ unsigned int truesize = ICE_RXBUF_2048;
+#endif
+ skb_add_rx_frag(skb, 0, rx_buf->page,
+ rx_buf->page_offset + headlen, size, truesize);
+ /* buffer is used by skb, update page_offset */
+ ice_rx_buf_adjust_pg_offset(rx_buf, truesize);
} else {
- /* we are not reusing the buffer so unmap it */
- dma_unmap_page(rx_ring->dev, rx_buf->dma, PAGE_SIZE,
- DMA_FROM_DEVICE);
+ /* buffer is unused, reset bias back to rx_buf; data was copied
+ * onto skb's linear part so there's no need for adjusting
+ * page offset and we can reuse this buffer as-is
+ */
+ rx_buf->pagecnt_bias++;
}
- /* clear contents of buffer_info */
- rx_buf->page = NULL;
-
return skb;
}
/**
- * ice_pull_tail - ice specific version of skb_pull_tail
- * @skb: pointer to current skb being adjusted
+ * ice_put_rx_buf - Clean up used buffer and either recycle or free
+ * @rx_ring: Rx descriptor ring to transact packets on
+ * @rx_buf: Rx buffer to pull data from
*
- * This function is an ice specific version of __pskb_pull_tail. The
- * main difference between this version and the original function is that
- * this function can make several assumptions about the state of things
- * that allow for significant optimizations versus the standard function.
- * As a result we can do things like drop a frag and maintain an accurate
- * truesize for the skb.
+ * This function will clean up the contents of the rx_buf. It will
+ * either recycle the buffer or unmap it and free the associated resources.
*/
-static void ice_pull_tail(struct sk_buff *skb)
+static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
{
- struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
- unsigned int pull_len;
- unsigned char *va;
-
- /* it is valid to use page_address instead of kmap since we are
- * working with pages allocated out of the lomem pool per
- * alloc_page(GFP_ATOMIC)
- */
- va = skb_frag_address(frag);
-
- /* we need the header to contain the greater of either ETH_HLEN or
- * 60 bytes if the skb->len is less than 60 for skb_pad.
- */
- pull_len = eth_get_headlen(va, ICE_RX_HDR_SIZE);
-
- /* align pull length to size of long to optimize memcpy performance */
- skb_copy_to_linear_data(skb, va, ALIGN(pull_len, sizeof(long)));
+ /* hand second half of page back to the ring */
+ if (ice_can_reuse_rx_page(rx_buf)) {
+ ice_reuse_rx_page(rx_ring, rx_buf);
+ rx_ring->rx_stats.page_reuse_count++;
+ } else {
+ /* we are not reusing the buffer so unmap it */
+ dma_unmap_page_attrs(rx_ring->dev, rx_buf->dma, PAGE_SIZE,
+ DMA_FROM_DEVICE, ICE_RX_DMA_ATTR);
+ __page_frag_cache_drain(rx_buf->page, rx_buf->pagecnt_bias);
+ }
- /* update all of the pointers */
- skb_frag_size_sub(frag, pull_len);
- frag->page_offset += pull_len;
- skb->data_len -= pull_len;
- skb->tail += pull_len;
+ /* clear contents of buffer_info */
+ rx_buf->page = NULL;
+ rx_buf->skb = NULL;
}
/**
@@ -730,10 +766,6 @@ static void ice_pull_tail(struct sk_buff *skb)
*/
static bool ice_cleanup_headers(struct sk_buff *skb)
{
- /* place header in linear portion of buffer */
- if (skb_is_nonlinear(skb))
- ice_pull_tail(skb);
-
/* if eth_skb_pad returns an error the skb was freed */
if (eth_skb_pad(skb))
return true;
@@ -751,8 +783,8 @@ static bool ice_cleanup_headers(struct sk_buff *skb)
* The status_error_len doesn't need to be shifted because it begins
* at offset zero.
*/
-static bool ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc,
- const u16 stat_err_bits)
+static bool
+ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc, const u16 stat_err_bits)
{
return !!(rx_desc->wb.status_error0 &
cpu_to_le16(stat_err_bits));
@@ -769,9 +801,9 @@ static bool ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc,
* sk_buff in the next buffer to be chained and return true indicating
* that this is in fact a non-EOP buffer.
*/
-static bool ice_is_non_eop(struct ice_ring *rx_ring,
- union ice_32b_rx_flex_desc *rx_desc,
- struct sk_buff *skb)
+static bool
+ice_is_non_eop(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc,
+ struct sk_buff *skb)
{
u32 ntc = rx_ring->next_to_clean + 1;
@@ -838,8 +870,9 @@ ice_rx_hash(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc,
*
* skb->protocol must be set before this function is called
*/
-static void ice_rx_csum(struct ice_vsi *vsi, struct sk_buff *skb,
- union ice_32b_rx_flex_desc *rx_desc, u8 ptype)
+static void
+ice_rx_csum(struct ice_vsi *vsi, struct sk_buff *skb,
+ union ice_32b_rx_flex_desc *rx_desc, u8 ptype)
{
struct ice_rx_ptype_decoded decoded;
u32 rx_error, rx_status;
@@ -909,9 +942,10 @@ checksum_fail:
* order to populate the hash, checksum, VLAN, protocol, and
* other fields within the skb.
*/
-static void ice_process_skb_fields(struct ice_ring *rx_ring,
- union ice_32b_rx_flex_desc *rx_desc,
- struct sk_buff *skb, u8 ptype)
+static void
+ice_process_skb_fields(struct ice_ring *rx_ring,
+ union ice_32b_rx_flex_desc *rx_desc,
+ struct sk_buff *skb, u8 ptype)
{
ice_rx_hash(rx_ring, rx_desc, skb, ptype);
@@ -930,13 +964,12 @@ static void ice_process_skb_fields(struct ice_ring *rx_ring,
* This function sends the completed packet (via. skb) up the stack using
* gro receive functions (with/without vlan tag)
*/
-static void ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb,
- u16 vlan_tag)
+static void
+ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb, u16 vlan_tag)
{
if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
- (vlan_tag & VLAN_VID_MASK)) {
+ (vlan_tag & VLAN_VID_MASK))
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
- }
napi_gro_receive(&rx_ring->q_vector->napi, skb);
}
@@ -961,7 +994,9 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
/* start the loop to process RX packets bounded by 'budget' */
while (likely(total_rx_pkts < (unsigned int)budget)) {
union ice_32b_rx_flex_desc *rx_desc;
+ struct ice_rx_buf *rx_buf;
struct sk_buff *skb;
+ unsigned int size;
u16 stat_err_bits;
u16 vlan_tag = 0;
u8 rx_ptype;
@@ -991,11 +1026,24 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
*/
dma_rmb();
+ size = le16_to_cpu(rx_desc->wb.pkt_len) &
+ ICE_RX_FLX_DESC_PKT_LEN_M;
+
+ rx_buf = ice_get_rx_buf(rx_ring, &skb, size);
/* allocate (if needed) and populate skb */
- skb = ice_fetch_rx_buf(rx_ring, rx_desc);
- if (!skb)
+ if (skb)
+ ice_add_rx_frag(rx_buf, skb, size);
+ else
+ skb = ice_construct_skb(rx_ring, rx_buf, size);
+
+ /* exit if we failed to retrieve a buffer */
+ if (!skb) {
+ rx_ring->rx_stats.alloc_buf_failed++;
+ rx_buf->pagecnt_bias++;
break;
+ }
+ ice_put_rx_buf(rx_ring, rx_buf);
cleaned_count++;
/* skip if it is NOP desc */
@@ -1048,18 +1096,257 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget)
return failure ? budget : (int)total_rx_pkts;
}
+static unsigned int ice_itr_divisor(struct ice_port_info *pi)
+{
+ switch (pi->phy.link_info.link_speed) {
+ case ICE_AQ_LINK_SPEED_40GB:
+ return ICE_ITR_ADAPTIVE_MIN_INC * 1024;
+ case ICE_AQ_LINK_SPEED_25GB:
+ case ICE_AQ_LINK_SPEED_20GB:
+ return ICE_ITR_ADAPTIVE_MIN_INC * 512;
+ case ICE_AQ_LINK_SPEED_100MB:
+ return ICE_ITR_ADAPTIVE_MIN_INC * 32;
+ default:
+ return ICE_ITR_ADAPTIVE_MIN_INC * 256;
+ }
+}
+
+/**
+ * ice_update_itr - update the adaptive ITR value based on statistics
+ * @q_vector: structure containing interrupt and ring information
+ * @rc: structure containing ring performance data
+ *
+ * Stores a new ITR value based on packets and byte
+ * counts during the last interrupt. The advantage of per interrupt
+ * computation is faster updates and more accurate ITR for the current
+ * traffic pattern. Constants in this function were computed
+ * based on theoretical maximum wire speed and thresholds were set based
+ * on testing data as well as attempting to minimize response time
+ * while increasing bulk throughput.
+ */
+static void
+ice_update_itr(struct ice_q_vector *q_vector, struct ice_ring_container *rc)
+{
+ unsigned int avg_wire_size, packets, bytes, itr;
+ unsigned long next_update = jiffies;
+ bool container_is_rx;
+
+ if (!rc->ring || !ITR_IS_DYNAMIC(rc->itr_setting))
+ return;
+
+ /* If itr_countdown is set it means we programmed an ITR within
+ * the last 4 interrupt cycles. This has a side effect of us
+ * potentially firing an early interrupt. In order to work around
+ * this we need to throw out any data received for a few
+ * interrupts following the update.
+ */
+ if (q_vector->itr_countdown) {
+ itr = rc->target_itr;
+ goto clear_counts;
+ }
+
+ container_is_rx = (&q_vector->rx == rc);
+ /* For Rx we want to push the delay up and default to low latency.
+ * for Tx we want to pull the delay down and default to high latency.
+ */
+ itr = container_is_rx ?
+ ICE_ITR_ADAPTIVE_MIN_USECS | ICE_ITR_ADAPTIVE_LATENCY :
+ ICE_ITR_ADAPTIVE_MAX_USECS | ICE_ITR_ADAPTIVE_LATENCY;
+
+ /* If we didn't update within up to 1 - 2 jiffies we can assume
+ * that either packets are coming in so slow there hasn't been
+ * any work, or that there is so much work that NAPI is dealing
+ * with interrupt moderation and we don't need to do anything.
+ */
+ if (time_after(next_update, rc->next_update))
+ goto clear_counts;
+
+ packets = rc->total_pkts;
+ bytes = rc->total_bytes;
+
+ if (container_is_rx) {
+ /* If Rx there are 1 to 4 packets and bytes are less than
+ * 9000 assume insufficient data to use bulk rate limiting
+ * approach unless Tx is already in bulk rate limiting. We
+ * are likely latency driven.
+ */
+ if (packets && packets < 4 && bytes < 9000 &&
+ (q_vector->tx.target_itr & ICE_ITR_ADAPTIVE_LATENCY)) {
+ itr = ICE_ITR_ADAPTIVE_LATENCY;
+ goto adjust_by_size;
+ }
+ } else if (packets < 4) {
+ /* If we have Tx and Rx ITR maxed and Tx ITR is running in
+ * bulk mode and we are receiving 4 or fewer packets just
+ * reset the ITR_ADAPTIVE_LATENCY bit for latency mode so
+ * that the Rx can relax.
+ */
+ if (rc->target_itr == ICE_ITR_ADAPTIVE_MAX_USECS &&
+ (q_vector->rx.target_itr & ICE_ITR_MASK) ==
+ ICE_ITR_ADAPTIVE_MAX_USECS)
+ goto clear_counts;
+ } else if (packets > 32) {
+ /* If we have processed over 32 packets in a single interrupt
+ * for Tx assume we need to switch over to "bulk" mode.
+ */
+ rc->target_itr &= ~ICE_ITR_ADAPTIVE_LATENCY;
+ }
+
+ /* We have no packets to actually measure against. This means
+ * either one of the other queues on this vector is active or
+ * we are a Tx queue doing TSO with too high of an interrupt rate.
+ *
+ * Between 4 and 56 we can assume that our current interrupt delay
+ * is only slightly too low. As such we should increase it by a small
+ * fixed amount.
+ */
+ if (packets < 56) {
+ itr = rc->target_itr + ICE_ITR_ADAPTIVE_MIN_INC;
+ if ((itr & ICE_ITR_MASK) > ICE_ITR_ADAPTIVE_MAX_USECS) {
+ itr &= ICE_ITR_ADAPTIVE_LATENCY;
+ itr += ICE_ITR_ADAPTIVE_MAX_USECS;
+ }
+ goto clear_counts;
+ }
+
+ if (packets <= 256) {
+ itr = min(q_vector->tx.current_itr, q_vector->rx.current_itr);
+ itr &= ICE_ITR_MASK;
+
+ /* Between 56 and 112 is our "goldilocks" zone where we are
+ * working out "just right". Just report that our current
+ * ITR is good for us.
+ */
+ if (packets <= 112)
+ goto clear_counts;
+
+ /* If packet count is 128 or greater we are likely looking
+ * at a slight overrun of the delay we want. Try halving
+ * our delay to see if that will cut the number of packets
+ * in half per interrupt.
+ */
+ itr >>= 1;
+ itr &= ICE_ITR_MASK;
+ if (itr < ICE_ITR_ADAPTIVE_MIN_USECS)
+ itr = ICE_ITR_ADAPTIVE_MIN_USECS;
+
+ goto clear_counts;
+ }
+
+ /* The paths below assume we are dealing with a bulk ITR since
+ * number of packets is greater than 256. We are just going to have
+ * to compute a value and try to bring the count under control,
+ * though for smaller packet sizes there isn't much we can do as
+ * NAPI polling will likely be kicking in sooner rather than later.
+ */
+ itr = ICE_ITR_ADAPTIVE_BULK;
+
+adjust_by_size:
+ /* If packet counts are 256 or greater we can assume we have a gross
+ * overestimation of what the rate should be. Instead of trying to fine
+ * tune it just use the formula below to try and dial in an exact value
+ * gives the current packet size of the frame.
+ */
+ avg_wire_size = bytes / packets;
+
+ /* The following is a crude approximation of:
+ * wmem_default / (size + overhead) = desired_pkts_per_int
+ * rate / bits_per_byte / (size + ethernet overhead) = pkt_rate
+ * (desired_pkt_rate / pkt_rate) * usecs_per_sec = ITR value
+ *
+ * Assuming wmem_default is 212992 and overhead is 640 bytes per
+ * packet, (256 skb, 64 headroom, 320 shared info), we can reduce the
+ * formula down to
+ *
+ * (170 * (size + 24)) / (size + 640) = ITR
+ *
+ * We first do some math on the packet size and then finally bitshift
+ * by 8 after rounding up. We also have to account for PCIe link speed
+ * difference as ITR scales based on this.
+ */
+ if (avg_wire_size <= 60) {
+ /* Start at 250k ints/sec */
+ avg_wire_size = 4096;
+ } else if (avg_wire_size <= 380) {
+ /* 250K ints/sec to 60K ints/sec */
+ avg_wire_size *= 40;
+ avg_wire_size += 1696;
+ } else if (avg_wire_size <= 1084) {
+ /* 60K ints/sec to 36K ints/sec */
+ avg_wire_size *= 15;
+ avg_wire_size += 11452;
+ } else if (avg_wire_size <= 1980) {
+ /* 36K ints/sec to 30K ints/sec */
+ avg_wire_size *= 5;
+ avg_wire_size += 22420;
+ } else {
+ /* plateau at a limit of 30K ints/sec */
+ avg_wire_size = 32256;
+ }
+
+ /* If we are in low latency mode halve our delay which doubles the
+ * rate to somewhere between 100K to 16K ints/sec
+ */
+ if (itr & ICE_ITR_ADAPTIVE_LATENCY)
+ avg_wire_size >>= 1;
+
+ /* Resultant value is 256 times larger than it needs to be. This
+ * gives us room to adjust the value as needed to either increase
+ * or decrease the value based on link speeds of 10G, 2.5G, 1G, etc.
+ *
+ * Use addition as we have already recorded the new latency flag
+ * for the ITR value.
+ */
+ itr += DIV_ROUND_UP(avg_wire_size,
+ ice_itr_divisor(q_vector->vsi->port_info)) *
+ ICE_ITR_ADAPTIVE_MIN_INC;
+
+ if ((itr & ICE_ITR_MASK) > ICE_ITR_ADAPTIVE_MAX_USECS) {
+ itr &= ICE_ITR_ADAPTIVE_LATENCY;
+ itr += ICE_ITR_ADAPTIVE_MAX_USECS;
+ }
+
+clear_counts:
+ /* write back value */
+ rc->target_itr = itr;
+
+ /* next update should occur within next jiffy */
+ rc->next_update = next_update + 1;
+
+ rc->total_bytes = 0;
+ rc->total_pkts = 0;
+}
+
/**
* ice_buildreg_itr - build value for writing to the GLINT_DYN_CTL register
* @itr_idx: interrupt throttling index
- * @reg_itr: interrupt throttling value adjusted based on ITR granularity
+ * @itr: interrupt throttling value in usecs
*/
-static u32 ice_buildreg_itr(int itr_idx, u16 reg_itr)
+static u32 ice_buildreg_itr(u16 itr_idx, u16 itr)
{
+ /* The itr value is reported in microseconds, and the register value is
+ * recorded in 2 microsecond units. For this reason we only need to
+ * shift by the GLINT_DYN_CTL_INTERVAL_S - ICE_ITR_GRAN_S to apply this
+ * granularity as a shift instead of division. The mask makes sure the
+ * ITR value is never odd so we don't accidentally write into the field
+ * prior to the ITR field.
