Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:
 "Highlights:

   1) Support TX_RING in AF_PACKET TPACKET_V3 mode, from Sowmini
      Varadhan.

   2) Simplify classifier state on sk_buff in order to shrink it a bit.
      From Willem de Bruijn.

   3) Introduce SIPHASH and it's usage for secure sequence numbers and
      syncookies. From Jason A. Donenfeld.

   4) Reduce CPU usage for ICMP replies we are going to limit or
      suppress, from Jesper Dangaard Brouer.

   5) Introduce Shared Memory Communications socket layer, from Ursula
      Braun.

   6) Add RACK loss detection and allow it to actually trigger fast
      recovery instead of just assisting after other algorithms have
      triggered it. From Yuchung Cheng.

   7) Add xmit_more and BQL support to mvneta driver, from Simon Guinot.

   8) skb_cow_data avoidance in esp4 and esp6, from Steffen Klassert.

   9) Export MPLS packet stats via netlink, from Robert Shearman.

  10) Significantly improve inet port bind conflict handling, especially
      when an application is restarted and changes it's setting of
      reuseport. From Josef Bacik.

  11) Implement TX batching in vhost_net, from Jason Wang.

  12) Extend the dummy device so that VF (virtual function) features,
      such as configuration, can be more easily tested. From Phil
      Sutter.

  13) Avoid two atomic ops per page on x86 in bnx2x driver, from Eric
      Dumazet.

  14) Add new bpf MAP, implementing a longest prefix match trie. From
      Daniel Mack.

  15) Packet sample offloading support in mlxsw driver, from Yotam Gigi.

  16) Add new aquantia driver, from David VomLehn.

  17) Add bpf tracepoints, from Daniel Borkmann.

  18) Add support for port mirroring to b53 and bcm_sf2 drivers, from
      Florian Fainelli.

  19) Remove custom busy polling in many drivers, it is done in the core
      networking since 4.5 times. From Eric Dumazet.

  20) Support XDP adjust_head in virtio_net, from John Fastabend.

  21) Fix several major holes in neighbour entry confirmation, from
      Julian Anastasov.

  22) Add XDP support to bnxt_en driver, from Michael Chan.

  23) VXLAN offloads for enic driver, from Govindarajulu Varadarajan.

  24) Add IPVTAP driver (IP-VLAN based tap driver) from Sainath Grandhi.

  25) Support GRO in IPSEC protocols, from Steffen Klassert"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1764 commits)
  Revert "ath10k: Search SMBIOS for OEM board file extension"
  net: socket: fix recvmmsg not returning error from sock_error
  bnxt_en: use eth_hw_addr_random()
  bpf: fix unlocking of jited image when module ronx not set
  arch: add ARCH_HAS_SET_MEMORY config
  net: napi_watchdog() can use napi_schedule_irqoff()
  tcp: Revert "tcp: tcp_probe: use spin_lock_bh()"
  net/hsr: use eth_hw_addr_random()
  net: mvpp2: enable building on 64-bit platforms
  net: mvpp2: switch to build_skb() in the RX path
  net: mvpp2: simplify MVPP2_PRS_RI_* definitions
  net: mvpp2: fix indentation of MVPP2_EXT_GLOBAL_CTRL_DEFAULT
  net: mvpp2: remove unused register definitions
  net: mvpp2: simplify mvpp2_bm_bufs_add()
  net: mvpp2: drop useless fields in mvpp2_bm_pool and related code
  net: mvpp2: remove unused 'tx_skb' field of 'struct mvpp2_tx_queue'
  net: mvpp2: release reference to txq_cpu[] entry after unmapping
  net: mvpp2: handle too large value in mvpp2_rx_time_coal_set()
  net: mvpp2: handle too large value handling in mvpp2_rx_pkts_coal_set()
  net: mvpp2: remove useless arguments in mvpp2_rx_{pkts, time}_coal_set
  ...

