Merge tag 'for-linville-20131001' of git://github.com/kvalo/ath

21 files changed, 984 insertions, 1015 deletions

diff --git a/drivers/net/wireless/ath/ath10k/bmi.c b/drivers/net/wireless/ath/ath10k/bmi.c
index 744da6d1c405..a1f099628850 100644
--- a/drivers/net/wireless/ath/ath10k/bmi.c
+++ b/drivers/net/wireless/ath/ath10k/bmi.c
@@ -22,7 +22,8 @@
 
 void ath10k_bmi_start(struct ath10k *ar)
 {
-	ath10k_dbg(ATH10K_DBG_CORE, "BMI started\n");
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi start\n");
+
 	ar->bmi.done_sent = false;
 }
 
@@ -32,8 +33,10 @@ int ath10k_bmi_done(struct ath10k *ar)
 	u32 cmdlen = sizeof(cmd.id) + sizeof(cmd.done);
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi done\n");
+
 	if (ar->bmi.done_sent) {
-		ath10k_dbg(ATH10K_DBG_CORE, "%s skipped\n", __func__);
+		ath10k_dbg(ATH10K_DBG_BMI, "bmi skipped\n");
 		return 0;
 	}
 
@@ -46,7 +49,6 @@ int ath10k_bmi_done(struct ath10k *ar)
 		return ret;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE, "BMI done\n");
 	return 0;
 }
 
@@ -59,6 +61,8 @@ int ath10k_bmi_get_target_info(struct ath10k *ar,
 	u32 resplen = sizeof(resp.get_target_info);
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi get target info\n");
+
 	if (ar->bmi.done_sent) {
 		ath10k_warn("BMI Get Target Info Command disallowed\n");
 		return -EBUSY;
@@ -80,6 +84,7 @@ int ath10k_bmi_get_target_info(struct ath10k *ar,
 
 	target_info->version = __le32_to_cpu(resp.get_target_info.version);
 	target_info->type    = __le32_to_cpu(resp.get_target_info.type);
+
 	return 0;
 }
 
@@ -92,15 +97,14 @@ int ath10k_bmi_read_memory(struct ath10k *ar,
 	u32 rxlen;
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi read address 0x%x length %d\n",
+		   address, length);
+
 	if (ar->bmi.done_sent) {
 		ath10k_warn("command disallowed\n");
 		return -EBUSY;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE,
-		   "%s: (device: 0x%p, address: 0x%x, length: %d)\n",
-		   __func__, ar, address, length);
-
 	while (length) {
 		rxlen = min_t(u32, length, BMI_MAX_DATA_SIZE);
 
@@ -133,15 +137,14 @@ int ath10k_bmi_write_memory(struct ath10k *ar,
 	u32 txlen;
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi write address 0x%x length %d\n",
+		   address, length);
+
 	if (ar->bmi.done_sent) {
 		ath10k_warn("command disallowed\n");
 		return -EBUSY;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE,
-		   "%s: (device: 0x%p, address: 0x%x, length: %d)\n",
-		   __func__, ar, address, length);
-
 	while (length) {
 		txlen = min(length, BMI_MAX_DATA_SIZE - hdrlen);
 
@@ -180,15 +183,14 @@ int ath10k_bmi_execute(struct ath10k *ar, u32 address, u32 *param)
 	u32 resplen = sizeof(resp.execute);
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi execute address 0x%x param 0x%x\n",
+		   address, *param);
+
 	if (ar->bmi.done_sent) {
 		ath10k_warn("command disallowed\n");
 		return -EBUSY;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE,
-		   "%s: (device: 0x%p, address: 0x%x, param: %d)\n",
-		   __func__, ar, address, *param);
-
 	cmd.id            = __cpu_to_le32(BMI_EXECUTE);
 	cmd.execute.addr  = __cpu_to_le32(address);
 	cmd.execute.param = __cpu_to_le32(*param);
@@ -216,6 +218,9 @@ int ath10k_bmi_lz_data(struct ath10k *ar, const void *buffer, u32 length)
 	u32 txlen;
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi lz data buffer 0x%p length %d\n",
+		   buffer, length);
+
 	if (ar->bmi.done_sent) {
 		ath10k_warn("command disallowed\n");
 		return -EBUSY;
@@ -250,6 +255,9 @@ int ath10k_bmi_lz_stream_start(struct ath10k *ar, u32 address)
 	u32 cmdlen = sizeof(cmd.id) + sizeof(cmd.lz_start);
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI, "bmi lz stream start address 0x%x\n",
+		   address);
+
 	if (ar->bmi.done_sent) {
 		ath10k_warn("command disallowed\n");
 		return -EBUSY;
@@ -275,6 +283,10 @@ int ath10k_bmi_fast_download(struct ath10k *ar,
 	u32 trailer_len = length - head_len;
 	int ret;
 