+ */
+ itr &= ICE_ITR_MASK;
+
return GLINT_DYN_CTL_INTENA_M | GLINT_DYN_CTL_CLEARPBA_M |
(itr_idx << GLINT_DYN_CTL_ITR_INDX_S) |
- (reg_itr << GLINT_DYN_CTL_INTERVAL_S);
+ (itr << (GLINT_DYN_CTL_INTERVAL_S - ICE_ITR_GRAN_S));
}
+/* The act of updating the ITR will cause it to immediately trigger. In order
+ * to prevent this from throwing off adaptive update statistics we defer the
+ * update so that it can only happen so often. So after either Tx or Rx are
+ * updated we make the adaptive scheme wait until either the ITR completely
+ * expires via the next_update expiration or we have been through at least
+ * 3 interrupts.
+ */
+#define ITR_COUNTDOWN_START 3
+
/**
* ice_update_ena_itr - Update ITR and re-enable MSIX interrupt
* @vsi: the VSI associated with the q_vector
@@ -1068,10 +1355,14 @@ static u32 ice_buildreg_itr(int itr_idx, u16 reg_itr)
static void
ice_update_ena_itr(struct ice_vsi *vsi, struct ice_q_vector *q_vector)
{
- struct ice_hw *hw = &vsi->back->hw;
- struct ice_ring_container *rc;
+ struct ice_ring_container *tx = &q_vector->tx;
+ struct ice_ring_container *rx = &q_vector->rx;
u32 itr_val;
+ /* This will do nothing if dynamic updates are not enabled */
+ ice_update_itr(q_vector, tx);
+ ice_update_itr(q_vector, rx);
+
/* This block of logic allows us to get away with only updating
* one ITR value with each interrupt. The idea is to perform a
* pseudo-lazy update with the following criteria.
@@ -1080,35 +1371,36 @@ ice_update_ena_itr(struct ice_vsi *vsi, struct ice_q_vector *q_vector)
* 2. If we must reduce an ITR that is given highest priority.
* 3. We then give priority to increasing ITR based on amount.
*/
- if (q_vector->rx.target_itr < q_vector->rx.current_itr) {
- rc = &q_vector->rx;
+ if (rx->target_itr < rx->current_itr) {
/* Rx ITR needs to be reduced, this is highest priority */
- itr_val = ice_buildreg_itr(rc->itr_idx, rc->target_itr);
- rc->current_itr = rc->target_itr;
- } else if ((q_vector->tx.target_itr < q_vector->tx.current_itr) ||
- ((q_vector->rx.target_itr - q_vector->rx.current_itr) <
- (q_vector->tx.target_itr - q_vector->tx.current_itr))) {
- rc = &q_vector->tx;
+ itr_val = ice_buildreg_itr(rx->itr_idx, rx->target_itr);
+ rx->current_itr = rx->target_itr;
+ q_vector->itr_countdown = ITR_COUNTDOWN_START;
+ } else if ((tx->target_itr < tx->current_itr) ||
+ ((rx->target_itr - rx->current_itr) <
+ (tx->target_itr - tx->current_itr))) {
/* Tx ITR needs to be reduced, this is second priority
* Tx ITR needs to be increased more than Rx, fourth priority
*/
- itr_val = ice_buildreg_itr(rc->itr_idx, rc->target_itr);
- rc->current_itr = rc->target_itr;
- } else if (q_vector->rx.current_itr != q_vector->rx.target_itr) {
- rc = &q_vector->rx;
+ itr_val = ice_buildreg_itr(tx->itr_idx, tx->target_itr);
+ tx->current_itr = tx->target_itr;
+ q_vector->itr_countdown = ITR_COUNTDOWN_START;
+ } else if (rx->current_itr != rx->target_itr) {
/* Rx ITR needs to be increased, third priority */
- itr_val = ice_buildreg_itr(rc->itr_idx, rc->target_itr);
- rc->current_itr = rc->target_itr;
+ itr_val = ice_buildreg_itr(rx->itr_idx, rx->target_itr);
+ rx->current_itr = rx->target_itr;
+ q_vector->itr_countdown = ITR_COUNTDOWN_START;
} else {
/* Still have to re-enable the interrupts */
itr_val = ice_buildreg_itr(ICE_ITR_NONE, 0);
+ if (q_vector->itr_countdown)
+ q_vector->itr_countdown--;
}
- if (!test_bit(__ICE_DOWN, vsi->state)) {
- int vector = vsi->hw_base_vector + q_vector->v_idx;
-
- wr32(hw, GLINT_DYN_CTL(vector), itr_val);
- }
+ if (!test_bit(__ICE_DOWN, vsi->state))
+ wr32(&vsi->back->hw,
+ GLINT_DYN_CTL(vsi->hw_base_vector + q_vector->v_idx),
+ itr_val);
}
/**
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index fc358ea81816..60131b84b021 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -47,6 +47,9 @@
#define ICE_TX_FLAGS_VLAN_M 0xffff0000
#define ICE_TX_FLAGS_VLAN_S 16
+#define ICE_RX_DMA_ATTR \
+ (DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
+
struct ice_tx_buf {
struct ice_tx_desc *next_to_watch;
struct sk_buff *skb;
@@ -73,6 +76,7 @@ struct ice_rx_buf {
dma_addr_t dma;
struct page *page;
unsigned int page_offset;
+ u16 pagecnt_bias;
};
struct ice_q_stats {
@@ -124,10 +128,17 @@ enum ice_rx_dtype {
#define ICE_ITR_DYNAMIC 0x8000 /* used as flag for itr_setting */
#define ITR_IS_DYNAMIC(setting) (!!((setting) & ICE_ITR_DYNAMIC))
#define ITR_TO_REG(setting) ((setting) & ~ICE_ITR_DYNAMIC)
-#define ICE_ITR_GRAN_S 1 /* Assume ITR granularity is 2us */
+#define ICE_ITR_GRAN_S 1 /* ITR granularity is always 2us */
+#define ICE_ITR_GRAN_US BIT(ICE_ITR_GRAN_S)
#define ICE_ITR_MASK 0x1FFE /* ITR register value alignment mask */
#define ITR_REG_ALIGN(setting) __ALIGN_MASK(setting, ~ICE_ITR_MASK)
+#define ICE_ITR_ADAPTIVE_MIN_INC 0x0002
+#define ICE_ITR_ADAPTIVE_MIN_USECS 0x0002
+#define ICE_ITR_ADAPTIVE_MAX_USECS 0x00FA
+#define ICE_ITR_ADAPTIVE_LATENCY 0x8000
+#define ICE_ITR_ADAPTIVE_BULK 0x0000
+
#define ICE_DFLT_INTRL 0
/* Legacy or Advanced Mode Queue */
@@ -173,21 +184,13 @@ struct ice_ring {
u16 next_to_alloc;
} ____cacheline_internodealigned_in_smp;
-enum ice_latency_range {
- ICE_LOWEST_LATENCY = 0,
- ICE_LOW_LATENCY = 1,
- ICE_BULK_LATENCY = 2,
- ICE_ULTRA_LATENCY = 3,
-};
-
struct ice_ring_container {
/* head of linked-list of rings */
struct ice_ring *ring;
unsigned long next_update; /* jiffies value of next queue update */
unsigned int total_bytes; /* total bytes processed this int */
unsigned int total_pkts; /* total packets processed this int */
- enum ice_latency_range latency_range;
- int itr_idx; /* index in the interrupt vector */
+ u16 itr_idx; /* index in the interrupt vector */
u16 target_itr; /* value in usecs divided by the hw->itr_gran */
u16 current_itr; /* value in usecs divided by the hw->itr_gran */
/* high bit set means dynamic ITR, rest is used to store user
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 17086d5b5c33..3a4e67484487 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -24,6 +24,7 @@ static inline bool ice_is_tc_ena(u8 bitmap, u8 tc)
/* debug masks - set these bits in hw->debug_mask to control output */
#define ICE_DBG_INIT BIT_ULL(1)
#define ICE_DBG_LINK BIT_ULL(4)
+#define ICE_DBG_PHY BIT_ULL(5)
#define ICE_DBG_QCTX BIT_ULL(6)
#define ICE_DBG_NVM BIT_ULL(7)
#define ICE_DBG_LAN BIT_ULL(8)
@@ -209,6 +210,9 @@ struct ice_nvm_info {
#define ICE_MAX_TRAFFIC_CLASS 8
#define ICE_TXSCHED_MAX_BRANCHES ICE_MAX_TRAFFIC_CLASS
+#define ice_for_each_traffic_class(_i) \
+ for ((_i) = 0; (_i) < ICE_MAX_TRAFFIC_CLASS; (_i)++)
+
struct ice_sched_node {
struct ice_sched_node *parent;
struct ice_sched_node *sibling; /* next sibling in the same layer */
@@ -247,7 +251,6 @@ struct ice_sched_vsi_info {
struct ice_sched_node *ag_node[ICE_MAX_TRAFFIC_CLASS];
struct list_head list_entry;
u16 max_lanq[ICE_MAX_TRAFFIC_CLASS];
- u16 vsi_id;
};
/* driver defines the policy */
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index 57155b4a59dc..84e51a0a0795 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -5,6 +5,37 @@
#include "ice_lib.h"
/**
+ * ice_err_to_virt err - translate errors for VF return code
+ * @ice_err: error return code
+ */
+static enum virtchnl_status_code ice_err_to_virt_err(enum ice_status ice_err)
+{
+ switch (ice_err) {
+ case ICE_SUCCESS:
+ return VIRTCHNL_STATUS_SUCCESS;
+ case ICE_ERR_BAD_PTR:
+ case ICE_ERR_INVAL_SIZE:
+ case ICE_ERR_DEVICE_NOT_SUPPORTED:
+ case ICE_ERR_PARAM:
+ case ICE_ERR_CFG:
+ return VIRTCHNL_STATUS_ERR_PARAM;
+ case ICE_ERR_NO_MEMORY:
+ return VIRTCHNL_STATUS_ERR_NO_MEMORY;
+ case ICE_ERR_NOT_READY:
+ case ICE_ERR_RESET_FAILED:
+ case ICE_ERR_FW_API_VER:
+ case ICE_ERR_AQ_ERROR:
+ case ICE_ERR_AQ_TIMEOUT:
+ case ICE_ERR_AQ_FULL:
+ case ICE_ERR_AQ_NO_WORK:
+ case ICE_ERR_AQ_EMPTY:
+ return VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR;
+ default:
+ return VIRTCHNL_STATUS_ERR_NOT_SUPPORTED;
+ }
+}
+
+/**
* ice_vc_vf_broadcast - Broadcast a message to all VFs on PF
* @pf: pointer to the PF structure
* @v_opcode: operation code
@@ -14,7 +45,7 @@
*/
static void
ice_vc_vf_broadcast(struct ice_pf *pf, enum virtchnl_ops v_opcode,
- enum ice_status v_retval, u8 *msg, u16 msglen)
+ enum virtchnl_status_code v_retval, u8 *msg, u16 msglen)
{
struct ice_hw *hw = &pf->hw;
struct ice_vf *vf = pf->vf;
@@ -104,7 +135,8 @@ static void ice_vc_notify_vf_link_state(struct ice_vf *vf)
ice_set_pfe_link(vf, &pfe, ls->link_speed, ls->link_info &
ICE_AQ_LINK_UP);
- ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, 0, (u8 *)&pfe,
+ ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT,
+ VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe,
sizeof(pfe), NULL);
}
@@ -343,11 +375,41 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr)
}
/**
- * ice_vsi_set_pvid - Set port VLAN id for the VSI
- * @vsi: the VSI being changed
+ * ice_vsi_set_pvid_fill_ctxt - Set VSI ctxt for add pvid
+ * @ctxt: the vsi ctxt to fill
* @vid: the VLAN id to set as a PVID
*/
-static int ice_vsi_set_pvid(struct ice_vsi *vsi, u16 vid)
+static void ice_vsi_set_pvid_fill_ctxt(struct ice_vsi_ctx *ctxt, u16 vid)
+{
+ ctxt->info.vlan_flags = (ICE_AQ_VSI_VLAN_MODE_UNTAGGED |
+ ICE_AQ_VSI_PVLAN_INSERT_PVID |
+ ICE_AQ_VSI_VLAN_EMOD_STR);
+ ctxt->info.pvid = cpu_to_le16(vid);
+ ctxt->info.sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
+ ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID |
+ ICE_AQ_VSI_PROP_SW_VALID);
+}
+
+/**
+ * ice_vsi_kill_pvid_fill_ctxt - Set VSI ctx for remove pvid
+ * @ctxt: the VSI ctxt to fill
+ */
+static void ice_vsi_kill_pvid_fill_ctxt(struct ice_vsi_ctx *ctxt)
+{
+ ctxt->info.vlan_flags = ICE_AQ_VSI_VLAN_EMOD_NOTHING;
+ ctxt->info.vlan_flags |= ICE_AQ_VSI_VLAN_MODE_ALL;
+ ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
+ ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID |
+ ICE_AQ_VSI_PROP_SW_VALID);
+}
+
+/**
+ * ice_vsi_manage_pvid - Enable or disable port VLAN for VSI
+ * @vsi: the VSI to update
+ * @vid: the VLAN id to set as a PVID
+ * @enable: true for enable pvid false for disable
+ */
+static int ice_vsi_manage_pvid(struct ice_vsi *vsi, u16 vid, bool enable)
{
struct device *dev = &vsi->back->pdev->dev;
struct ice_hw *hw = &vsi->back->hw;
@@ -359,46 +421,27 @@ static int ice_vsi_set_pvid(struct ice_vsi *vsi, u16 vid)
if (!ctxt)
return -ENOMEM;
- ctxt->info.vlan_flags = (ICE_AQ_VSI_VLAN_MODE_UNTAGGED |
- ICE_AQ_VSI_PVLAN_INSERT_PVID |
- ICE_AQ_VSI_VLAN_EMOD_STR);
- ctxt->info.pvid = cpu_to_le16(vid);
- ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID);
+ ctxt->info = vsi->info;
+ if (enable)
+ ice_vsi_set_pvid_fill_ctxt(ctxt, vid);
+ else
+ ice_vsi_kill_pvid_fill_ctxt(ctxt);
status = ice_update_vsi(hw, vsi->idx, ctxt, NULL);
if (status) {
- dev_info(dev, "update VSI for VLAN insert failed, err %d aq_err %d\n",
+ dev_info(dev, "update VSI for port VLAN failed, err %d aq_err %d\n",
status, hw->adminq.sq_last_status);
ret = -EIO;
goto out;
}
- vsi->info.pvid = ctxt->info.pvid;
- vsi->info.vlan_flags = ctxt->info.vlan_flags;
+ vsi->info = ctxt->info;
out:
devm_kfree(dev, ctxt);
return ret;
}
/**
- * ice_vsi_kill_pvid - Remove port VLAN id from the VSI
- * @vsi: the VSI being changed
- */
-static int ice_vsi_kill_pvid(struct ice_vsi *vsi)
-{
- struct ice_pf *pf = vsi->back;
-
- if (ice_vsi_manage_vlan_stripping(vsi, false)) {
- dev_err(&pf->pdev->dev, "Error removing Port VLAN on VSI %i\n",
- vsi->vsi_num);
- return -ENODEV;
- }
-
- vsi->info.pvid = 0;
- return 0;
-}
-
-/**
* ice_vf_vsi_setup - Set up a VF VSI
* @pf: board private structure
* @pi: pointer to the port_info instance
@@ -446,8 +489,10 @@ static int ice_alloc_vsi_res(struct ice_vf *vf)
vsi->hw_base_vector += 1;
/* Check if port VLAN exist before, and restore it accordingly */
- if (vf->port_vlan_id)
- ice_vsi_set_pvid(vsi, vf->port_vlan_id);
+ if (vf->port_vlan_id) {
+ ice_vsi_manage_pvid(vsi, vf->port_vlan_id, true);
+ ice_vsi_add_vlan(vsi, vf->port_vlan_id & ICE_VLAN_M);
+ }
eth_broadcast_addr(broadcast);
@@ -484,6 +529,8 @@ ice_alloc_vsi_res_exit:
*/
static int ice_alloc_vf_res(struct ice_vf *vf)
{
+ struct ice_pf *pf = vf->pf;
+ int tx_rx_queue_left;
int status;
/* setup VF VSI and necessary resources */
@@ -491,6 +538,15 @@ static int ice_alloc_vf_res(struct ice_vf *vf)
if (status)
goto ice_alloc_vf_res_exit;
+ /* Update number of VF queues, in case VF had requested for queue
+ * changes
+ */
+ tx_rx_queue_left = min_t(int, pf->q_left_tx, pf->q_left_rx);
+ tx_rx_queue_left += ICE_DFLT_QS_PER_VF;
+ if (vf->num_req_qs && vf->num_req_qs <= tx_rx_queue_left &&
+ vf->num_req_qs != vf->num_vf_qs)
+ vf->num_vf_qs = vf->num_req_qs;
+
if (vf->trusted)
set_bit(ICE_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps);
else
@@ -548,6 +604,10 @@ static void ice_ena_vf_mappings(struct ice_vf *vf)
wr32(hw, GLINT_VECT2FUNC(v), reg);
}
+ /* Map mailbox interrupt. We put an explicit 0 here to remind us that
+ * VF admin queue interrupts will go to VF MSI-X vector 0.