1 files changed, 108 insertions, 128 deletions

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 66957ac12ca4..dd2b4a300819 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -102,14 +102,13 @@ iwl_mvm_bar_check_trigger(struct iwl_mvm *mvm, const u8 *addr,
 #define OPT_HDR(type, skb, off) \
 	(type *)(skb_network_header(skb) + (off))
 
-static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
-			    struct ieee80211_hdr *hdr,
-			    struct ieee80211_tx_info *info,
-			    struct iwl_tx_cmd *tx_cmd)
+static u16 iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
+			   struct ieee80211_hdr *hdr,
+			   struct ieee80211_tx_info *info)
 {
+	u16 offload_assist = 0;
 #if IS_ENABLED(CONFIG_INET)
 	u16 mh_len = ieee80211_hdrlen(hdr->frame_control);
-	u16 offload_assist = le16_to_cpu(tx_cmd->offload_assist);
 	u8 protocol = 0;
 
 	/*
@@ -117,7 +116,7 @@ static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 	 * compute it
 	 */
 	if (skb->ip_summed != CHECKSUM_PARTIAL || IWL_MVM_SW_TX_CSUM_OFFLOAD)
-		return;
+		goto out;
 
 	/* We do not expect to be requested to csum stuff we do not support */
 	if (WARN_ONCE(!(mvm->hw->netdev_features & IWL_TX_CSUM_NETIF_FLAGS) ||
@@ -125,7 +124,7 @@ static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 		       skb->protocol != htons(ETH_P_IPV6)),
 		      "No support for requested checksum\n")) {
 		skb_checksum_help(skb);
-		return;
+		goto out;
 	}
 
 	if (skb->protocol == htons(ETH_P_IP)) {
@@ -145,7 +144,7 @@ static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 			    protocol != NEXTHDR_HOP &&
 			    protocol != NEXTHDR_DEST) {
 				skb_checksum_help(skb);
-				return;
+				goto out;
 			}
 
 			hp = OPT_HDR(struct ipv6_opt_hdr, skb, off);
@@ -159,7 +158,7 @@ static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (protocol != IPPROTO_TCP && protocol != IPPROTO_UDP) {
 		WARN_ON_ONCE(1);
 		skb_checksum_help(skb);
-		return;
+		goto out;
 	}
 
 	/* enable L4 csum */
@@ -191,8 +190,9 @@ static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 	mh_len /= 2;
 	offload_assist |= mh_len << TX_CMD_OFFLD_MH_SIZE;
 
-	tx_cmd->offload_assist = cpu_to_le16(offload_assist);
+out:
 #endif
+	return offload_assist;
 }
 
 /*
@@ -202,7 +202,6 @@ void iwl_mvm_set_tx_cmd(struct iwl_mvm *mvm, struct sk_buff *skb,
 			struct iwl_tx_cmd *tx_cmd,
 			struct ieee80211_tx_info *info, u8 sta_id)
 {
-	struct ieee80211_tx_info *skb_info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	__le16 fc = hdr->frame_control;
 	u32 tx_flags = le32_to_cpu(tx_cmd->tx_flags);
@@ -284,9 +283,8 @@ void iwl_mvm_set_tx_cmd(struct iwl_mvm *mvm, struct sk_buff *skb,
 		tx_flags |= TX_CMD_FLG_WRITE_TX_POWER;
 
 	tx_cmd->tx_flags = cpu_to_le32(tx_flags);
-	/* Total # bytes to be transmitted */
-	tx_cmd->len = cpu_to_le16((u16)skb->len +
-		(uintptr_t)skb_info->driver_data[0]);
+	/* Total # bytes to be transmitted - PCIe code will adjust for A-MSDU */
+	tx_cmd->len = cpu_to_le16((u16)skb->len);
 	tx_cmd->life_time = cpu_to_le32(TX_CMD_LIFE_TIME_INFINITE);
 	tx_cmd->sta_id = sta_id;
 
@@ -295,7 +293,52 @@ void iwl_mvm_set_tx_cmd(struct iwl_mvm *mvm, struct sk_buff *skb,
 	    !(tx_cmd->offload_assist & cpu_to_le16(BIT(TX_CMD_OFFLD_AMSDU))))
 		tx_cmd->offload_assist |= cpu_to_le16(BIT(TX_CMD_OFFLD_PAD));
 