+	ath10k_dbg(ATH10K_DBG_BMI,
+		   "bmi fast download address 0x%x buffer 0x%p length %d\n",
+		   address, buffer, length);
+
 	ret = ath10k_bmi_lz_stream_start(ar, address);
 	if (ret)
 		return ret;
diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index afe5df404ce6..834e29ea236c 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -338,33 +338,19 @@ int ath10k_ce_send(struct ath10k_ce_pipe *ce_state,
 	return ret;
 }
 
-void ath10k_ce_sendlist_buf_add(struct ce_sendlist *sendlist, u32 buffer,
-				unsigned int nbytes, u32 flags)
-{
-	unsigned int num_items = sendlist->num_items;
-	struct ce_sendlist_item *item;
-
-	item = &sendlist->item[num_items];
-	item->data = buffer;
-	item->u.nbytes = nbytes;
-	item->flags = flags;
-	sendlist->num_items++;
-}
-
 int ath10k_ce_sendlist_send(struct ath10k_ce_pipe *ce_state,
 			    void *per_transfer_context,
-			    struct ce_sendlist *sendlist,
-			    unsigned int transfer_id)
+			    unsigned int transfer_id,
+			    u32 paddr, unsigned int nbytes,
+			    u32 flags)
 {
 	struct ath10k_ce_ring *src_ring = ce_state->src_ring;
-	struct ce_sendlist_item *item;
 	struct ath10k *ar = ce_state->ar;
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	unsigned int nentries_mask = src_ring->nentries_mask;
-	unsigned int num_items = sendlist->num_items;
 	unsigned int sw_index;
 	unsigned int write_index;
-	int i, delta, ret = -ENOMEM;
+	int delta, ret = -ENOMEM;
 
 	spin_lock_bh(&ar_pci->ce_lock);
 
@@ -373,30 +359,12 @@ int ath10k_ce_sendlist_send(struct ath10k_ce_pipe *ce_state,
 
 	delta = CE_RING_DELTA(nentries_mask, write_index, sw_index - 1);
 
-	if (delta >= num_items) {
-		/*
-		 * Handle all but the last item uniformly.
-		 */
-		for (i = 0; i < num_items - 1; i++) {
-			item = &sendlist->item[i];
-			ret = ath10k_ce_send_nolock(ce_state,
-						    CE_SENDLIST_ITEM_CTXT,
-						    (u32) item->data,
-						    item->u.nbytes, transfer_id,
-						    item->flags |
-						    CE_SEND_FLAG_GATHER);
-			if (ret)
-				ath10k_warn("CE send failed for item: %d\n", i);
-		}
-		/*
-		 * Provide valid context pointer for final item.
-		 */
-		item = &sendlist->item[i];
+	if (delta >= 1) {
 		ret = ath10k_ce_send_nolock(ce_state, per_transfer_context,
-					    (u32) item->data, item->u.nbytes,
-					    transfer_id, item->flags);
+					    paddr, nbytes,
+					    transfer_id, flags);
 		if (ret)
-			ath10k_warn("CE send failed for last item: %d\n", i);
+			ath10k_warn("CE send failed: %d\n", ret);
 	}
 
 	spin_unlock_bh(&ar_pci->ce_lock);
@@ -742,11 +710,6 @@ void ath10k_ce_per_engine_service(struct ath10k *ar, unsigned int ce_id)
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_ce_pipe *ce_state = &ar_pci->ce_states[ce_id];
 	u32 ctrl_addr = ce_state->ctrl_addr;
-	void *transfer_context;
-	u32 buf;
-	unsigned int nbytes;
-	unsigned int id;
-	unsigned int flags;
 	int ret;
 
 	ret = ath10k_pci_wake(ar);
@@ -759,38 +722,15 @@ void ath10k_ce_per_engine_service(struct ath10k *ar, unsigned int ce_id)
 	ath10k_ce_engine_int_status_clear(ar, ctrl_addr,
 					  HOST_IS_COPY_COMPLETE_MASK);
 