+ */
+ wr32(hw, VPINT_MBX_CTL(abs_vf_id), VPINT_MBX_CTL_CAUSE_ENA_M | 0);
/* set regardless of mapping mode */
wr32(hw, VPLAN_TXQ_MAPENA(vf->vf_id), VPLAN_TXQ_MAPENA_TX_ENA_M);
@@ -750,6 +810,47 @@ static void ice_cleanup_and_realloc_vf(struct ice_vf *vf)
}
/**
+ * ice_vf_set_vsi_promisc - set given VF VSI to given promiscuous mode(s)
+ * @vf: pointer to the VF info
+ * @vsi: the VSI being configured
+ * @promisc_m: mask of promiscuous config bits
+ * @rm_promisc: promisc flag request from the VF to remove or add filter
+ *
+ * This function configures VF VSI promiscuous mode, based on the VF requests,
+ * for Unicast, Multicast and VLAN
+ */
+static enum ice_status
+ice_vf_set_vsi_promisc(struct ice_vf *vf, struct ice_vsi *vsi, u8 promisc_m,
+ bool rm_promisc)
+{
+ struct ice_pf *pf = vf->pf;
+ enum ice_status status = 0;
+ struct ice_hw *hw;
+
+ hw = &pf->hw;
+ if (vf->num_vlan) {
+ status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m,
+ rm_promisc);
+ } else if (vf->port_vlan_id) {
+ if (rm_promisc)
+ status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m,
+ vf->port_vlan_id);
+ else
+ status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m,
+ vf->port_vlan_id);
+ } else {
+ if (rm_promisc)
+ status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m,
+ 0);
+ else
+ status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m,
+ 0);
+ }
+
+ return status;
+}
+
+/**
* ice_reset_all_vfs - reset all allocated VFs in one go
* @pf: pointer to the PF structure
* @is_vflr: true if VFLR was issued, false if not
@@ -764,6 +865,7 @@ static void ice_cleanup_and_realloc_vf(struct ice_vf *vf)
bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
{
struct ice_hw *hw = &pf->hw;
+ struct ice_vf *vf;
int v, i;
/* If we don't have any VFs, then there is nothing to reset */
@@ -778,12 +880,17 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
for (v = 0; v < pf->num_alloc_vfs; v++)
ice_trigger_vf_reset(&pf->vf[v], is_vflr);
- /* Call Disable LAN Tx queue AQ call with VFR bit set and 0
- * queues to inform Firmware about VF reset.
- */
- for (v = 0; v < pf->num_alloc_vfs; v++)
- ice_dis_vsi_txq(pf->vsi[0]->port_info, 0, NULL, NULL,
- ICE_VF_RESET, v, NULL);
+ for (v = 0; v < pf->num_alloc_vfs; v++) {
+ struct ice_vsi *vsi;
+
+ vf = &pf->vf[v];
+ vsi = pf->vsi[vf->lan_vsi_idx];
+ if (test_bit(ICE_VF_STATE_ENA, vf->vf_states)) {
+ ice_vsi_stop_lan_tx_rings(vsi, ICE_VF_RESET, vf->vf_id);
+ ice_vsi_stop_rx_rings(vsi);
+ clear_bit(ICE_VF_STATE_ENA, vf->vf_states);
+ }
+ }
/* HW requires some time to make sure it can flush the FIFO for a VF
* when it resets it. Poll the VPGEN_VFRSTAT register for each VF in
@@ -796,9 +903,9 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
/* Check each VF in sequence */
while (v < pf->num_alloc_vfs) {
- struct ice_vf *vf = &pf->vf[v];
u32 reg;
+ vf = &pf->vf[v];
reg = rd32(hw, VPGEN_VFRSTAT(vf->vf_id));
if (!(reg & VPGEN_VFRSTAT_VFRD_M))
break;
@@ -818,8 +925,18 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
usleep_range(10000, 20000);
/* free VF resources to begin resetting the VSI state */
- for (v = 0; v < pf->num_alloc_vfs; v++)
- ice_free_vf_res(&pf->vf[v]);
+ for (v = 0; v < pf->num_alloc_vfs; v++) {
+ vf = &pf->vf[v];
+
+ ice_free_vf_res(vf);
+
+ /* Free VF queues as well, and reallocate later.
+ * If a given VF has different number of queues
+ * configured, the request for update will come
+ * via mailbox communication.
+ */
+ vf->num_vf_qs = 0;
+ }
if (ice_check_avail_res(pf)) {
dev_err(&pf->pdev->dev,
@@ -828,8 +945,15 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
}
/* Finish the reset on each VF */
- for (v = 0; v < pf->num_alloc_vfs; v++)
- ice_cleanup_and_realloc_vf(&pf->vf[v]);
+ for (v = 0; v < pf->num_alloc_vfs; v++) {
+ vf = &pf->vf[v];
+
+ vf->num_vf_qs = pf->num_vf_qps;
+ dev_dbg(&pf->pdev->dev,
+ "VF-id %d has %d queues configured\n",
+ vf->vf_id, vf->num_vf_qs);
+ ice_cleanup_and_realloc_vf(vf);
+ }
ice_flush(hw);
clear_bit(__ICE_VF_DIS, pf->state);
@@ -847,9 +971,10 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr)
static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr)
{
struct ice_pf *pf = vf->pf;
- struct ice_hw *hw = &pf->hw;
struct ice_vsi *vsi;
+ struct ice_hw *hw;
bool rsd = false;
+ u8 promisc_m;
u32 reg;
int i;
@@ -875,6 +1000,7 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr)
vf->vf_id, NULL);
}
+ hw = &pf->hw;
/* poll VPGEN_VFRSTAT reg to make sure
* that reset is complete
*/
@@ -900,6 +1026,21 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr)
usleep_range(10000, 20000);
+ /* disable promiscuous modes in case they were enabled
+ * ignore any error if disabling process failed
+ */
+ if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) ||
+ test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) {
+ if (vf->port_vlan_id || vf->num_vlan)
+ promisc_m = ICE_UCAST_VLAN_PROMISC_BITS;
+ else
+ promisc_m = ICE_UCAST_PROMISC_BITS;
+
+ vsi = pf->vsi[vf->lan_vsi_idx];
+ if (ice_vf_set_vsi_promisc(vf, vsi, promisc_m, true))
+ dev_err(&pf->pdev->dev, "disabling promiscuous mode failed\n");
+ }
+
/* free VF resources to begin resetting the VSI state */
ice_free_vf_res(vf);
@@ -938,7 +1079,7 @@ void ice_vc_notify_reset(struct ice_pf *pf)
pfe.event = VIRTCHNL_EVENT_RESET_IMPENDING;
pfe.severity = PF_EVENT_SEVERITY_CERTAIN_DOOM;
- ice_vc_vf_broadcast(pf, VIRTCHNL_OP_EVENT, ICE_SUCCESS,
+ ice_vc_vf_broadcast(pf, VIRTCHNL_OP_EVENT, VIRTCHNL_STATUS_SUCCESS,
(u8 *)&pfe, sizeof(struct virtchnl_pf_event));
}
@@ -961,8 +1102,9 @@ static void ice_vc_notify_vf_reset(struct ice_vf *vf)
pfe.event = VIRTCHNL_EVENT_RESET_IMPENDING;
pfe.severity = PF_EVENT_SEVERITY_CERTAIN_DOOM;
- ice_aq_send_msg_to_vf(&vf->pf->hw, vf->vf_id, VIRTCHNL_OP_EVENT, 0,
- (u8 *)&pfe, sizeof(pfe), NULL);
+ ice_aq_send_msg_to_vf(&vf->pf->hw, vf->vf_id, VIRTCHNL_OP_EVENT,
+ VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe, sizeof(pfe),
+ NULL);
}
/**
@@ -1012,7 +1154,7 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs)
pf->num_alloc_vfs = num_alloc_vfs;
/* VF resources get allocated during reset */
- if (!ice_reset_all_vfs(pf, false))
+ if (!ice_reset_all_vfs(pf, true))
goto err_unroll_sriov;
goto err_unroll_intr;
@@ -1182,8 +1324,9 @@ static void ice_vc_dis_vf(struct ice_vf *vf)
*
* send msg to VF
*/
-static int ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode,
- enum ice_status v_retval, u8 *msg, u16 msglen)
+static int
+ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode,
+ enum virtchnl_status_code v_retval, u8 *msg, u16 msglen)
{
enum ice_status aq_ret;
struct ice_pf *pf;
@@ -1243,8 +1386,8 @@ static int ice_vc_get_ver_msg(struct ice_vf *vf, u8 *msg)
if (VF_IS_V10(&vf->vf_ver))
info.minor = VIRTCHNL_VERSION_MINOR_NO_VF_CAPS;
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_VERSION, ICE_SUCCESS,
- (u8 *)&info,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_VERSION,
+ VIRTCHNL_STATUS_SUCCESS, (u8 *)&info,
sizeof(struct virtchnl_version_info));
}
@@ -1257,15 +1400,15 @@ static int ice_vc_get_ver_msg(struct ice_vf *vf, u8 *msg)
*/
static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_vf_resource *vfres = NULL;
- enum ice_status aq_ret = 0;
struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi;
int len = 0;
int ret;
if (!test_bit(ICE_VF_STATE_INIT, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto err;
}
@@ -1273,7 +1416,7 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg)
vfres = devm_kzalloc(&pf->pdev->dev, len, GFP_KERNEL);
if (!vfres) {
- aq_ret = ICE_ERR_NO_MEMORY;
+ v_ret = VIRTCHNL_STATUS_ERR_NO_MEMORY;
len = 0;
goto err;
}
@@ -1286,6 +1429,11 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg)
vfres->vf_cap_flags = VIRTCHNL_VF_OFFLOAD_L2;
vsi = pf->vsi[vf->lan_vsi_idx];
+ if (!vsi) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ goto err;
+ }
+
if (!vsi->info.pvid)
vfres->vf_cap_flags |= VIRTCHNL_VF_OFFLOAD_VLAN;
@@ -1336,7 +1484,7 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg)
err:
/* send the response back to the VF */
- ret = ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_VF_RESOURCES, aq_ret,
+ ret = ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_VF_RESOURCES, v_ret,
(u8 *)vfres, len);
devm_kfree(&pf->pdev->dev, vfres);
@@ -1368,7 +1516,7 @@ static struct ice_vsi *ice_find_vsi_from_id(struct ice_pf *pf, u16 id)
{
int i;
- for (i = 0; i < pf->num_alloc_vsi; i++)
+ ice_for_each_vsi(pf, i)
if (pf->vsi[i] && pf->vsi[i]->vsi_num == id)
return pf->vsi[i];
@@ -1416,42 +1564,42 @@ static bool ice_vc_isvalid_q_id(struct ice_vf *vf, u16 vsi_id, u8 qid)
*/
static int ice_vc_config_rss_key(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_rss_key *vrk =
(struct virtchnl_rss_key *)msg;
+ struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi = NULL;
- enum ice_status aq_ret;
- int ret;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, vrk->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, vrk->vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (vrk->key_len != ICE_VSIQF_HKEY_ARRAY_SIZE) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- ret = ice_set_rss(vsi, vrk->key, NULL, 0);
- aq_ret = ret ? ICE_ERR_PARAM : ICE_SUCCESS;
+ if (ice_set_rss(vsi, vrk->key, NULL, 0))
+ v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR;
error_param:
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_KEY, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_KEY, v_ret,
NULL, 0);
}
@@ -1465,40 +1613,40 @@ error_param:
static int ice_vc_config_rss_lut(struct ice_vf *vf, u8 *msg)
{
struct virtchnl_rss_lut *vrl = (struct virtchnl_rss_lut *)msg;
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
+ struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi = NULL;
- enum ice_status aq_ret;
- int ret;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, vrl->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, vrl->vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (vrl->lut_entries != ICE_VSIQF_HLUT_ARRAY_SIZE) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- ret = ice_set_rss(vsi, NULL, vrl->lut, ICE_VSIQF_HLUT_ARRAY_SIZE);
- aq_ret = ret ? ICE_ERR_PARAM : ICE_SUCCESS;
+ if (ice_set_rss(vsi, NULL, vrl->lut, ICE_VSIQF_HLUT_ARRAY_SIZE))
+ v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR;
error_param:
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_LUT, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_LUT, v_ret,
NULL, 0);
}
@@ -1511,25 +1659,26 @@ error_param:
*/
static int ice_vc_get_stats_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_queue_select *vqs =
(struct virtchnl_queue_select *)msg;
- enum ice_status aq_ret = 0;
+ struct ice_pf *pf = vf->pf;
struct ice_eth_stats stats;
struct ice_vsi *vsi;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, vqs->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, vqs->vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -1540,7 +1689,7 @@ static int ice_vc_get_stats_msg(struct ice_vf *vf, u8 *msg)
error_param:
/* send the response to the VF */
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_STATS, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_STATS, v_ret,
(u8 *)&stats, sizeof(stats));
}
@@ -1553,29 +1702,30 @@ error_param:
*/
static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_queue_select *vqs =
(struct virtchnl_queue_select *)msg;
- enum ice_status aq_ret = 0;
+ struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, vqs->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!vqs->rx_queues && !vqs->tx_queues) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, vqs->vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -1584,15 +1734,15 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg)
* programmed using ice_vsi_cfg_txqs
*/
if (ice_vsi_start_rx_rings(vsi))
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
/* Set flag to indicate that queues are enabled */
- if (!aq_ret)
+ if (v_ret == VIRTCHNL_STATUS_SUCCESS)
set_bit(ICE_VF_STATE_ENA, vf->vf_states);
error_param:
/* send the response to the VF */
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_QUEUES, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_QUEUES, v_ret,
NULL, 0);
}
@@ -1606,30 +1756,31 @@ error_param:
*/
static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_queue_select *vqs =
(struct virtchnl_queue_select *)msg;
- enum ice_status aq_ret = 0;
+ struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) &&
!test_bit(ICE_VF_STATE_ENA, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, vqs->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!vqs->rx_queues && !vqs->tx_queues) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, vqs->vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -1637,23 +1788,23 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg)
dev_err(&vsi->back->pdev->dev,
"Failed to stop tx rings on VSI %d\n",
vsi->vsi_num);
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
}
if (ice_vsi_stop_rx_rings(vsi)) {
dev_err(&vsi->back->pdev->dev,
"Failed to stop rx rings on VSI %d\n",
vsi->vsi_num);
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
}
/* Clear enabled queues flag */
- if (!aq_ret)
+ if (v_ret == VIRTCHNL_STATUS_SUCCESS)
clear_bit(ICE_VF_STATE_ENA, vf->vf_states);
error_param:
/* send the response to the VF */
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_QUEUES, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_QUEUES, v_ret,
NULL, 0);
}
@@ -1666,18 +1817,18 @@ error_param:
*/
static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_irq_map_info *irqmap_info =
(struct virtchnl_irq_map_info *)msg;
u16 vsi_id, vsi_q_id, vector_id;
struct virtchnl_vector_map *map;
struct ice_vsi *vsi = NULL;
struct ice_pf *pf = vf->pf;
- enum ice_status aq_ret = 0;
unsigned long qmap;
int i;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -1689,13 +1840,13 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg)
/* validate msg params */
if (!(vector_id < pf->hw.func_caps.common_cap
.num_msix_vectors) || !ice_vc_isvalid_vsi_id(vf, vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -1705,7 +1856,7 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg)
struct ice_q_vector *q_vector;
if (!ice_vc_isvalid_q_id(vf, vsi_id, vsi_q_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
q_vector = vsi->q_vectors[i];
@@ -1719,7 +1870,7 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg)
struct ice_q_vector *q_vector;
if (!ice_vc_isvalid_q_id(vf, vsi_id, vsi_q_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
q_vector = vsi->q_vectors[i];
@@ -1733,7 +1884,7 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg)
ice_vsi_cfg_msix(vsi);
error_param:
/* send the response to the VF */
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_IRQ_MAP, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_IRQ_MAP, v_ret,
NULL, 0);
}
@@ -1746,26 +1897,34 @@ error_param:
*/
static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_vsi_queue_config_info *qci =
(struct virtchnl_vsi_queue_config_info *)msg;
struct virtchnl_queue_pair_info *qpi;
- enum ice_status aq_ret = 0;
+ struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi;
int i;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, qci->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- vsi = ice_find_vsi_from_id(vf->pf, qci->vsi_id);
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ goto error_param;
+ }
+
+ if (qci->num_queue_pairs > ICE_MAX_BASE_QS_PER_VF) {
+ dev_err(&pf->pdev->dev,
+ "VF-%d requesting more than supported number of queues: %d\n",
+ vf->vf_id, qci->num_queue_pairs);
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -1775,7 +1934,7 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg)
qpi->rxq.vsi_id != qci->vsi_id ||
qpi->rxq.queue_id != qpi->txq.queue_id ||
!ice_vc_isvalid_q_id(vf, qci->vsi_id, qpi->txq.queue_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
/* copy Tx queue info from VF into VSI */
@@ -1785,13 +1944,13 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg)
vsi->rx_rings[i]->dma = qpi->rxq.dma_ring_addr;
vsi->rx_rings[i]->count = qpi->rxq.ring_len;
if (qpi->rxq.databuffer_size > ((16 * 1024) - 128)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
vsi->rx_buf_len = qpi->rxq.databuffer_size;
if (qpi->rxq.max_pkt_size >= (16 * 1024) ||
qpi->rxq.max_pkt_size < 64) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
vsi->max_frame = qpi->rxq.max_pkt_size;
@@ -1802,15 +1961,16 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg)
*/
vsi->num_txq = qci->num_queue_pairs;
vsi->num_rxq = qci->num_queue_pairs;
+ /* All queues of VF VSI are in TC 0 */
+ vsi->tc_cfg.tc_info[0].qcount_tx = qci->num_queue_pairs;
+ vsi->tc_cfg.tc_info[0].qcount_rx = qci->num_queue_pairs;
- if (!ice_vsi_cfg_lan_txqs(vsi) && !ice_vsi_cfg_rxqs(vsi))
- aq_ret = 0;
- else
- aq_ret = ICE_ERR_PARAM;
+ if (ice_vsi_cfg_lan_txqs(vsi) || ice_vsi_cfg_rxqs(vsi))
+ v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR;
error_param:
/* send the response to the VF */
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_VSI_QUEUES, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_VSI_QUEUES, v_ret,
NULL, 0);
}
@@ -1852,11 +2012,11 @@ static bool ice_can_vf_change_mac(struct ice_vf *vf)
static int
ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_ether_addr_list *al =
(struct virtchnl_ether_addr_list *)msg;
struct ice_pf *pf = vf->pf;
enum virtchnl_ops vc_op;
- enum ice_status ret;
LIST_HEAD(mac_list);
struct ice_vsi *vsi;
int mac_count = 0;
@@ -1869,19 +2029,27 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) ||
!ice_vc_isvalid_vsi_id(vf, al->vsi_id)) {
- ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto handle_mac_exit;
}
if (set && !ice_is_vf_trusted(vf) &&
(vf->num_mac + al->num_elements) > ICE_MAX_MACADDR_PER_VF) {
dev_err(&pf->pdev->dev,
- "Can't add more MAC addresses, because VF is not trusted, switch the VF to trusted mode in order to add more functionalities\n");
- ret = ICE_ERR_PARAM;
+ "Can't add more MAC addresses, because VF-%d is not trusted, switch the VF to trusted mode in order to add more functionalities\n",
+ vf->vf_id);
+ /* There is no need to let VF know about not being trusted
+ * to add more MAC addr, so we can just return success message.