-	iwl_mvm_tx_csum(mvm, skb, hdr, info, tx_cmd);
+	tx_cmd->offload_assist |=
+		cpu_to_le16(iwl_mvm_tx_csum(mvm, skb, hdr, info));
+}
+
+static u32 iwl_mvm_get_tx_rate(struct iwl_mvm *mvm,
+			       struct ieee80211_tx_info *info,
+			       struct ieee80211_sta *sta)
+{
+	int rate_idx;
+	u8 rate_plcp;
+	u32 rate_flags;
+
+	/* HT rate doesn't make sense for a non data frame */
+	WARN_ONCE(info->control.rates[0].flags & IEEE80211_TX_RC_MCS,
+		  "Got an HT rate (flags:0x%x/mcs:%d) for a non data frame\n",
+		  info->control.rates[0].flags,
+		  info->control.rates[0].idx);
+
+	rate_idx = info->control.rates[0].idx;
+	/* if the rate isn't a well known legacy rate, take the lowest one */
+	if (rate_idx < 0 || rate_idx >= IWL_RATE_COUNT_LEGACY)
+		rate_idx = rate_lowest_index(
+				&mvm->nvm_data->bands[info->band], sta);
+
+	/* For 5 GHZ band, remap mac80211 rate indices into driver indices */
+	if (info->band == NL80211_BAND_5GHZ)
+		rate_idx += IWL_FIRST_OFDM_RATE;
+
+	/* For 2.4 GHZ band, check that there is no need to remap */
+	BUILD_BUG_ON(IWL_FIRST_CCK_RATE != 0);
+
+	/* Get PLCP rate for tx_cmd->rate_n_flags */
+	rate_plcp = iwl_mvm_mac80211_idx_to_hwrate(rate_idx);
+
+	if (info->band == NL80211_BAND_2GHZ &&
+	    !iwl_mvm_bt_coex_is_shared_ant_avail(mvm))
+		rate_flags = mvm->cfg->non_shared_ant << RATE_MCS_ANT_POS;
+	else
+		rate_flags =
+			BIT(mvm->mgmt_last_antenna_idx) << RATE_MCS_ANT_POS;
+
+	/* Set CCK flag as needed */
+	if ((rate_idx >= IWL_FIRST_CCK_RATE) && (rate_idx <= IWL_LAST_CCK_RATE))
+		rate_flags |= RATE_MCS_CCK_MSK;
+
+	return (u32)rate_plcp | rate_flags;
 }
 
 /*
@@ -305,10 +348,6 @@ void iwl_mvm_set_tx_cmd_rate(struct iwl_mvm *mvm, struct iwl_tx_cmd *tx_cmd,
 			    struct ieee80211_tx_info *info,
 			    struct ieee80211_sta *sta, __le16 fc)
 {
-	u32 rate_flags;
-	int rate_idx;
-	u8 rate_plcp;
-
 	/* Set retry limit on RTS packets */
 	tx_cmd->rts_retry_limit = IWL_RTS_DFAULT_RETRY_LIMIT;
 
@@ -337,46 +376,12 @@ void iwl_mvm_set_tx_cmd_rate(struct iwl_mvm *mvm, struct iwl_tx_cmd *tx_cmd,
 			cpu_to_le32(TX_CMD_FLG_ACK | TX_CMD_FLG_BAR);
 	}
 