-	if (ce_state->recv_cb) {
-		/*
-		 * Pop completed recv buffers and call the registered
-		 * recv callback for each
-		 */
-		while (ath10k_ce_completed_recv_next_nolock(ce_state,
-							    &transfer_context,
-							    &buf, &nbytes,
-							    &id, &flags) == 0) {
-			spin_unlock_bh(&ar_pci->ce_lock);
-			ce_state->recv_cb(ce_state, transfer_context, buf,
-					  nbytes, id, flags);
-			spin_lock_bh(&ar_pci->ce_lock);
-		}
-	}
+	spin_unlock_bh(&ar_pci->ce_lock);
 
-	if (ce_state->send_cb) {
-		/*
-		 * Pop completed send buffers and call the registered
-		 * send callback for each
-		 */
-		while (ath10k_ce_completed_send_next_nolock(ce_state,
-							    &transfer_context,
-							    &buf,
-							    &nbytes,
-							    &id) == 0) {
-			spin_unlock_bh(&ar_pci->ce_lock);
-			ce_state->send_cb(ce_state, transfer_context,
-					  buf, nbytes, id);
-			spin_lock_bh(&ar_pci->ce_lock);
-		}
-	}
+	if (ce_state->recv_cb)
+		ce_state->recv_cb(ce_state);
+
+	if (ce_state->send_cb)
+		ce_state->send_cb(ce_state);
+
+	spin_lock_bh(&ar_pci->ce_lock);
 
 	/*
 	 * Misc CE interrupts are not being handled, but still need
@@ -881,11 +821,7 @@ void ath10k_ce_disable_interrupts(struct ath10k *ar)
 }
 
 void ath10k_ce_send_cb_register(struct ath10k_ce_pipe *ce_state,
-				void (*send_cb)(struct ath10k_ce_pipe *ce_state,
-						void *transfer_context,
-						u32 buffer,
-						unsigned int nbytes,
-						unsigned int transfer_id),
+				void (*send_cb)(struct ath10k_ce_pipe *),
 				int disable_interrupts)
 {
 	struct ath10k *ar = ce_state->ar;
@@ -898,12 +834,7 @@ void ath10k_ce_send_cb_register(struct ath10k_ce_pipe *ce_state,
 }
 
 void ath10k_ce_recv_cb_register(struct ath10k_ce_pipe *ce_state,
-				void (*recv_cb)(struct ath10k_ce_pipe *ce_state,
-						void *transfer_context,
-						u32 buffer,
-						unsigned int nbytes,
-						unsigned int transfer_id,
-						unsigned int flags))
+				void (*recv_cb)(struct ath10k_ce_pipe *))
 {
 	struct ath10k *ar = ce_state->ar;
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
@@ -1010,6 +941,10 @@ static int ath10k_ce_init_src_ring(struct ath10k *ar,
 	ath10k_ce_src_ring_lowmark_set(ar, ctrl_addr, 0);
 	ath10k_ce_src_ring_highmark_set(ar, ctrl_addr, nentries);
 
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot ce src ring id %d entries %d base_addr %p\n",
+		   ce_id, nentries, src_ring->base_addr_owner_space);
+
 	return 0;
 }
 
@@ -1091,6 +1026,10 @@ static int ath10k_ce_init_dest_ring(struct ath10k *ar,
 	ath10k_ce_dest_ring_lowmark_set(ar, ctrl_addr, 0);
 	ath10k_ce_dest_ring_highmark_set(ar, ctrl_addr, nentries);
 
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot ce dest ring id %d entries %d base_addr %p\n",
+		   ce_id, nentries, dest_ring->base_addr_owner_space);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ath/ath10k/ce.h b/drivers/net/wireless/ath/ath10k/ce.h
index cfef7d0fc5c4..aec802868341 100644
--- a/drivers/net/wireless/ath/ath10k/ce.h
+++ b/drivers/net/wireless/ath/ath10k/ce.h
@@ -27,7 +27,6 @@
 