+ */
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto handle_mac_exit;
}
vsi = pf->vsi[vf->lan_vsi_idx];
+ if (!vsi) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ goto handle_mac_exit;
+ }
for (i = 0; i < al->num_elements; i++) {
u8 *maddr = al->list[i].addr;
@@ -1893,40 +2061,39 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
* already added. Just continue.
*/
dev_info(&pf->pdev->dev,
- "mac %pM already set for VF %d\n",
+ "MAC %pM already set for VF %d\n",
maddr, vf->vf_id);
continue;
} else {
/* VF can't remove dflt_lan_addr/bcast mac */
dev_err(&pf->pdev->dev,
- "can't remove mac %pM for VF %d\n",
+ "VF can't remove default MAC address or MAC %pM programmed by PF for VF %d\n",
maddr, vf->vf_id);
- ret = ICE_ERR_PARAM;
- goto handle_mac_exit;
+ continue;
}
}
/* check for the invalid cases and bail if necessary */
if (is_zero_ether_addr(maddr)) {
dev_err(&pf->pdev->dev,
- "invalid mac %pM provided for VF %d\n",
+ "invalid MAC %pM provided for VF %d\n",
maddr, vf->vf_id);
- ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto handle_mac_exit;
}
if (is_unicast_ether_addr(maddr) &&
!ice_can_vf_change_mac(vf)) {
dev_err(&pf->pdev->dev,
- "can't change unicast mac for untrusted VF %d\n",
+ "can't change unicast MAC for untrusted VF %d\n",
vf->vf_id);
- ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto handle_mac_exit;
}
/* get here if maddr is multicast or if VF can change mac */
if (ice_add_mac_to_list(vsi, &mac_list, al->list[i].addr)) {
- ret = ICE_ERR_NO_MEMORY;
+ v_ret = VIRTCHNL_STATUS_ERR_NO_MEMORY;
goto handle_mac_exit;
}
mac_count++;
@@ -1934,14 +2101,14 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
/* program the updated filter list */
if (set)
- ret = ice_add_mac(&pf->hw, &mac_list);
+ v_ret = ice_err_to_virt_err(ice_add_mac(&pf->hw, &mac_list));
else
- ret = ice_remove_mac(&pf->hw, &mac_list);
+ v_ret = ice_err_to_virt_err(ice_remove_mac(&pf->hw, &mac_list));
- if (ret) {
+ if (v_ret) {
dev_err(&pf->pdev->dev,
- "can't update mac filters for VF %d, error %d\n",
- vf->vf_id, ret);
+ "can't update MAC filters for VF %d, error %d\n",
+ vf->vf_id, v_ret);
} else {
if (set)
vf->num_mac += mac_count;
@@ -1952,7 +2119,7 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set)
handle_mac_exit:
ice_free_fltr_list(&pf->pdev->dev, &mac_list);
/* send the response to the VF */
- return ice_vc_send_msg_to_vf(vf, vc_op, ret, NULL, 0);
+ return ice_vc_send_msg_to_vf(vf, vc_op, v_ret, NULL, 0);
}
/**
@@ -1991,35 +2158,38 @@ static int ice_vc_del_mac_addr_msg(struct ice_vf *vf, u8 *msg)
*/
static int ice_vc_request_qs_msg(struct ice_vf *vf, u8 *msg)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_vf_res_request *vfres =
(struct virtchnl_vf_res_request *)msg;
int req_queues = vfres->num_queue_pairs;
- enum ice_status aq_ret = 0;
struct ice_pf *pf = vf->pf;
+ int max_allowed_vf_queues;
int tx_rx_queue_left;
int cur_queues;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
- cur_queues = pf->num_vf_qps;
+ cur_queues = vf->num_vf_qs;
tx_rx_queue_left = min_t(int, pf->q_left_tx, pf->q_left_rx);
+ max_allowed_vf_queues = tx_rx_queue_left + cur_queues;
if (req_queues <= 0) {
dev_err(&pf->pdev->dev,
"VF %d tried to request %d queues. Ignoring.\n",
vf->vf_id, req_queues);
- } else if (req_queues > ICE_MAX_QS_PER_VF) {
+ } else if (req_queues > ICE_MAX_BASE_QS_PER_VF) {
dev_err(&pf->pdev->dev,
"VF %d tried to request more than %d queues.\n",
- vf->vf_id, ICE_MAX_QS_PER_VF);
- vfres->num_queue_pairs = ICE_MAX_QS_PER_VF;
+ vf->vf_id, ICE_MAX_BASE_QS_PER_VF);
+ vfres->num_queue_pairs = ICE_MAX_BASE_QS_PER_VF;
} else if (req_queues - cur_queues > tx_rx_queue_left) {
dev_warn(&pf->pdev->dev,
"VF %d requested %d more queues, but only %d left.\n",
vf->vf_id, req_queues - cur_queues, tx_rx_queue_left);
- vfres->num_queue_pairs = tx_rx_queue_left + cur_queues;
+ vfres->num_queue_pairs = min_t(int, max_allowed_vf_queues,
+ ICE_MAX_BASE_QS_PER_VF);
} else {
/* request is successful, then reset VF */
vf->num_req_qs = req_queues;
@@ -2033,7 +2203,7 @@ static int ice_vc_request_qs_msg(struct ice_vf *vf, u8 *msg)
error_param:
/* send the response to the VF */
return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_REQUEST_QUEUES,
- aq_ret, (u8 *)vfres, sizeof(*vfres));
+ v_ret, (u8 *)vfres, sizeof(*vfres));
}
/**
@@ -2093,11 +2263,12 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos,
VLAN_VID_MASK));
if (vlan_id || qos) {
- ret = ice_vsi_set_pvid(vsi, vlanprio);
+ ret = ice_vsi_manage_pvid(vsi, vlanprio, true);
if (ret)
goto error_set_pvid;
} else {
- ice_vsi_kill_pvid(vsi);
+ ice_vsi_manage_pvid(vsi, 0, false);
+ vsi->info.pvid = 0;
}
if (vlan_id) {
@@ -2129,48 +2300,57 @@ error_set_pvid:
*/
static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
{
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct virtchnl_vlan_filter_list *vfl =
(struct virtchnl_vlan_filter_list *)msg;
- enum ice_status aq_ret = 0;
struct ice_pf *pf = vf->pf;
+ bool vlan_promisc = false;
struct ice_vsi *vsi;
+ struct ice_hw *hw;
+ int status = 0;
+ u8 promisc_m;
int i;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (!ice_vc_isvalid_vsi_id(vf, vfl->vsi_id)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (add_v && !ice_is_vf_trusted(vf) &&
vf->num_vlan >= ICE_MAX_VLAN_PER_VF) {
dev_info(&pf->pdev->dev,
- "VF is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n");
- aq_ret = ICE_ERR_PARAM;
+ "VF-%d is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n",
+ vf->vf_id);
+ /* There is no need to let VF know about being not trusted,
+ * so we can just return success message here
+ */
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
for (i = 0; i < vfl->num_elements; i++) {
if (vfl->vlan_id[i] > ICE_MAX_VLANID) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
dev_err(&pf->pdev->dev,
"invalid VF VLAN id %d\n", vfl->vlan_id[i]);
goto error_param;
}
}
- vsi = ice_find_vsi_from_id(vf->pf, vfl->vsi_id);
+ hw = &pf->hw;
+ vsi = pf->vsi[vf->lan_vsi_idx];
if (!vsi) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
if (vsi->info.pvid) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
@@ -2178,23 +2358,47 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
dev_err(&pf->pdev->dev,
"%sable VLAN stripping failed for VSI %i\n",
add_v ? "en" : "dis", vsi->vsi_num);
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
+ if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) ||
+ test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states))
+ vlan_promisc = true;
+
if (add_v) {
for (i = 0; i < vfl->num_elements; i++) {
u16 vid = vfl->vlan_id[i];
- if (!ice_vsi_add_vlan(vsi, vid)) {
- vf->num_vlan++;
+ if (ice_vsi_add_vlan(vsi, vid)) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ goto error_param;
+ }
- /* Enable VLAN pruning when VLAN 0 is added */
- if (unlikely(!vid))
- if (ice_cfg_vlan_pruning(vsi, true))
- aq_ret = ICE_ERR_PARAM;
+ vf->num_vlan++;
+ /* Enable VLAN pruning when VLAN is added */
+ if (!vlan_promisc) {
+ status = ice_cfg_vlan_pruning(vsi, true, false);
+ if (status) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ dev_err(&pf->pdev->dev,
+ "Enable VLAN pruning on VLAN ID: %d failed error-%d\n",
+ vid, status);
+ goto error_param;
+ }
} else {
- aq_ret = ICE_ERR_PARAM;
+ /* Enable Ucast/Mcast VLAN promiscuous mode */
+ promisc_m = ICE_PROMISC_VLAN_TX |
+ ICE_PROMISC_VLAN_RX;
+
+ status = ice_set_vsi_promisc(hw, vsi->idx,
+ promisc_m, vid);
+ if (status) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ dev_err(&pf->pdev->dev,
+ "Enable Unicast/multicast promiscuous mode on VLAN ID:%d failed error-%d\n",
+ vid, status);
+ }
}
}
} else {
@@ -2204,12 +2408,22 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
/* Make sure ice_vsi_kill_vlan is successful before
* updating VLAN information
*/
- if (!ice_vsi_kill_vlan(vsi, vid)) {
- vf->num_vlan--;
+ if (ice_vsi_kill_vlan(vsi, vid)) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ goto error_param;
+ }
+
+ vf->num_vlan--;
+ /* Disable VLAN pruning when removing VLAN */
+ ice_cfg_vlan_pruning(vsi, false, false);
- /* Disable VLAN pruning when removing VLAN 0 */
- if (unlikely(!vid))
- ice_cfg_vlan_pruning(vsi, false);
+ /* Disable Unicast/Multicast VLAN promiscuous mode */
+ if (vlan_promisc) {
+ promisc_m = ICE_PROMISC_VLAN_TX |
+ ICE_PROMISC_VLAN_RX;
+
+ ice_clear_vsi_promisc(hw, vsi->idx,
+ promisc_m, vid);
}
}
}
@@ -2217,10 +2431,10 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
error_param:
/* send the response to the VF */
if (add_v)
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ADD_VLAN, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ADD_VLAN, v_ret,
NULL, 0);
else
- return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DEL_VLAN, aq_ret,
+ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DEL_VLAN, v_ret,
NULL, 0);
}
@@ -2256,22 +2470,22 @@ static int ice_vc_remove_vlan_msg(struct ice_vf *vf, u8 *msg)
*/
static int ice_vc_ena_vlan_stripping(struct ice_vf *vf)
{
- enum ice_status aq_ret = 0;
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
vsi = pf->vsi[vf->lan_vsi_idx];
if (ice_vsi_manage_vlan_stripping(vsi, true))
- aq_ret = ICE_ERR_AQ_ERROR;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
error_param:
return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_VLAN_STRIPPING,
- aq_ret, NULL, 0);
+ v_ret, NULL, 0);
}
/**
@@ -2282,22 +2496,27 @@ error_param:
*/
static int ice_vc_dis_vlan_stripping(struct ice_vf *vf)
{
- enum ice_status aq_ret = 0;
+ enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS;
struct ice_pf *pf = vf->pf;
struct ice_vsi *vsi;
if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) {
- aq_ret = ICE_ERR_PARAM;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
goto error_param;
}
vsi = pf->vsi[vf->lan_vsi_idx];
+ if (!vsi) {
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+ goto error_param;
+ }
+
if (ice_vsi_manage_vlan_stripping(vsi, false))
- aq_ret = ICE_ERR_AQ_ERROR;
+ v_ret = VIRTCHNL_STATUS_ERR_PARAM;
error_param:
return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_VLAN_STRIPPING,
- aq_ret, NULL, 0);
+ v_ret, NULL, 0);
}
/**
@@ -2333,7 +2552,7 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
/* Perform basic checks on the msg */
err = virtchnl_vc_validate_vf_msg(&vf->vf_ver, v_opcode, msg, msglen);
if (err) {
- if (err == VIRTCHNL_ERR_PARAM)
+ if (err == VIRTCHNL_STATUS_ERR_PARAM)
err = -EPERM;
else
err = -EINVAL;
@@ -2355,7 +2574,8 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
error_handler:
if (err) {
- ice_vc_send_msg_to_vf(vf, v_opcode, ICE_ERR_PARAM, NULL, 0);
+ ice_vc_send_msg_to_vf(vf, v_opcode, VIRTCHNL_STATUS_ERR_PARAM,
+ NULL, 0);
dev_err(&pf->pdev->dev, "Invalid message from VF %d, opcode %d, len %d, error %d\n",
vf_id, v_opcode, msglen, err);
return;
@@ -2418,7 +2638,8 @@ error_handler:
default:
dev_err(&pf->pdev->dev, "Unsupported opcode %d from VF %d\n",
v_opcode, vf_id);
- err = ice_vc_send_msg_to_vf(vf, v_opcode, ICE_ERR_NOT_IMPL,
+ err = ice_vc_send_msg_to_vf(vf, v_opcode,
+ VIRTCHNL_STATUS_ERR_NOT_SUPPORTED,
NULL, 0);
break;
}
@@ -2427,7 +2648,7 @@ error_handler:
* as it is busy with pending work.