-	/* HT rate doesn't make sense for a non data frame */
-	WARN_ONCE(info->control.rates[0].flags & IEEE80211_TX_RC_MCS,
-		  "Got an HT rate (flags:0x%x/mcs:%d) for a non data frame (fc:0x%x)\n",
-		  info->control.rates[0].flags,
-		  info->control.rates[0].idx,
-		  le16_to_cpu(fc));
-
-	rate_idx = info->control.rates[0].idx;
-	/* if the rate isn't a well known legacy rate, take the lowest one */
-	if (rate_idx < 0 || rate_idx >= IWL_RATE_COUNT_LEGACY)
-		rate_idx = rate_lowest_index(
-				&mvm->nvm_data->bands[info->band], sta);
-
-	/* For 5 GHZ band, remap mac80211 rate indices into driver indices */
-	if (info->band == NL80211_BAND_5GHZ)
-		rate_idx += IWL_FIRST_OFDM_RATE;
-
-	/* For 2.4 GHZ band, check that there is no need to remap */
-	BUILD_BUG_ON(IWL_FIRST_CCK_RATE != 0);
-
-	/* Get PLCP rate for tx_cmd->rate_n_flags */
-	rate_plcp = iwl_mvm_mac80211_idx_to_hwrate(rate_idx);
-
 	mvm->mgmt_last_antenna_idx =
 		iwl_mvm_next_antenna(mvm, iwl_mvm_get_valid_tx_ant(mvm),
 				     mvm->mgmt_last_antenna_idx);
 
-	if (info->band == NL80211_BAND_2GHZ &&
-	    !iwl_mvm_bt_coex_is_shared_ant_avail(mvm))
-		rate_flags = mvm->cfg->non_shared_ant << RATE_MCS_ANT_POS;
-	else
-		rate_flags =
-			BIT(mvm->mgmt_last_antenna_idx) << RATE_MCS_ANT_POS;
-
-	/* Set CCK flag as needed */
-	if ((rate_idx >= IWL_FIRST_CCK_RATE) && (rate_idx <= IWL_LAST_CCK_RATE))
-		rate_flags |= RATE_MCS_CCK_MSK;
-
 	/* Set the rate in the TX cmd */
-	tx_cmd->rate_n_flags = cpu_to_le32((u32)rate_plcp | rate_flags);
+	tx_cmd->rate_n_flags = cpu_to_le32(iwl_mvm_get_tx_rate(mvm, info, sta));
 }
 
 static inline void iwl_mvm_set_tx_cmd_pn(struct ieee80211_tx_info *info,
@@ -459,7 +464,6 @@ iwl_mvm_set_tx_params(struct iwl_mvm *mvm, struct sk_buff *skb,
 		      struct ieee80211_sta *sta, u8 sta_id)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
-	struct ieee80211_tx_info *skb_info = IEEE80211_SKB_CB(skb);
 	struct iwl_device_cmd *dev_cmd;
 	struct iwl_tx_cmd *tx_cmd;
 
@@ -479,12 +483,18 @@ iwl_mvm_set_tx_params(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	iwl_mvm_set_tx_cmd_rate(mvm, tx_cmd, info, sta, hdr->frame_control);
 
+	return dev_cmd;
+}
+
+static void iwl_mvm_skb_prepare_status(struct sk_buff *skb,
+				       struct iwl_device_cmd *cmd)
+{
+	struct ieee80211_tx_info *skb_info = IEEE80211_SKB_CB(skb);
+
 	memset(&skb_info->status, 0, sizeof(skb_info->status));
 	memset(skb_info->driver_data, 0, sizeof(skb_info->driver_data));
 
-	skb_info->driver_data[1] = dev_cmd;
-
-	return dev_cmd;
+	skb_info->driver_data[1] = cmd;
 }
 
 static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
@@ -496,15 +506,17 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm,
 	switch (info->control.vif->type) {
 	case NL80211_IFTYPE_AP:
 		/*
-		 * handle legacy hostapd as well, where station may be added
-		 * only after assoc.
+		 * Handle legacy hostapd as well, where station may be added
+		 * only after assoc. Take care of the case where we send a
+		 * deauth to a station that we don't have.
 		 */
-		if (ieee80211_is_probe_resp(fc) || ieee80211_is_auth(fc))
+		if (ieee80211_is_probe_resp(fc) || ieee80211_is_auth(fc) ||
+		    ieee80211_is_deauth(fc))
 			return IWL_MVM_DQA_AP_PROBE_RESP_QUEUE;
 		if (info->hw_queue == info->control.vif->cab_queue)
 			return info->hw_queue;
 