*/
dev_info(&pf->pdev->dev,
- "PF failed to honor VF %d, opcode %d\n, error %d\n",
+ "PF failed to honor VF %d, opcode %d, error %d\n",
vf_id, v_opcode, err);
}
}
@@ -2440,8 +2661,8 @@ error_handler:
*
* return VF configuration
*/
-int ice_get_vf_cfg(struct net_device *netdev, int vf_id,
- struct ifla_vf_info *ivi)
+int
+ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi)
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
@@ -2587,7 +2808,7 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
ether_addr_copy(vf->dflt_lan_addr.addr, mac);
vf->pf_set_mac = true;
netdev_info(netdev,
- "mac on VF %d set to %pM\n. VF driver will be reinitialized\n",
+ "MAC on VF %d set to %pM. VF driver will be reinitialized\n",
vf_id, mac);
ice_vc_dis_vf(vf);
@@ -2690,7 +2911,8 @@ int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state)
ice_set_pfe_link(vf, &pfe, ls->link_speed, vf->link_up);
/* Notify the VF of its new link state */
- ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, 0, (u8 *)&pfe,
+ ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT,
+ VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe,
sizeof(pfe), NULL);
return 0;
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
index 01470a8ee03a..932e2ab3380b 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h
@@ -70,6 +70,7 @@ struct ice_vf {
u8 spoofchk;
u16 num_mac;
u16 num_vlan;
+ u16 num_vf_qs; /* num of queue configured per VF */
u8 num_req_qs; /* num of queue pairs requested by VF */
};
@@ -77,8 +78,8 @@ struct ice_vf {
void ice_process_vflr_event(struct ice_pf *pf);
int ice_sriov_configure(struct pci_dev *pdev, int num_vfs);
int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac);
-int ice_get_vf_cfg(struct net_device *netdev, int vf_id,
- struct ifla_vf_info *ivi);
+int
+ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi);
void ice_free_vfs(struct ice_pf *pf);
void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event);
@@ -86,11 +87,9 @@ void ice_vc_notify_link_state(struct ice_pf *pf);
void ice_vc_notify_reset(struct ice_pf *pf);
bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr);
-int ice_set_vf_port_vlan(struct net_device *netdev, int vf_id,
- u16 vlan_id, u8 qos, __be16 vlan_proto);
-
-int ice_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate,
- int max_tx_rate);
+int
+ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos,
+ __be16 vlan_proto);
int ice_set_vf_trust(struct net_device *netdev, int vf_id, bool trusted);
@@ -162,12 +161,5 @@ ice_set_vf_link_state(struct net_device __always_unused *netdev,
return -EOPNOTSUPP;
}
-static inline int
-ice_set_vf_bw(struct net_device __always_unused *netdev,
- int __always_unused vf_id, int __always_unused min_tx_rate,
- int __always_unused max_tx_rate)
-{
- return -EOPNOTSUPP;
-}
#endif /* CONFIG_PCI_IOV */
#endif /* _ICE_VIRTCHNL_PF_H_ */
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index c57671068245..c645d9e648e0 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -3158,8 +3158,8 @@ static int igb_set_eee(struct net_device *netdev,
} else if (!edata->eee_enabled) {
dev_err(&adapter->pdev->dev,
"Setting EEE options are not supported with EEE disabled\n");
- return -EINVAL;
- }
+ return -EINVAL;
+ }
adapter->eee_advert = ethtool_adv_to_mmd_eee_adv_t(edata->advertised);
if (hw->dev_spec._82575.eee_disable != !edata->eee_enabled) {
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 69b230c53fed..bea7175d171b 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2480,7 +2480,7 @@ static int igb_set_features(struct net_device *netdev,
else
igb_reset(adapter);
- return 0;
+ return 1;
}
static int igb_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
@@ -3452,6 +3452,9 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
break;
}
}
+
+ dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP);
+
pm_runtime_put_noidle(&pdev->dev);
return 0;
diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 80faccc34cda..0f5534ce27b0 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -29,9 +29,15 @@ unsigned int igc_get_max_rss_queues(struct igc_adapter *adapter);
void igc_set_flag_queue_pairs(struct igc_adapter *adapter,
const u32 max_rss_queues);
int igc_reinit_queues(struct igc_adapter *adapter);
+void igc_write_rss_indir_tbl(struct igc_adapter *adapter);
bool igc_has_link(struct igc_adapter *adapter);
void igc_reset(struct igc_adapter *adapter);
int igc_set_spd_dplx(struct igc_adapter *adapter, u32 spd, u8 dplx);
+int igc_add_mac_steering_filter(struct igc_adapter *adapter,
+ const u8 *addr, u8 queue, u8 flags);
+int igc_del_mac_steering_filter(struct igc_adapter *adapter,
+ const u8 *addr, u8 queue, u8 flags);
+void igc_update_stats(struct igc_adapter *adapter);
extern char igc_driver_name[];
extern char igc_driver_version[];
@@ -51,6 +57,13 @@ extern char igc_driver_version[];
#define IGC_FLAG_VLAN_PROMISC BIT(15)
#define IGC_FLAG_RX_LEGACY BIT(16)
+#define IGC_FLAG_RSS_FIELD_IPV4_UDP BIT(6)
+#define IGC_FLAG_RSS_FIELD_IPV6_UDP BIT(7)
+
+#define IGC_MRQC_ENABLE_RSS_MQ 0x00000002
+#define IGC_MRQC_RSS_FIELD_IPV4_UDP 0x00400000
+#define IGC_MRQC_RSS_FIELD_IPV6_UDP 0x00800000
+
#define IGC_START_ITR 648 /* ~6000 ints/sec */
#define IGC_4K_ITR 980
#define IGC_20K_ITR 196
@@ -284,15 +297,50 @@ struct igc_q_vector {
struct igc_ring ring[0] ____cacheline_internodealigned_in_smp;
};
+#define MAX_ETYPE_FILTER (4 - 1)
+
+enum igc_filter_match_flags {
+ IGC_FILTER_FLAG_ETHER_TYPE = 0x1,
+ IGC_FILTER_FLAG_VLAN_TCI = 0x2,
+ IGC_FILTER_FLAG_SRC_MAC_ADDR = 0x4,
+ IGC_FILTER_FLAG_DST_MAC_ADDR = 0x8,
+};
+
+/* RX network flow classification data structure */
+struct igc_nfc_input {
+ /* Byte layout in order, all values with MSB first:
+ * match_flags - 1 byte
+ * etype - 2 bytes
+ * vlan_tci - 2 bytes
+ */
+ u8 match_flags;
+ __be16 etype;
+ __be16 vlan_tci;
+ u8 src_addr[ETH_ALEN];
+ u8 dst_addr[ETH_ALEN];
+};
+
+struct igc_nfc_filter {
+ struct hlist_node nfc_node;
+ struct igc_nfc_input filter;
+ unsigned long cookie;
+ u16 etype_reg_index;
+ u16 sw_idx;
+ u16 action;
+};
+
struct igc_mac_addr {
u8 addr[ETH_ALEN];
u8 queue;
u8 state; /* bitmask */
};
-#define IGC_MAC_STATE_DEFAULT 0x1
-#define IGC_MAC_STATE_MODIFIED 0x2
-#define IGC_MAC_STATE_IN_USE 0x4
+#define IGC_MAC_STATE_DEFAULT 0x1
+#define IGC_MAC_STATE_IN_USE 0x2
+#define IGC_MAC_STATE_SRC_ADDR 0x4
+#define IGC_MAC_STATE_QUEUE_STEERING 0x8
+
+#define IGC_MAX_RXNFC_FILTERS 16
/* Board specific private data structure */
struct igc_adapter {
@@ -356,12 +404,22 @@ struct igc_adapter {
u16 tx_ring_count;
u16 rx_ring_count;
+ u32 tx_hwtstamp_timeouts;
+ u32 tx_hwtstamp_skipped;
+ u32 rx_hwtstamp_cleared;
u32 *shadow_vfta;
u32 rss_queues;
+ u32 rss_indir_tbl_init;
+
+ /* RX network flow classification support */
+ struct hlist_head nfc_filter_list;
+ struct hlist_head cls_flower_list;
+ unsigned int nfc_filter_count;
/* lock for RX network flow classification filter */
spinlock_t nfc_lock;
+ bool etype_bitmap[MAX_ETYPE_FILTER];
struct igc_mac_addr *mac_table;
@@ -447,6 +505,10 @@ static inline s32 igc_read_phy_reg(struct igc_hw *hw, u32 offset, u16 *data)
/* forward declaration */
void igc_reinit_locked(struct igc_adapter *);
+int igc_add_filter(struct igc_adapter *adapter,
+ struct igc_nfc_filter *input);
+int igc_erase_filter(struct igc_adapter *adapter,
+ struct igc_nfc_filter *input);
#define igc_rx_pg_size(_ring) (PAGE_SIZE << igc_rx_pg_order(_ring))
diff --git a/drivers/net/ethernet/intel/igc/igc_base.h b/drivers/net/ethernet/intel/igc/igc_base.h
index 76d4991d7284..58d1109d7f3f 100644
--- a/drivers/net/ethernet/intel/igc/igc_base.h
+++ b/drivers/net/ethernet/intel/igc/igc_base.h
@@ -1,8 +1,8 @@
/* SPDX-License-Identifier: GPL-2.0 */
/* Copyright (c) 2018 Intel Corporation */
-#ifndef _IGC_BASE_H
-#define _IGC_BASE_H
+#ifndef _IGC_BASE_H_
+#define _IGC_BASE_H_
/* forward declaration */
void igc_rx_fifo_flush_base(struct igc_hw *hw);
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 7d1bdcd1225a..a9a30268de59 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -310,6 +310,12 @@
IGC_RXDEXT_STATERR_CXE | \
IGC_RXDEXT_STATERR_RXE)
+#define IGC_MRQC_RSS_FIELD_IPV4_TCP 0x00010000
+#define IGC_MRQC_RSS_FIELD_IPV4 0x00020000
+#define IGC_MRQC_RSS_FIELD_IPV6_TCP_EX 0x00040000
+#define IGC_MRQC_RSS_FIELD_IPV6 0x00100000
+#define IGC_MRQC_RSS_FIELD_IPV6_TCP 0x00200000
+
/* Header split receive */
#define IGC_RFCTL_IPV6_EX_DIS 0x00010000
#define IGC_RFCTL_LEF 0x00040000
@@ -325,6 +331,10 @@
#define I225_RXPBSIZE_DEFAULT 0x000000A2 /* RXPBSIZE default */
#define I225_TXPBSIZE_DEFAULT 0x04000014 /* TXPBSIZE default */
+/* Receive Checksum Control */
+#define IGC_RXCSUM_CRCOFL 0x00000800 /* CRC32 offload enable */
+#define IGC_RXCSUM_PCSD 0x00002000 /* packet checksum disabled */
+
/* GPY211 - I225 defines */
#define GPY_MMD_MASK 0xFFFF0000
#define GPY_MMD_SHIFT 16
@@ -390,4 +400,11 @@
#define IGC_N0_QUEUE -1
+#define IGC_MAX_MAC_HDR_LEN 127
+#define IGC_MAX_NETWORK_HDR_LEN 511
+
+#define IGC_VLAPQF_QUEUE_SEL(_n, q_idx) ((q_idx) << ((_n) * 4))
+#define IGC_VLAPQF_P_VALID(_n) (0x1 << (3 + (_n) * 4))
+#define IGC_VLAPQF_QUEUE_MASK 0x03
+
#endif /* _IGC_DEFINES_H_ */
diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index eff37a6c0afa..ac98f1d96892 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -2,10 +2,120 @@
/* Copyright (c) 2018 Intel Corporation */
/* ethtool support for igc */
+#include <linux/if_vlan.h>
#include <linux/pm_runtime.h>
#include "igc.h"
+/* forward declaration */
+struct igc_stats {
+ char stat_string[ETH_GSTRING_LEN];
+ int sizeof_stat;
+ int stat_offset;
+};
+
+#define IGC_STAT(_name, _stat) { \
+ .stat_string = _name, \
+ .sizeof_stat = FIELD_SIZEOF(struct igc_adapter, _stat), \
+ .stat_offset = offsetof(struct igc_adapter, _stat) \
+}
+
+static const struct igc_stats igc_gstrings_stats[] = {
+ IGC_STAT("rx_packets", stats.gprc),
+ IGC_STAT("tx_packets", stats.gptc),
+ IGC_STAT("rx_bytes", stats.gorc),
+ IGC_STAT("tx_bytes", stats.gotc),
+ IGC_STAT("rx_broadcast", stats.bprc),
+ IGC_STAT("tx_broadcast", stats.bptc),
+ IGC_STAT("rx_multicast", stats.mprc),
+ IGC_STAT("tx_multicast", stats.mptc),
+ IGC_STAT("multicast", stats.mprc),
+ IGC_STAT("collisions", stats.colc),
+ IGC_STAT("rx_crc_errors", stats.crcerrs),
+ IGC_STAT("rx_no_buffer_count", stats.rnbc),
+ IGC_STAT("rx_missed_errors", stats.mpc),
+ IGC_STAT("tx_aborted_errors", stats.ecol),
+ IGC_STAT("tx_carrier_errors", stats.tncrs),
+ IGC_STAT("tx_window_errors", stats.latecol),
+ IGC_STAT("tx_abort_late_coll", stats.latecol),
+ IGC_STAT("tx_deferred_ok", stats.dc),
+ IGC_STAT("tx_single_coll_ok", stats.scc),
+ IGC_STAT("tx_multi_coll_ok", stats.mcc),
+ IGC_STAT("tx_timeout_count", tx_timeout_count),
+ IGC_STAT("rx_long_length_errors", stats.roc),
+ IGC_STAT("rx_short_length_errors", stats.ruc),
+ IGC_STAT("rx_align_errors", stats.algnerrc),
+ IGC_STAT("tx_tcp_seg_good", stats.tsctc),
+ IGC_STAT("tx_tcp_seg_failed", stats.tsctfc),
+ IGC_STAT("rx_flow_control_xon", stats.xonrxc),
+ IGC_STAT("rx_flow_control_xoff", stats.xoffrxc),
+ IGC_STAT("tx_flow_control_xon", stats.xontxc),
+ IGC_STAT("tx_flow_control_xoff", stats.xofftxc),
+ IGC_STAT("rx_long_byte_count", stats.gorc),
+ IGC_STAT("tx_dma_out_of_sync", stats.doosync),
+ IGC_STAT("tx_smbus", stats.mgptc),
+ IGC_STAT("rx_smbus", stats.mgprc),
+ IGC_STAT("dropped_smbus", stats.mgpdc),
+ IGC_STAT("os2bmc_rx_by_bmc", stats.o2bgptc),
+ IGC_STAT("os2bmc_tx_by_bmc", stats.b2ospc),
+ IGC_STAT("os2bmc_tx_by_host", stats.o2bspc),
+ IGC_STAT("os2bmc_rx_by_host", stats.b2ogprc),
+ IGC_STAT("tx_hwtstamp_timeouts", tx_hwtstamp_timeouts),
+ IGC_STAT("tx_hwtstamp_skipped", tx_hwtstamp_skipped),
+ IGC_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared),
+};
+
+#define IGC_NETDEV_STAT(_net_stat) { \
+ .stat_string = __stringify(_net_stat), \
+ .sizeof_stat = FIELD_SIZEOF(struct rtnl_link_stats64, _net_stat), \
+ .stat_offset = offsetof(struct rtnl_link_stats64, _net_stat) \
+}
+
+static const struct igc_stats igc_gstrings_net_stats[] = {
+ IGC_NETDEV_STAT(rx_errors),
+ IGC_NETDEV_STAT(tx_errors),
+ IGC_NETDEV_STAT(tx_dropped),
+ IGC_NETDEV_STAT(rx_length_errors),
+ IGC_NETDEV_STAT(rx_over_errors),
+ IGC_NETDEV_STAT(rx_frame_errors),
+ IGC_NETDEV_STAT(rx_fifo_errors),
+ IGC_NETDEV_STAT(tx_fifo_errors),
+ IGC_NETDEV_STAT(tx_heartbeat_errors)
+};
+
+enum igc_diagnostics_results {
+ TEST_REG = 0,
+ TEST_EEP,
+ TEST_IRQ,
+ TEST_LOOP,
+ TEST_LINK
+};
+
+static const char igc_gstrings_test[][ETH_GSTRING_LEN] = {
+ [TEST_REG] = "Register test (offline)",
+ [TEST_EEP] = "Eeprom test (offline)",
+ [TEST_IRQ] = "Interrupt test (offline)",
+ [TEST_LOOP] = "Loopback test (offline)",
+ [TEST_LINK] = "Link test (on/offline)"
+};
+
+#define IGC_TEST_LEN (sizeof(igc_gstrings_test) / ETH_GSTRING_LEN)
+
+#define IGC_GLOBAL_STATS_LEN \
+ (sizeof(igc_gstrings_stats) / sizeof(struct igc_stats))
+#define IGC_NETDEV_STATS_LEN \
+ (sizeof(igc_gstrings_net_stats) / sizeof(struct igc_stats))
+#define IGC_RX_QUEUE_STATS_LEN \
+ (sizeof(struct igc_rx_queue_stats) / sizeof(u64))
+#define IGC_TX_QUEUE_STATS_LEN 3 /* packets, bytes, restart_queue */
+#define IGC_QUEUE_STATS_LEN \
+ ((((struct igc_adapter *)netdev_priv(netdev))->num_rx_queues * \
+ IGC_RX_QUEUE_STATS_LEN) + \
+ (((struct igc_adapter *)netdev_priv(netdev))->num_tx_queues * \
+ IGC_TX_QUEUE_STATS_LEN))
+#define IGC_STATS_LEN \
+ (IGC_GLOBAL_STATS_LEN + IGC_NETDEV_STATS_LEN + IGC_QUEUE_STATS_LEN)
+
static const char igc_priv_flags_strings[][ETH_GSTRING_LEN] = {
#define IGC_PRIV_FLAGS_LEGACY_RX BIT(0)
"legacy-rx",
@@ -545,6 +655,127 @@ static int igc_set_pauseparam(struct net_device *netdev,
return retval;
}
+static void igc_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+{
+ struct igc_adapter *adapter = netdev_priv(netdev);
+ u8 *p = data;
+ int i;
+
+ switch (stringset) {
+ case ETH_SS_TEST:
+ memcpy(data, *igc_gstrings_test,
+ IGC_TEST_LEN * ETH_GSTRING_LEN);
+ break;
+ case ETH_SS_STATS:
+ for (i = 0; i < IGC_GLOBAL_STATS_LEN; i++) {
+ memcpy(p, igc_gstrings_stats[i].stat_string,
+ ETH_GSTRING_LEN);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < IGC_NETDEV_STATS_LEN; i++) {
+ memcpy(p, igc_gstrings_net_stats[i].stat_string,
+ ETH_GSTRING_LEN);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < adapter->num_tx_queues; i++) {
+ sprintf(p, "tx_queue_%u_packets", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "tx_queue_%u_bytes", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "tx_queue_%u_restart", i);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < adapter->num_rx_queues; i++) {
+ sprintf(p, "rx_queue_%u_packets", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "rx_queue_%u_bytes", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "rx_queue_%u_drops", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "rx_queue_%u_csum_err", i);
+ p += ETH_GSTRING_LEN;
+ sprintf(p, "rx_queue_%u_alloc_failed", i);
+ p += ETH_GSTRING_LEN;
+ }
+ /* BUG_ON(p - data != IGC_STATS_LEN * ETH_GSTRING_LEN); */
+ break;
+ case ETH_SS_PRIV_FLAGS:
+ memcpy(data, igc_priv_flags_strings,
+ IGC_PRIV_FLAGS_STR_LEN * ETH_GSTRING_LEN);
+ break;
+ }
+}
+
+static int igc_get_sset_count(struct net_device *netdev, int sset)
+{
+ switch (sset) {
+ case ETH_SS_STATS:
+ return IGC_STATS_LEN;
+ case ETH_SS_TEST:
+ return IGC_TEST_LEN;
+ case ETH_SS_PRIV_FLAGS:
+ return IGC_PRIV_FLAGS_STR_LEN;
+ default:
+ return -ENOTSUPP;
+ }
+}
+
+static void igc_get_ethtool_stats(struct net_device *netdev,
+ struct ethtool_stats *stats, u64 *data)
+{
+ struct igc_adapter *adapter = netdev_priv(netdev);
+ struct rtnl_link_stats64 *net_stats = &adapter->stats64;
+ unsigned int start;
+ struct igc_ring *ring;
+ int i, j;
+ char *p;
+
+ spin_lock(&adapter->stats64_lock);
+ igc_update_stats(adapter);
+
+ for (i = 0; i < IGC_GLOBAL_STATS_LEN; i++) {
+ p = (char *)adapter + igc_gstrings_stats[i].stat_offset;
+ data[i] = (igc_gstrings_stats[i].sizeof_stat ==
+ sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+ }
+ for (j = 0; j < IGC_NETDEV_STATS_LEN; j++, i++) {
+ p = (char *)net_stats + igc_gstrings_net_stats[j].stat_offset;
+ data[i] = (igc_gstrings_net_stats[j].sizeof_stat ==
+ sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+ }
+ for (j = 0; j < adapter->num_tx_queues; j++) {
+ u64 restart2;
+
+ ring = adapter->tx_ring[j];
+ do {
+ start = u64_stats_fetch_begin_irq(&ring->tx_syncp);
+ data[i] = ring->tx_stats.packets;
+ data[i + 1] = ring->tx_stats.bytes;
+ data[i + 2] = ring->tx_stats.restart_queue;
+ } while (u64_stats_fetch_retry_irq(&ring->tx_syncp, start));
+ do {
+ start = u64_stats_fetch_begin_irq(&ring->tx_syncp2);
+ restart2 = ring->tx_stats.restart_queue2;
+ } while (u64_stats_fetch_retry_irq(&ring->tx_syncp2, start));
+ data[i + 2] += restart2;
+
+ i += IGC_TX_QUEUE_STATS_LEN;
+ }
+ for (j = 0; j < adapter->num_rx_queues; j++) {
+ ring = adapter->rx_ring[j];
+ do {
+ start = u64_stats_fetch_begin_irq(&ring->rx_syncp);
+ data[i] = ring->rx_stats.packets;
+ data[i + 1] = ring->rx_stats.bytes;
+ data[i + 2] = ring->rx_stats.drops;
+ data[i + 3] = ring->rx_stats.csum_err;
+ data[i + 4] = ring->rx_stats.alloc_failed;
+ } while (u64_stats_fetch_retry_irq(&ring->rx_syncp, start));
+ i += IGC_RX_QUEUE_STATS_LEN;
+ }
+ spin_unlock(&adapter->stats64_lock);
+}
+
static int igc_get_coalesce(struct net_device *netdev,
struct ethtool_coalesce *ec)
{
@@ -643,6 +874,605 @@ static int igc_set_coalesce(struct net_device *netdev,
return 0;
}
+#define ETHER_TYPE_FULL_MASK ((__force __be16)~0)
+static int igc_get_ethtool_nfc_entry(struct igc_adapter *adapter,
+ struct ethtool_rxnfc *cmd)
+{
+ struct ethtool_rx_flow_spec *fsp = &cmd->fs;
+ struct igc_nfc_filter *rule = NULL;
+
+ /* report total rule count */
+ cmd->data = IGC_MAX_RXNFC_FILTERS;
+
+ hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) {
+ if (fsp->location <= rule->sw_idx)
+ break;
+ }
+
+ if (!rule || fsp->location != rule->sw_idx)
+ return -EINVAL;
+
+ if (rule->filter.match_flags) {
+ fsp->flow_type = ETHER_FLOW;
+ fsp->ring_cookie = rule->action;
+ if (rule->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) {
+ fsp->h_u.ether_spec.h_proto = rule->filter.etype;
+ fsp->m_u.ether_spec.h_proto = ETHER_TYPE_FULL_MASK;
+ }
+ if (rule->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) {
+ fsp->flow_type |= FLOW_EXT;
+ fsp->h_ext.vlan_tci = rule->filter.vlan_tci;
+ fsp->m_ext.vlan_tci = htons(VLAN_PRIO_MASK);
+ }
+ if (rule->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) {
+ ether_addr_copy(fsp->h_u.ether_spec.h_dest,
+ rule->filter.dst_addr);
+ /* As we only support matching by the full
+ * mask, return the mask to userspace
+ */
+ eth_broadcast_addr(fsp->m_u.ether_spec.h_dest);
+ }
+ if (rule->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) {
+ ether_addr_copy(fsp->h_u.ether_spec.h_source,
+ rule->filter.src_addr);
+ /* As we only support matching by the full
+ * mask, return the mask to userspace
+ */
+ eth_broadcast_addr(fsp->m_u.ether_spec.h_source);
+ }
+
+ return 0;
+ }
+ return -EINVAL;
+}
+
+static int igc_get_ethtool_nfc_all(struct igc_adapter *adapter,
+ struct ethtool_rxnfc *cmd,
+ u32 *rule_locs)
+{
+ struct igc_nfc_filter *rule;
+ int cnt = 0;
+
+ /* report total rule count */
+ cmd->data = IGC_MAX_RXNFC_FILTERS;
+
+ hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) {
+ if (cnt == cmd->rule_cnt)
+ return -EMSGSIZE;
+ rule_locs[cnt] = rule->sw_idx;
+ cnt++;
+ }
+
+ cmd->rule_cnt = cnt;
+
+ return 0;
+}
+
+static int igc_get_rss_hash_opts(struct igc_adapter *adapter,
+ struct ethtool_rxnfc *cmd)
+{
+ cmd->data = 0;
+
+ /* Report default options for RSS on igc */
+ switch (cmd->flow_type) {
+ case TCP_V4_FLOW:
+ cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ /* Fall through */
+ case UDP_V4_FLOW:
+ if (adapter->flags & IGC_FLAG_RSS_FIELD_IPV4_UDP)
+ cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ /* Fall through */
+ case SCTP_V4_FLOW:
+ /* Fall through */
+ case AH_ESP_V4_FLOW:
+ /* Fall through */
+ case AH_V4_FLOW:
+ /* Fall through */
+ case ESP_V4_FLOW:
+ /* Fall through */
+ case IPV4_FLOW:
+ cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+ break;
+ case TCP_V6_FLOW:
+ cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ /* Fall through */
+ case UDP_V6_FLOW:
+ if (adapter->flags & IGC_FLAG_RSS_FIELD_IPV6_UDP)
+ cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ /* Fall through */
+ case SCTP_V6_FLOW:
+ /* Fall through */
+ case AH_ESP_V6_FLOW:
+ /* Fall through */
+ case AH_V6_FLOW:
+ /* Fall through */
+ case ESP_V6_FLOW:
+ /* Fall through */
+ case IPV6_FLOW:
+ cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int igc_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
+ u32 *rule_locs)
+{
+ struct igc_adapter *adapter = netdev_priv(dev);
+ int ret = -EOPNOTSUPP;
+
+ switch (cmd->cmd) {
+ case ETHTOOL_GRXRINGS:
+ cmd->data = adapter->num_rx_queues;
+ ret = 0;
+ break;
+ case ETHTOOL_GRXCLSRLCNT:
+ cmd->rule_cnt = adapter->nfc_filter_count;
+ ret = 0;
+ break;
+ case ETHTOOL_GRXCLSRULE:
+ ret = igc_get_ethtool_nfc_entry(adapter, cmd);
+ break;
+ case ETHTOOL_GRXCLSRLALL:
+ ret = igc_get_ethtool_nfc_all(adapter, cmd, rule_locs);
+ break;
+ case ETHTOOL_GRXFH:
+ ret = igc_get_rss_hash_opts(adapter, cmd);
+ break;
+ default:
+ break;
+ }
+
+ return ret;
+}
+
+#define UDP_RSS_FLAGS (IGC_FLAG_RSS_FIELD_IPV4_UDP | \
+ IGC_FLAG_RSS_FIELD_IPV6_UDP)
+static int igc_set_rss_hash_opt(struct igc_adapter *adapter,
+ struct ethtool_rxnfc *nfc)
+{
+ u32 flags = adapter->flags;
+
+ /* RSS does not support anything other than hashing
+ * to queues on src and dst IPs and ports
+ */
+ if (nfc->data & ~(RXH_IP_SRC | RXH_IP_DST |
+ RXH_L4_B_0_1 | RXH_L4_B_2_3))
+ return -EINVAL;
+
+ switch (nfc->flow_type) {
+ case TCP_V4_FLOW:
+ case TCP_V6_FLOW:
+ if (!(nfc->data & RXH_IP_SRC) ||
+ !(nfc->data & RXH_IP_DST) ||
+ !(nfc->data & RXH_L4_B_0_1) ||
+ !(nfc->data & RXH_L4_B_2_3))
+ return -EINVAL;
+ break;
+ case UDP_V4_FLOW:
+ if (!(nfc->data & RXH_IP_SRC) ||
+ !(nfc->data & RXH_IP_DST))
+ return -EINVAL;
+ switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+ case 0:
+ flags &= ~IGC_FLAG_RSS_FIELD_IPV4_UDP;
+ break;
+ case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+ flags |= IGC_FLAG_RSS_FIELD_IPV4_UDP;
+ break;
+ default:
+ return -EINVAL;
+ }
+ break;
+ case UDP_V6_FLOW:
+ if (!(nfc->data & RXH_IP_SRC) ||
+ !(nfc->data & RXH_IP_DST))
+ return -EINVAL;
+ switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+ case 0:
+ flags &= ~IGC_FLAG_RSS_FIELD_IPV6_UDP;
+ break;
+ case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+ flags |= IGC_FLAG_RSS_FIELD_IPV6_UDP;
+ break;
+ default:
+ return -EINVAL;
+ }
+ break;
+ case AH_ESP_V4_FLOW:
+ case AH_V4_FLOW:
+ case ESP_V4_FLOW:
+ case SCTP_V4_FLOW:
+ case AH_ESP_V6_FLOW:
+ case AH_V6_FLOW:
+ case ESP_V6_FLOW:
+ case SCTP_V6_FLOW:
+ if (!(nfc->data & RXH_IP_SRC) ||
+ !(nfc->data & RXH_IP_DST) ||
+ (nfc->data & RXH_L4_B_0_1) ||
+ (nfc->data & RXH_L4_B_2_3))
+ return -EINVAL;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ /* if we changed something we need to update flags */
+ if (flags != adapter->flags) {
+ struct igc_hw *hw = &adapter->hw;
+ u32 mrqc = rd32(IGC_MRQC);
+
+ if ((flags & UDP_RSS_FLAGS) &&
+ !(adapter->flags & UDP_RSS_FLAGS))
+ dev_err(&adapter->pdev->dev,
+ "enabling UDP RSS: fragmented packets may arrive out of order to the stack above\n");
+
+ adapter->flags = flags;
+
+ /* Perform hash on these packet types */
+ mrqc |= IGC_MRQC_RSS_FIELD_IPV4 |
+ IGC_MRQC_RSS_FIELD_IPV4_TCP |
+ IGC_MRQC_RSS_FIELD_IPV6 |
+ IGC_MRQC_RSS_FIELD_IPV6_TCP;
+
+ mrqc &= ~(IGC_MRQC_RSS_FIELD_IPV4_UDP |
+ IGC_MRQC_RSS_FIELD_IPV6_UDP);
+
+ if (flags & IGC_FLAG_RSS_FIELD_IPV4_UDP)
+ mrqc |= IGC_MRQC_RSS_FIELD_IPV4_UDP;
+
+ if (flags & IGC_FLAG_RSS_FIELD_IPV6_UDP)
+ mrqc |= IGC_MRQC_RSS_FIELD_IPV6_UDP;
+
+ wr32(IGC_MRQC, mrqc);
+ }
+
+ return 0;
+}
+
+static int igc_rxnfc_write_etype_filter(struct igc_adapter *adapter,
+ struct igc_nfc_filter *input)
+{
+ struct igc_hw *hw = &adapter->hw;
+ u8 i;
+ u32 etqf;
+ u16 etype;
+