-		WARN_ON_ONCE(1);
+		WARN_ONCE(1, "fc=0x%02x", le16_to_cpu(fc));
 		return IWL_MVM_DQA_AP_PROBE_RESP_QUEUE;
 	case NL80211_IFTYPE_P2P_DEVICE:
 		if (ieee80211_is_mgmt(fc))
@@ -536,9 +548,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 	 * queue. STATION (HS2.0) uses the auxiliary context of the FW,
 	 * and hence needs to be sent on the aux queue
 	 */
-	if (IEEE80211_SKB_CB(skb)->hw_queue == IWL_MVM_OFFCHANNEL_QUEUE &&
+	if (skb_info->hw_queue == IWL_MVM_OFFCHANNEL_QUEUE &&
 	    skb_info->control.vif->type == NL80211_IFTYPE_STATION)
-		IEEE80211_SKB_CB(skb)->hw_queue = mvm->aux_queue;
+		skb_info->hw_queue = mvm->aux_queue;
 
 	memcpy(&info, skb->cb, sizeof(info));
 
@@ -550,9 +562,6 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 			  info.hw_queue != info.control.vif->cab_queue)))
 		return -1;
 
-	/* This holds the amsdu headers length */
-	skb_info->driver_data[0] = (void *)(uintptr_t)0;
-
 	queue = info.hw_queue;
 
 	/*
@@ -563,9 +572,10 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 	 * (this is not possible for unicast packets as a TLDS discovery
 	 * response are sent without a station entry); otherwise use the
 	 * AUX station.
-	 * In DQA mode, if vif is of type STATION and frames are not multicast,
-	 * they should be sent from the BSS queue. For example, TDLS setup
-	 * frames should be sent on this queue, as they go through the AP.
+	 * In DQA mode, if vif is of type STATION and frames are not multicast
+	 * or offchannel, they should be sent from the BSS queue.
+	 * For example, TDLS setup frames should be sent on this queue,
+	 * as they go through the AP.
 	 */
 	sta_id = mvm->aux_sta.sta_id;
 	if (info.control.vif) {
@@ -587,7 +597,8 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 			if (ap_sta_id != IWL_MVM_STATION_COUNT)
 				sta_id = ap_sta_id;
 		} else if (iwl_mvm_is_dqa_supported(mvm) &&
-			   info.control.vif->type == NL80211_IFTYPE_STATION) {
+			   info.control.vif->type == NL80211_IFTYPE_STATION &&
+			   queue != mvm->aux_queue) {
 			queue = IWL_MVM_DQA_BSS_CLIENT_QUEUE;
 		}
 	}
@@ -598,6 +609,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 	if (!dev_cmd)
 		return -1;
 
+	/* From now on, we cannot access info->control */
+	iwl_mvm_skb_prepare_status(skb, dev_cmd);
+
 	tx_cmd = (struct iwl_tx_cmd *)dev_cmd->payload;
 
 	/* Copy MAC header from skb into command buffer */
@@ -634,7 +648,7 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	unsigned int num_subframes, tcp_payload_len, subf_len, max_amsdu_len;
 	bool ipv4 = (skb->protocol == htons(ETH_P_IP));
 	u16 ip_base_id = ipv4 ? ntohs(ip_hdr(skb)->id) : 0;
-	u16 amsdu_add, snap_ip_tcp, pad, i = 0;
+	u16 snap_ip_tcp, pad, i = 0;
 	unsigned int dbg_max_amsdu_len;
 	netdev_features_t netdev_features = NETIF_F_CSUM_MASK | NETIF_F_SG;
 	u8 *qc, tid, txf;
@@ -736,21 +750,6 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	/* This skb fits in one single A-MSDU */
 	if (num_subframes * mss >= tcp_payload_len) {
-		struct ieee80211_tx_info *skb_info = IEEE80211_SKB_CB(skb);
-
-		/*
-		 * Compute the length of all the data added for the A-MSDU.
-		 * This will be used to compute the length to write in the TX
-		 * command. We have: SNAP + IP + TCP for n -1 subframes and
-		 * ETH header for n subframes. Note that the original skb
-		 * already had one set of SNAP / IP / TCP headers.
-		 */
-		num_subframes = DIV_ROUND_UP(tcp_payload_len, mss);
-		amsdu_add = num_subframes * sizeof(struct ethhdr) +
-			(num_subframes - 1) * (snap_ip_tcp + pad);
-		/* This holds the amsdu headers length */
-		skb_info->driver_data[0] = (void *)(uintptr_t)amsdu_add;
-
 		__skb_queue_tail(mpdus_skb, skb);
 		return 0;
 	}
@@ -789,14 +788,6 @@ segment:
 			ip_hdr(tmp)->id = htons(ip_base_id + i * num_subframes);
 