+ /* find an empty etype filter register */
+ for (i = 0; i < MAX_ETYPE_FILTER; ++i) {
+ if (!adapter->etype_bitmap[i])
+ break;
+ }
+ if (i == MAX_ETYPE_FILTER) {
+ dev_err(&adapter->pdev->dev, "ethtool -N: etype filters are all used.\n");
+ return -EINVAL;
+ }
+
+ adapter->etype_bitmap[i] = true;
+
+ etqf = rd32(IGC_ETQF(i));
+ etype = ntohs(input->filter.etype & ETHER_TYPE_FULL_MASK);
+
+ etqf |= IGC_ETQF_FILTER_ENABLE;
+ etqf &= ~IGC_ETQF_ETYPE_MASK;
+ etqf |= (etype & IGC_ETQF_ETYPE_MASK);
+
+ etqf &= ~IGC_ETQF_QUEUE_MASK;
+ etqf |= ((input->action << IGC_ETQF_QUEUE_SHIFT)
+ & IGC_ETQF_QUEUE_MASK);
+ etqf |= IGC_ETQF_QUEUE_ENABLE;
+
+ wr32(IGC_ETQF(i), etqf);
+
+ input->etype_reg_index = i;
+
+ return 0;
+}
+
+static int igc_rxnfc_write_vlan_prio_filter(struct igc_adapter *adapter,
+ struct igc_nfc_filter *input)
+{
+ struct igc_hw *hw = &adapter->hw;
+ u8 vlan_priority;
+ u16 queue_index;
+ u32 vlapqf;
+
+ vlapqf = rd32(IGC_VLAPQF);
+ vlan_priority = (ntohs(input->filter.vlan_tci) & VLAN_PRIO_MASK)
+ >> VLAN_PRIO_SHIFT;
+ queue_index = (vlapqf >> (vlan_priority * 4)) & IGC_VLAPQF_QUEUE_MASK;
+
+ /* check whether this vlan prio is already set */
+ if (vlapqf & IGC_VLAPQF_P_VALID(vlan_priority) &&
+ queue_index != input->action) {
+ dev_err(&adapter->pdev->dev, "ethtool rxnfc set vlan prio filter failed.\n");
+ return -EEXIST;
+ }
+
+ vlapqf |= IGC_VLAPQF_P_VALID(vlan_priority);
+ vlapqf |= IGC_VLAPQF_QUEUE_SEL(vlan_priority, input->action);
+
+ wr32(IGC_VLAPQF, vlapqf);
+
+ return 0;
+}
+
+int igc_add_filter(struct igc_adapter *adapter, struct igc_nfc_filter *input)
+{
+ struct igc_hw *hw = &adapter->hw;
+ int err = -EINVAL;
+
+ if (hw->mac.type == igc_i225 &&
+ !(input->filter.match_flags & ~IGC_FILTER_FLAG_SRC_MAC_ADDR)) {
+ dev_err(&adapter->pdev->dev,
+ "i225 doesn't support flow classification rules specifying only source addresses.\n");
+ return -EOPNOTSUPP;
+ }
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) {
+ err = igc_rxnfc_write_etype_filter(adapter, input);
+ if (err)
+ return err;
+ }
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) {
+ err = igc_add_mac_steering_filter(adapter,
+ input->filter.dst_addr,
+ input->action, 0);
+ err = min_t(int, err, 0);
+ if (err)
+ return err;
+ }
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) {
+ err = igc_add_mac_steering_filter(adapter,
+ input->filter.src_addr,
+ input->action,
+ IGC_MAC_STATE_SRC_ADDR);
+ err = min_t(int, err, 0);
+ if (err)
+ return err;
+ }
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI)
+ err = igc_rxnfc_write_vlan_prio_filter(adapter, input);
+
+ return err;
+}
+
+static void igc_clear_etype_filter_regs(struct igc_adapter *adapter,
+ u16 reg_index)
+{
+ struct igc_hw *hw = &adapter->hw;
+ u32 etqf = rd32(IGC_ETQF(reg_index));
+
+ etqf &= ~IGC_ETQF_QUEUE_ENABLE;
+ etqf &= ~IGC_ETQF_QUEUE_MASK;
+ etqf &= ~IGC_ETQF_FILTER_ENABLE;
+
+ wr32(IGC_ETQF(reg_index), etqf);
+
+ adapter->etype_bitmap[reg_index] = false;
+}
+
+static void igc_clear_vlan_prio_filter(struct igc_adapter *adapter,
+ u16 vlan_tci)
+{
+ struct igc_hw *hw = &adapter->hw;
+ u8 vlan_priority;
+ u32 vlapqf;
+
+ vlan_priority = (vlan_tci & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
+
+ vlapqf = rd32(IGC_VLAPQF);
+ vlapqf &= ~IGC_VLAPQF_P_VALID(vlan_priority);
+ vlapqf &= ~IGC_VLAPQF_QUEUE_SEL(vlan_priority,
+ IGC_VLAPQF_QUEUE_MASK);
+
+ wr32(IGC_VLAPQF, vlapqf);
+}
+
+int igc_erase_filter(struct igc_adapter *adapter, struct igc_nfc_filter *input)
+{
+ if (input->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE)
+ igc_clear_etype_filter_regs(adapter,
+ input->etype_reg_index);
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI)
+ igc_clear_vlan_prio_filter(adapter,
+ ntohs(input->filter.vlan_tci));
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR)
+ igc_del_mac_steering_filter(adapter, input->filter.src_addr,
+ input->action,
+ IGC_MAC_STATE_SRC_ADDR);
+
+ if (input->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR)
+ igc_del_mac_steering_filter(adapter, input->filter.dst_addr,
+ input->action, 0);
+
+ return 0;
+}
+
+static int igc_update_ethtool_nfc_entry(struct igc_adapter *adapter,
+ struct igc_nfc_filter *input,
+ u16 sw_idx)
+{
+ struct igc_nfc_filter *rule, *parent;
+ int err = -EINVAL;
+
+ parent = NULL;
+ rule = NULL;
+
+ hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) {
+ /* hash found, or no matching entry */
+ if (rule->sw_idx >= sw_idx)
+ break;
+ parent = rule;
+ }
+
+ /* if there is an old rule occupying our place remove it */
+ if (rule && rule->sw_idx == sw_idx) {
+ if (!input)
+ err = igc_erase_filter(adapter, rule);
+
+ hlist_del(&rule->nfc_node);
+ kfree(rule);
+ adapter->nfc_filter_count--;
+ }
+
+ /* If no input this was a delete, err should be 0 if a rule was
+ * successfully found and removed from the list else -EINVAL
+ */
+ if (!input)
+ return err;
+
+ /* initialize node */
+ INIT_HLIST_NODE(&input->nfc_node);
+
+ /* add filter to the list */
+ if (parent)
+ hlist_add_behind(&input->nfc_node, &parent->nfc_node);
+ else
+ hlist_add_head(&input->nfc_node, &adapter->nfc_filter_list);
+
+ /* update counts */
+ adapter->nfc_filter_count++;
+
+ return 0;
+}
+
+static int igc_add_ethtool_nfc_entry(struct igc_adapter *adapter,
+ struct ethtool_rxnfc *cmd)
+{
+ struct net_device *netdev = adapter->netdev;
+ struct ethtool_rx_flow_spec *fsp =
+ (struct ethtool_rx_flow_spec *)&cmd->fs;
+ struct igc_nfc_filter *input, *rule;
+ int err = 0;
+
+ if (!(netdev->hw_features & NETIF_F_NTUPLE))
+ return -EOPNOTSUPP;
+
+ /* Don't allow programming if the action is a queue greater than
+ * the number of online Rx queues.
+ */
+ if (fsp->ring_cookie == RX_CLS_FLOW_DISC ||
+ fsp->ring_cookie >= adapter->num_rx_queues) {
+ dev_err(&adapter->pdev->dev, "ethtool -N: The specified action is invalid\n");
+ return -EINVAL;
+ }
+
+ /* Don't allow indexes to exist outside of available space */
+ if (fsp->location >= IGC_MAX_RXNFC_FILTERS) {
+ dev_err(&adapter->pdev->dev, "Location out of range\n");
+ return -EINVAL;
+ }
+
+ if ((fsp->flow_type & ~FLOW_EXT) != ETHER_FLOW)
+ return -EINVAL;
+
+ input = kzalloc(sizeof(*input), GFP_KERNEL);
+ if (!input)
+ return -ENOMEM;
+
+ if (fsp->m_u.ether_spec.h_proto == ETHER_TYPE_FULL_MASK) {
+ input->filter.etype = fsp->h_u.ether_spec.h_proto;
+ input->filter.match_flags = IGC_FILTER_FLAG_ETHER_TYPE;
+ }
+
+ /* Only support matching addresses by the full mask */
+ if (is_broadcast_ether_addr(fsp->m_u.ether_spec.h_source)) {
+ input->filter.match_flags |= IGC_FILTER_FLAG_SRC_MAC_ADDR;
+ ether_addr_copy(input->filter.src_addr,
+ fsp->h_u.ether_spec.h_source);
+ }
+
+ /* Only support matching addresses by the full mask */
+ if (is_broadcast_ether_addr(fsp->m_u.ether_spec.h_dest)) {
+ input->filter.match_flags |= IGC_FILTER_FLAG_DST_MAC_ADDR;
+ ether_addr_copy(input->filter.dst_addr,
+ fsp->h_u.ether_spec.h_dest);
+ }
+
+ if ((fsp->flow_type & FLOW_EXT) && fsp->m_ext.vlan_tci) {
+ if (fsp->m_ext.vlan_tci != htons(VLAN_PRIO_MASK)) {
+ err = -EINVAL;
+ goto err_out;
+ }
+ input->filter.vlan_tci = fsp->h_ext.vlan_tci;
+ input->filter.match_flags |= IGC_FILTER_FLAG_VLAN_TCI;
+ }
+
+ input->action = fsp->ring_cookie;
+ input->sw_idx = fsp->location;
+
+ spin_lock(&adapter->nfc_lock);
+
+ hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) {
+ if (!memcmp(&input->filter, &rule->filter,
+ sizeof(input->filter))) {
+ err = -EEXIST;
+ dev_err(&adapter->pdev->dev,
+ "ethtool: this filter is already set\n");
+ goto err_out_w_lock;
+ }
+ }
+
+ err = igc_add_filter(adapter, input);
+ if (err)
+ goto err_out_w_lock;
+
+ igc_update_ethtool_nfc_entry(adapter, input, input->sw_idx);
+
+ spin_unlock(&adapter->nfc_lock);
+ return 0;
+
+err_out_w_lock:
+ spin_unlock(&adapter->nfc_lock);
+err_out:
+ kfree(input);
+ return err;
+}
+
+static int igc_del_ethtool_nfc_entry(struct igc_adapter *adapter,
+ struct ethtool_rxnfc *cmd)
+{
+ struct ethtool_rx_flow_spec *fsp =
+ (struct ethtool_rx_flow_spec *)&cmd->fs;
+ int err;
+
+ spin_lock(&adapter->nfc_lock);
+ err = igc_update_ethtool_nfc_entry(adapter, NULL, fsp->location);
+ spin_unlock(&adapter->nfc_lock);
+
+ return err;
+}
+
+static int igc_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+ struct igc_adapter *adapter = netdev_priv(dev);
+ int ret = -EOPNOTSUPP;
+
+ switch (cmd->cmd) {
+ case ETHTOOL_SRXFH:
+ ret = igc_set_rss_hash_opt(adapter, cmd);
+ break;
+ case ETHTOOL_SRXCLSRLINS:
+ ret = igc_add_ethtool_nfc_entry(adapter, cmd);
+ break;
+ case ETHTOOL_SRXCLSRLDEL:
+ ret = igc_del_ethtool_nfc_entry(adapter, cmd);
+ default:
+ break;
+ }
+
+ return ret;
+}
+
void igc_write_rss_indir_tbl(struct igc_adapter *adapter)
{
struct igc_hw *hw = &adapter->hw;
@@ -885,17 +1715,13 @@ static int igc_get_link_ksettings(struct net_device *netdev,
if (hw->mac.type == igc_i225 &&
(status & IGC_STATUS_SPEED_2500)) {
speed = SPEED_2500;
- hw_dbg("2500 Mbs, ");
} else {
speed = SPEED_1000;
- hw_dbg("1000 Mbs, ");
}
} else if (status & IGC_STATUS_SPEED_100) {
speed = SPEED_100;
- hw_dbg("100 Mbs, ");
} else {
speed = SPEED_10;
- hw_dbg("10 Mbs, ");
}
if ((status & IGC_STATUS_FD) ||
hw->phy.media_type != igc_media_type_copper)
@@ -1011,8 +1837,13 @@ static const struct ethtool_ops igc_ethtool_ops = {
.set_ringparam = igc_set_ringparam,
.get_pauseparam = igc_get_pauseparam,
.set_pauseparam = igc_set_pauseparam,
+ .get_strings = igc_get_strings,
+ .get_sset_count = igc_get_sset_count,
+ .get_ethtool_stats = igc_get_ethtool_stats,
.get_coalesce = igc_get_coalesce,
.set_coalesce = igc_set_coalesce,
+ .get_rxnfc = igc_get_rxnfc,
+ .set_rxnfc = igc_set_rxnfc,
.get_rxfh_indir_size = igc_get_rxfh_indir_size,
.get_rxfh = igc_get_rxfh,
.set_rxfh = igc_set_rxfh,
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 87a11879bf2d..a883b3f357e7 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -620,6 +620,55 @@ static void igc_configure_tx(struct igc_adapter *adapter)
*/
static void igc_setup_mrqc(struct igc_adapter *adapter)
{
+ struct igc_hw *hw = &adapter->hw;
+ u32 j, num_rx_queues;
+ u32 mrqc, rxcsum;
+ u32 rss_key[10];
+
+ netdev_rss_key_fill(rss_key, sizeof(rss_key));
+ for (j = 0; j < 10; j++)
+ wr32(IGC_RSSRK(j), rss_key[j]);
+
+ num_rx_queues = adapter->rss_queues;
+
+ if (adapter->rss_indir_tbl_init != num_rx_queues) {
+ for (j = 0; j < IGC_RETA_SIZE; j++)
+ adapter->rss_indir_tbl[j] =
+ (j * num_rx_queues) / IGC_RETA_SIZE;
+ adapter->rss_indir_tbl_init = num_rx_queues;
+ }
+ igc_write_rss_indir_tbl(adapter);
+
+ /* Disable raw packet checksumming so that RSS hash is placed in
+ * descriptor on writeback. No need to enable TCP/UDP/IP checksum
+ * offloads as they are enabled by default
+ */
+ rxcsum = rd32(IGC_RXCSUM);
+ rxcsum |= IGC_RXCSUM_PCSD;
+
+ /* Enable Receive Checksum Offload for SCTP */
+ rxcsum |= IGC_RXCSUM_CRCOFL;
+
+ /* Don't need to set TUOFL or IPOFL, they default to 1 */
+ wr32(IGC_RXCSUM, rxcsum);
+
+ /* Generate RSS hash based on packet types, TCP/UDP
+ * port numbers and/or IPv4/v6 src and dst addresses
+ */
+ mrqc = IGC_MRQC_RSS_FIELD_IPV4 |
+ IGC_MRQC_RSS_FIELD_IPV4_TCP |
+ IGC_MRQC_RSS_FIELD_IPV6 |
+ IGC_MRQC_RSS_FIELD_IPV6_TCP |
+ IGC_MRQC_RSS_FIELD_IPV6_TCP_EX;
+
+ if (adapter->flags & IGC_FLAG_RSS_FIELD_IPV4_UDP)
+ mrqc |= IGC_MRQC_RSS_FIELD_IPV4_UDP;
+ if (adapter->flags & IGC_FLAG_RSS_FIELD_IPV6_UDP)
+ mrqc |= IGC_MRQC_RSS_FIELD_IPV6_UDP;
+
+ mrqc |= IGC_MRQC_ENABLE_RSS_MQ;
+
+ wr32(IGC_MRQC, mrqc);
}
/**
@@ -1738,12 +1787,200 @@ void igc_up(struct igc_adapter *adapter)
* igc_update_stats - Update the board statistics counters
* @adapter: board private structure
*/
-static void igc_update_stats(struct igc_adapter *adapter)
+void igc_update_stats(struct igc_adapter *adapter)
{
+ struct rtnl_link_stats64 *net_stats = &adapter->stats64;
+ struct pci_dev *pdev = adapter->pdev;
+ struct igc_hw *hw = &adapter->hw;
+ u64 _bytes, _packets;
+ u64 bytes, packets;
+ unsigned int start;
+ u32 mpc;
+ int i;
+
+ /* Prevent stats update while adapter is being reset, or if the pci
+ * connection is down.
+ */
+ if (adapter->link_speed == 0)
+ return;
+ if (pci_channel_offline(pdev))
+ return;
+
+ packets = 0;
+ bytes = 0;
+
+ rcu_read_lock();
+ for (i = 0; i < adapter->num_rx_queues; i++) {
+ struct igc_ring *ring = adapter->rx_ring[i];
+ u32 rqdpc = rd32(IGC_RQDPC(i));
+
+ if (hw->mac.type >= igc_i225)
+ wr32(IGC_RQDPC(i), 0);
+
+ if (rqdpc) {
+ ring->rx_stats.drops += rqdpc;
+ net_stats->rx_fifo_errors += rqdpc;
+ }
+
+ do {
+ start = u64_stats_fetch_begin_irq(&ring->rx_syncp);
+ _bytes = ring->rx_stats.bytes;
+ _packets = ring->rx_stats.packets;
+ } while (u64_stats_fetch_retry_irq(&ring->rx_syncp, start));
+ bytes += _bytes;
+ packets += _packets;
+ }
+
+ net_stats->rx_bytes = bytes;
+ net_stats->rx_packets = packets;
+
+ packets = 0;
+ bytes = 0;
+ for (i = 0; i < adapter->num_tx_queues; i++) {
+ struct igc_ring *ring = adapter->tx_ring[i];
+
+ do {
+ start = u64_stats_fetch_begin_irq(&ring->tx_syncp);
+ _bytes = ring->tx_stats.bytes;
+ _packets = ring->tx_stats.packets;
+ } while (u64_stats_fetch_retry_irq(&ring->tx_syncp, start));
+ bytes += _bytes;
+ packets += _packets;
+ }
+ net_stats->tx_bytes = bytes;
+ net_stats->tx_packets = packets;
+ rcu_read_unlock();
+
+ /* read stats registers */
+ adapter->stats.crcerrs += rd32(IGC_CRCERRS);
+ adapter->stats.gprc += rd32(IGC_GPRC);
+ adapter->stats.gorc += rd32(IGC_GORCL);
+ rd32(IGC_GORCH); /* clear GORCL */
+ adapter->stats.bprc += rd32(IGC_BPRC);
+ adapter->stats.mprc += rd32(IGC_MPRC);
+ adapter->stats.roc += rd32(IGC_ROC);
+
+ adapter->stats.prc64 += rd32(IGC_PRC64);
+ adapter->stats.prc127 += rd32(IGC_PRC127);
+ adapter->stats.prc255 += rd32(IGC_PRC255);
+ adapter->stats.prc511 += rd32(IGC_PRC511);
+ adapter->stats.prc1023 += rd32(IGC_PRC1023);
+ adapter->stats.prc1522 += rd32(IGC_PRC1522);
+ adapter->stats.symerrs += rd32(IGC_SYMERRS);
+ adapter->stats.sec += rd32(IGC_SEC);
+
+ mpc = rd32(IGC_MPC);
+ adapter->stats.mpc += mpc;
+ net_stats->rx_fifo_errors += mpc;
+ adapter->stats.scc += rd32(IGC_SCC);
+ adapter->stats.ecol += rd32(IGC_ECOL);
+ adapter->stats.mcc += rd32(IGC_MCC);
+ adapter->stats.latecol += rd32(IGC_LATECOL);
+ adapter->stats.dc += rd32(IGC_DC);
+ adapter->stats.rlec += rd32(IGC_RLEC);
+ adapter->stats.xonrxc += rd32(IGC_XONRXC);
+ adapter->stats.xontxc += rd32(IGC_XONTXC);
+ adapter->stats.xoffrxc += rd32(IGC_XOFFRXC);
+ adapter->stats.xofftxc += rd32(IGC_XOFFTXC);
+ adapter->stats.fcruc += rd32(IGC_FCRUC);
+ adapter->stats.gptc += rd32(IGC_GPTC);
+ adapter->stats.gotc += rd32(IGC_GOTCL);
+ rd32(IGC_GOTCH); /* clear GOTCL */
+ adapter->stats.rnbc += rd32(IGC_RNBC);
+ adapter->stats.ruc += rd32(IGC_RUC);
+ adapter->stats.rfc += rd32(IGC_RFC);
+ adapter->stats.rjc += rd32(IGC_RJC);
+ adapter->stats.tor += rd32(IGC_TORH);
+ adapter->stats.tot += rd32(IGC_TOTH);
+ adapter->stats.tpr += rd32(IGC_TPR);
+
+ adapter->stats.ptc64 += rd32(IGC_PTC64);
+ adapter->stats.ptc127 += rd32(IGC_PTC127);
+ adapter->stats.ptc255 += rd32(IGC_PTC255);
+ adapter->stats.ptc511 += rd32(IGC_PTC511);
+ adapter->stats.ptc1023 += rd32(IGC_PTC1023);
+ adapter->stats.ptc1522 += rd32(IGC_PTC1522);
+
+ adapter->stats.mptc += rd32(IGC_MPTC);
+ adapter->stats.bptc += rd32(IGC_BPTC);
+
+ adapter->stats.tpt += rd32(IGC_TPT);
+ adapter->stats.colc += rd32(IGC_COLC);
+
+ adapter->stats.algnerrc += rd32(IGC_ALGNERRC);
+
+ adapter->stats.tsctc += rd32(IGC_TSCTC);
+ adapter->stats.tsctfc += rd32(IGC_TSCTFC);
+
+ adapter->stats.iac += rd32(IGC_IAC);
+ adapter->stats.icrxoc += rd32(IGC_ICRXOC);
+ adapter->stats.icrxptc += rd32(IGC_ICRXPTC);
+ adapter->stats.icrxatc += rd32(IGC_ICRXATC);
+ adapter->stats.ictxptc += rd32(IGC_ICTXPTC);
+ adapter->stats.ictxatc += rd32(IGC_ICTXATC);
+ adapter->stats.ictxqec += rd32(IGC_ICTXQEC);
+ adapter->stats.ictxqmtc += rd32(IGC_ICTXQMTC);
+ adapter->stats.icrxdmtc += rd32(IGC_ICRXDMTC);
+
+ /* Fill out the OS statistics structure */
+ net_stats->multicast = adapter->stats.mprc;
+ net_stats->collisions = adapter->stats.colc;
+
+ /* Rx Errors */
+
+ /* RLEC on some newer hardware can be incorrect so build
+ * our own version based on RUC and ROC
+ */
+ net_stats->rx_errors = adapter->stats.rxerrc +
+ adapter->stats.crcerrs + adapter->stats.algnerrc +
+ adapter->stats.ruc + adapter->stats.roc +
+ adapter->stats.cexterr;
+ net_stats->rx_length_errors = adapter->stats.ruc +
+ adapter->stats.roc;
+ net_stats->rx_crc_errors = adapter->stats.crcerrs;
+ net_stats->rx_frame_errors = adapter->stats.algnerrc;
+ net_stats->rx_missed_errors = adapter->stats.mpc;
+
+ /* Tx Errors */
+ net_stats->tx_errors = adapter->stats.ecol +
+ adapter->stats.latecol;
+ net_stats->tx_aborted_errors = adapter->stats.ecol;
+ net_stats->tx_window_errors = adapter->stats.latecol;
+ net_stats->tx_carrier_errors = adapter->stats.tncrs;
+
+ /* Tx Dropped needs to be maintained elsewhere */
+
+ /* Management Stats */
+ adapter->stats.mgptc += rd32(IGC_MGTPTC);
+ adapter->stats.mgprc += rd32(IGC_MGTPRC);
+ adapter->stats.mgpdc += rd32(IGC_MGTPDC);
}
static void igc_nfc_filter_exit(struct igc_adapter *adapter)
{
+ struct igc_nfc_filter *rule;
+
+ spin_lock(&adapter->nfc_lock);
+
+ hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
+ igc_erase_filter(adapter, rule);
+
+ hlist_for_each_entry(rule, &adapter->cls_flower_list, nfc_node)
+ igc_erase_filter(adapter, rule);
+
+ spin_unlock(&adapter->nfc_lock);
+}
+
+static void igc_nfc_filter_restore(struct igc_adapter *adapter)
+{
+ struct igc_nfc_filter *rule;
+
+ spin_lock(&adapter->nfc_lock);
+
+ hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
+ igc_add_filter(adapter, rule);
+
+ spin_unlock(&adapter->nfc_lock);
}
/**
@@ -1890,6 +2127,86 @@ static struct net_device_stats *igc_get_stats(struct net_device *netdev)
return &netdev->stats;
}
+static netdev_features_t igc_fix_features(struct net_device *netdev,
+ netdev_features_t features)
+{
+ /* Since there is no support for separate Rx/Tx vlan accel
+ * enable/disable make sure Tx flag is always in same state as Rx.
+ */
+ if (features & NETIF_F_HW_VLAN_CTAG_RX)
+ features |= NETIF_F_HW_VLAN_CTAG_TX;
+ else
+ features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+
+ return features;
+}
+
+static int igc_set_features(struct net_device *netdev,
+ netdev_features_t features)
+{
+ netdev_features_t changed = netdev->features ^ features;
+ struct igc_adapter *adapter = netdev_priv(netdev);
+
+ /* Add VLAN support */
+ if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE)))
+ return 0;
+
+ if (!(features & NETIF_F_NTUPLE)) {
+ struct hlist_node *node2;
+ struct igc_nfc_filter *rule;
+
+ spin_lock(&adapter->nfc_lock);
+ hlist_for_each_entry_safe(rule, node2,
+ &adapter->nfc_filter_list, nfc_node) {
+ igc_erase_filter(adapter, rule);
+ hlist_del(&rule->nfc_node);
+ kfree(rule);
+ }
+ spin_unlock(&adapter->nfc_lock);
+ adapter->nfc_filter_count = 0;
+ }
+
+ netdev->features = features;
+
+ if (netif_running(netdev))
+ igc_reinit_locked(adapter);
+ else
+ igc_reset(adapter);
+
+ return 1;
+}
+
+static netdev_features_t
+igc_features_check(struct sk_buff *skb, struct net_device *dev,
+ netdev_features_t features)
+{
+ unsigned int network_hdr_len, mac_hdr_len;
+
+ /* Make certain the headers can be described by a context descriptor */
+ mac_hdr_len = skb_network_header(skb) - skb->data;
+ if (unlikely(mac_hdr_len > IGC_MAX_MAC_HDR_LEN))
+ return features & ~(NETIF_F_HW_CSUM |
+ NETIF_F_SCTP_CRC |
+ NETIF_F_HW_VLAN_CTAG_TX |
+ NETIF_F_TSO |
+ NETIF_F_TSO6);
+
+ network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
+ if (unlikely(network_hdr_len > IGC_MAX_NETWORK_HDR_LEN))
+ return features & ~(NETIF_F_HW_CSUM |
+ NETIF_F_SCTP_CRC |
+ NETIF_F_TSO |
+ NETIF_F_TSO6);
+
+ /* We can only support IPv4 TSO in tunnels if we can mangle the
+ * inner IP ID field, so strip TSO if MANGLEID is not supported.
+ */
+ if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
+ features &= ~NETIF_F_TSO;
+
+ return features;
+}
+
/**
* igc_configure - configure the hardware for RX and TX
* @adapter: private board structure
@@ -1906,6 +2223,7 @@ static void igc_configure(struct igc_adapter *adapter)
igc_setup_mrqc(adapter);
igc_setup_rctl(adapter);
+ igc_nfc_filter_restore(adapter);
igc_configure_tx(adapter);
igc_configure_rx(adapter);
@@ -1967,6 +2285,127 @@ static void igc_set_default_mac_filter(struct igc_adapter *adapter)
igc_rar_set_index(adapter, 0);
}
+/* If the filter to be added and an already existing filter express
+ * the same address and address type, it should be possible to only
+ * override the other configurations, for example the queue to steer
+ * traffic.
+ */
+static bool igc_mac_entry_can_be_used(const struct igc_mac_addr *entry,
+ const u8 *addr, const u8 flags)
+{
+ if (!(entry->state & IGC_MAC_STATE_IN_USE))
+ return true;
+
+ if ((entry->state & IGC_MAC_STATE_SRC_ADDR) !=
+ (flags & IGC_MAC_STATE_SRC_ADDR))
+ return false;
+
+ if (!ether_addr_equal(addr, entry->addr))
+ return false;
+
+ return true;
+}
+
+/* Add a MAC filter for 'addr' directing matching traffic to 'queue',
+ * 'flags' is used to indicate what kind of match is made, match is by
+ * default for the destination address, if matching by source address
+ * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used.
+ */
+static int igc_add_mac_filter_flags(struct igc_adapter *adapter,
+ const u8 *addr, const u8 queue,
+ const u8 flags)
+{
+ struct igc_hw *hw = &adapter->hw;
+ int rar_entries = hw->mac.rar_entry_count;
+ int i;
+
+ if (is_zero_ether_addr(addr))
+ return -EINVAL;
+
+ /* Search for the first empty entry in the MAC table.
+ * Do not touch entries at the end of the table reserved for the VF MAC
+ * addresses.
+ */
+ for (i = 0; i < rar_entries; i++) {
+ if (!igc_mac_entry_can_be_used(&adapter->mac_table[i],
+ addr, flags))
+ continue;
+
+ ether_addr_copy(adapter->mac_table[i].addr, addr);
+ adapter->mac_table[i].queue = queue;
+ adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE | flags;
+
+ igc_rar_set_index(adapter, i);
+ return i;
+ }
+
+ return -ENOSPC;
+}
+
+int igc_add_mac_steering_filter(struct igc_adapter *adapter,
+ const u8 *addr, u8 queue, u8 flags)
+{
+ return igc_add_mac_filter_flags(adapter, addr, queue,
+ IGC_MAC_STATE_QUEUE_STEERING | flags);
+}
+
+/* Remove a MAC filter for 'addr' directing matching traffic to
+ * 'queue', 'flags' is used to indicate what kind of match need to be
+ * removed, match is by default for the destination address, if
+ * matching by source address is to be removed the flag
+ * IGC_MAC_STATE_SRC_ADDR can be used.
+ */
+static int igc_del_mac_filter_flags(struct igc_adapter *adapter,
+ const u8 *addr, const u8 queue,
+ const u8 flags)
+{
+ struct igc_hw *hw = &adapter->hw;
+ int rar_entries = hw->mac.rar_entry_count;
+ int i;
+
+ if (is_zero_ether_addr(addr))
+ return -EINVAL;
+
+ /* Search for matching entry in the MAC table based on given address
+ * and queue. Do not touch entries at the end of the table reserved
+ * for the VF MAC addresses.
+ */
+ for (i = 0; i < rar_entries; i++) {
+ if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE))
+ continue;
+ if ((adapter->mac_table[i].state & flags) != flags)
+ continue;
+ if (adapter->mac_table[i].queue != queue)
+ continue;
+ if (!ether_addr_equal(adapter->mac_table[i].addr, addr))
+ continue;
+
+ /* When a filter for the default address is "deleted",
+ * we return it to its initial configuration
+ */
+ if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) {
+ adapter->mac_table[i].state =
+ IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE;
+ } else {
+ adapter->mac_table[i].state = 0;
+ adapter->mac_table[i].queue = 0;
+ memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
+ }
+
+ igc_rar_set_index(adapter, i);
+ return 0;
+ }
+
+ return -ENOENT;
+}
+
+int igc_del_mac_steering_filter(struct igc_adapter *adapter,
+ const u8 *addr, u8 queue, u8 flags)
+{
+ return igc_del_mac_filter_flags(adapter, addr, queue,
+ IGC_MAC_STATE_QUEUE_STEERING | flags);
+}
+
/**
* igc_set_rx_mode - Secondary Unicast, Multicast and Promiscuous mode set
* @netdev: network interface device structure
@@ -3434,6 +3873,9 @@ static const struct net_device_ops igc_netdev_ops = {
.ndo_set_mac_address = igc_set_mac,
.ndo_change_mtu = igc_change_mtu,
.ndo_get_stats = igc_get_stats,
+ .ndo_fix_features = igc_fix_features,
+ .ndo_set_features = igc_set_features,
+ .ndo_features_check = igc_features_check,
};
/* PCIe configuration access */
@@ -3663,6 +4105,9 @@ static int igc_probe(struct pci_dev *pdev,
if (err)
goto err_sw_init;
+ /* copy netdev features into list of user selectable features */
+ netdev->hw_features |= NETIF_F_NTUPLE;
+
/* MTU range: 68 - 9216 */
netdev->min_mtu = ETH_MIN_MTU;
netdev->max_mtu = MAX_STD_JUMBO_FRAME_SIZE;
diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h
index 5afe7a8d3faf..50d7c04dccf5 100644
--- a/drivers/net/ethernet/intel/igc/igc_regs.h
+++ b/drivers/net/ethernet/intel/igc/igc_regs.h
@@ -80,8 +80,23 @@
/* MSI-X Table Register Descriptions */
#define IGC_PBACL 0x05B68 /* MSIx PBA Clear - R/W 1 to clear */
+/* RSS registers */
+#define IGC_MRQC 0x05818 /* Multiple Receive Control - RW */
+
+/* Filtering Registers */
+#define IGC_ETQF(_n) (0x05CB0 + (4 * (_n))) /* EType Queue Fltr */
+
+/* ETQF register bit definitions */
+#define IGC_ETQF_FILTER_ENABLE BIT(26)
+#define IGC_ETQF_QUEUE_ENABLE BIT(31)
+#define IGC_ETQF_QUEUE_SHIFT 16
+#define IGC_ETQF_QUEUE_MASK 0x00070000
+#define IGC_ETQF_ETYPE_MASK 0x0000FFFF
+
/* Redirection Table - RW Array */
#define IGC_RETA(_i) (0x05C00 + ((_i) * 4))
+/* RSS Random Key - RW Array */
+#define IGC_RSSRK(_i) (0x05C80 + ((_i) * 4))
/* Receive Register Descriptions */
#define IGC_RCTL 0x00100 /* Rx Control - RW */
@@ -101,6 +116,7 @@
#define IGC_UTA 0x0A000 /* Unicast Table Array - RW */
#define IGC_RAL(_n) (0x05400 + ((_n) * 0x08))
#define IGC_RAH(_n) (0x05404 + ((_n) * 0x08))
+#define IGC_VLAPQF 0x055B0 /* VLAN Priority Queue Filter VLAPQF */
/* Transmit Register Descriptions */
#define IGC_TCTL 0x00400 /* Tx Control - RW */
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index e100054a3765..16c728984164 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8483,8 +8483,7 @@ static void ixgbe_atr(struct ixgbe_ring *ring,
#ifdef IXGBE_FCOE
static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb,
- struct net_device *sb_dev,
- select_queue_fallback_t fallback)
+ struct net_device *sb_dev)
{
struct ixgbe_adapter *adapter;
struct ixgbe_ring_feature *f;
@@ -8514,7 +8513,7 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb,
break;
/* fall through */
default:
- return fallback(dev, skb, sb_dev);
+ return netdev_pick_tx(dev, skb, sb_dev);
}
f = &adapter->ring_feature[RING_F_FCOE];
@@ -9796,7 +9795,7 @@ static int ixgbe_set_features(struct net_device *netdev,
NETIF_F_HW_VLAN_CTAG_FILTER))
ixgbe_set_rx_mode(netdev);
- return 0;
+ return 1;
}
/**