 		if (tcp_payload_len > mss) {
-			struct ieee80211_tx_info *skb_info =
-				IEEE80211_SKB_CB(tmp);
-
-			num_subframes = DIV_ROUND_UP(tcp_payload_len, mss);
-			amsdu_add = num_subframes * sizeof(struct ethhdr) +
-				(num_subframes - 1) * (snap_ip_tcp + pad);
-			skb_info->driver_data[0] =
-				(void *)(uintptr_t)amsdu_add;
 			skb_shinfo(tmp)->gso_size = mss;
 		} else {
 			qc = ieee80211_get_qos_ctl((void *)tmp->data);
@@ -908,7 +899,6 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 		goto drop;
 
 	tx_cmd = (struct iwl_tx_cmd *)dev_cmd->payload;
-	/* From now on, we cannot access info->control */
 
 	/*
 	 * we handle that entirely ourselves -- for uAPSD the firmware
@@ -919,6 +909,10 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	spin_lock(&mvmsta->lock);
 
+	/* nullfunc frames should go to the MGMT queue regardless of QOS,
+	 * the condition of !ieee80211_is_qos_nullfunc(fc) keeps the default
+	 * assignment of MGMT TID
+	 */
 	if (ieee80211_is_data_qos(fc) && !ieee80211_is_qos_nullfunc(fc)) {
 		u8 *qc = NULL;
 		qc = ieee80211_get_qos_ctl(hdr);
@@ -931,27 +925,13 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 		hdr->seq_ctrl &= cpu_to_le16(IEEE80211_SCTL_FRAG);
 		hdr->seq_ctrl |= cpu_to_le16(seq_number);
 		is_ampdu = info->flags & IEEE80211_TX_CTL_AMPDU;
-	} else if (iwl_mvm_is_dqa_supported(mvm) &&
-		   (ieee80211_is_qos_nullfunc(fc) ||
-		    ieee80211_is_nullfunc(fc))) {
-		/*
-		 * nullfunc frames should go to the MGMT queue regardless of QOS
-		 */
-		tid = IWL_MAX_TID_COUNT;
+		if (WARN_ON_ONCE(is_ampdu &&
+				 mvmsta->tid_data[tid].state != IWL_AGG_ON))
+			goto drop_unlock_sta;
 	}
 
-	if (iwl_mvm_is_dqa_supported(mvm)) {
+	if (iwl_mvm_is_dqa_supported(mvm) || is_ampdu)
 		txq_id = mvmsta->tid_data[tid].txq_id;
-
-		if (ieee80211_is_mgmt(fc))
-			tx_cmd->tid_tspec = IWL_TID_NON_QOS;
-	}
-
-	/* Copy MAC header from skb into command buffer */
-	memcpy(tx_cmd->hdr, hdr, hdrlen);
-
-	WARN_ON_ONCE(info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM);
-
 	if (sta->tdls && !iwl_mvm_is_dqa_supported(mvm)) {
 		/* default to TID 0 for non-QoS packets */
 		u8 tdls_tid = tid == IWL_MAX_TID_COUNT ? 0 : tid;
@@ -959,11 +939,10 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 		txq_id = mvmsta->hw_queue[tid_to_mac80211_ac[tdls_tid]];
 	}
 
-	if (is_ampdu) {
-		if (WARN_ON_ONCE(mvmsta->tid_data[tid].state != IWL_AGG_ON))
-			goto drop_unlock_sta;
-		txq_id = mvmsta->tid_data[tid].txq_id;
-	}
+	/* Copy MAC header from skb into command buffer */
+	memcpy(tx_cmd->hdr, hdr, hdrlen);
+
+	WARN_ON_ONCE(info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM);
 
 	/* Check if TXQ needs to be allocated or re-activated */
 	if (unlikely(txq_id == IEEE80211_INVAL_HW_QUEUE ||
@@ -1015,6 +994,9 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 	IWL_DEBUG_TX(mvm, "TX to [%d|%d] Q:%d - seq: 0x%x\n", mvmsta->sta_id,
 		     tid, txq_id, IEEE80211_SEQ_TO_SN(seq_number));
 
+	/* From now on, we cannot access info->control */
+	iwl_mvm_skb_prepare_status(skb, dev_cmd);
+
 	if (iwl_trans_tx(mvm->trans, skb, dev_cmd, txq_id))
 		goto drop_unlock_sta;
 
@@ -1024,7 +1006,10 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb,
 	spin_unlock(&mvmsta->lock);
 
 	/* Increase pending frames count if this isn't AMPDU */
-	if (!is_ampdu)
+	if ((iwl_mvm_is_dqa_supported(mvm) &&
+	     mvmsta->tid_data[tx_cmd->tid_tspec].state != IWL_AGG_ON &&
+	     mvmsta->tid_data[tx_cmd->tid_tspec].state != IWL_AGG_STARTING) ||
+	    (!iwl_mvm_is_dqa_supported(mvm) && !is_ampdu))
 		atomic_inc(&mvm->pending_frames[mvmsta->sta_id]);
 
 	return 0;
@@ -1040,7 +1025,6 @@ int iwl_mvm_tx_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 		   struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
-	struct ieee80211_tx_info *skb_info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_tx_info info;
 	struct sk_buff_head mpdus_skbs;
 	unsigned int payload_len;
@@ -1054,9 +1038,6 @@ int iwl_mvm_tx_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	memcpy(&info, skb->cb, sizeof(info));
 
-	/* This holds the amsdu headers length */
-	skb_info->driver_data[0] = (void *)(uintptr_t)0;
-
 	if (!skb_is_gso(skb))
 		return iwl_mvm_tx_mpdu(mvm, skb, &info, sta);
 
@@ -1295,8 +1276,6 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 
 		memset(&info->status, 0, sizeof(info->status));
 
-		info->flags &= ~IEEE80211_TX_CTL_AMPDU;
-
 		/* inform mac80211 about what happened with the frame */
 		switch (status & TX_STATUS_MSK) {
 		case TX_STATUS_SUCCESS:
@@ -1319,10 +1298,11 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 			(void *)(uintptr_t)le32_to_cpu(tx_resp->initial_rate);
 
 		/* Single frame failure in an AMPDU queue => send BAR */
-		if (txq_id >= mvm->first_agg_queue &&
+		if (info->flags & IEEE80211_TX_CTL_AMPDU &&
 		    !(info->flags & IEEE80211_TX_STAT_ACK) &&
 		    !(info->flags & IEEE80211_TX_STAT_TX_FILTERED))
 			info->flags |= IEEE80211_TX_STAT_AMPDU_NO_BACK;
+		info->flags &= ~IEEE80211_TX_CTL_AMPDU;
 
 		/* W/A FW bug: seq_ctl is wrong when the status isn't success */
 		if (status != TX_STATUS_SUCCESS) {
@@ -1357,7 +1337,7 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 		ieee80211_tx_status(mvm->hw, skb);
 	}
 
-	if (txq_id >= mvm->first_agg_queue) {
+	if (iwl_mvm_is_dqa_supported(mvm) || txq_id >= mvm->first_agg_queue) {
 		/* If this is an aggregation queue, we use the ssn since:
 		 * ssn = wifi seq_num % 256.
 		 * The seq_ctl is the sequence control of the packet to which