 /* Descriptor rings must be aligned to this boundary */
 #define CE_DESC_RING_ALIGN	8
-#define CE_SENDLIST_ITEMS_MAX	12
 #define CE_SEND_FLAG_GATHER	0x00010000
 
 /*
@@ -116,41 +115,14 @@ struct ath10k_ce_pipe {
 
 	u32 ctrl_addr;
 
-	void (*send_cb) (struct ath10k_ce_pipe *ce_state,
-			 void *per_transfer_send_context,
-			 u32 buffer,
-			 unsigned int nbytes,
-			 unsigned int transfer_id);
-	void (*recv_cb) (struct ath10k_ce_pipe *ce_state,
-			 void *per_transfer_recv_context,
-			 u32 buffer,
-			 unsigned int nbytes,
-			 unsigned int transfer_id,
-			 unsigned int flags);
+	void (*send_cb)(struct ath10k_ce_pipe *);
+	void (*recv_cb)(struct ath10k_ce_pipe *);
 
 	unsigned int src_sz_max;
 	struct ath10k_ce_ring *src_ring;
 	struct ath10k_ce_ring *dest_ring;
 };
 
-struct ce_sendlist_item {
-	/* e.g. buffer or desc list */
-	dma_addr_t data;
-	union {
-		/* simple buffer */
-		unsigned int nbytes;
-		/* Rx descriptor list */
-		unsigned int ndesc;
-	} u;
-	/* externally-specified flags; OR-ed with internal flags */
-	u32 flags;
-};
-
-struct ce_sendlist {
-	unsigned int num_items;
-	struct ce_sendlist_item item[CE_SENDLIST_ITEMS_MAX];
-};
-
 /* Copy Engine settable attributes */
 struct ce_attr;
 
@@ -181,20 +153,9 @@ int ath10k_ce_send(struct ath10k_ce_pipe *ce_state,
 		   unsigned int flags);
 
 void ath10k_ce_send_cb_register(struct ath10k_ce_pipe *ce_state,
-				void (*send_cb)(struct ath10k_ce_pipe *ce_state,
-						void *transfer_context,
-						u32 buffer,
-						unsigned int nbytes,
-						unsigned int transfer_id),
+				void (*send_cb)(struct ath10k_ce_pipe *),
 				int disable_interrupts);
 
-/* Append a simple buffer (address/length) to a sendlist. */
-void ath10k_ce_sendlist_buf_add(struct ce_sendlist *sendlist,
-				u32 buffer,
-				unsigned int nbytes,
-				/* OR-ed with internal flags */
-				u32 flags);
-
 /*
  * Queue a "sendlist" of buffers to be sent using gather to a single
  * anonymous destination buffer
@@ -206,10 +167,10 @@ void ath10k_ce_sendlist_buf_add(struct ce_sendlist *sendlist,
  * Implemenation note: Pushes multiple buffers with Gather to Source ring.
  */
 int ath10k_ce_sendlist_send(struct ath10k_ce_pipe *ce_state,
-			    void *per_transfer_send_context,
-			    struct ce_sendlist *sendlist,
-			    /* 14 bits */
-			    unsigned int transfer_id);
+			    void *per_transfer_context,
+			    unsigned int transfer_id,
+			    u32 paddr, unsigned int nbytes,
+			    u32 flags);
 
 /*==================Recv=======================*/
 
@@ -228,12 +189,7 @@ int ath10k_ce_recv_buf_enqueue(struct ath10k_ce_pipe *ce_state,
 			       u32 buffer);
 
 void ath10k_ce_recv_cb_register(struct ath10k_ce_pipe *ce_state,
-				void (*recv_cb)(struct ath10k_ce_pipe *ce_state,
-						void *transfer_context,
-						u32 buffer,
-						unsigned int nbytes,
-						unsigned int transfer_id,
-						unsigned int flags));
+				void (*recv_cb)(struct ath10k_ce_pipe *));
 
 /* recv flags */
 /* Data is byte-swapped */
@@ -325,16 +281,6 @@ struct ce_attr {
 	unsigned int dest_nentries;
 };
 
-/*
- * When using sendlist_send to transfer multiple buffer fragments, the
- * transfer context of each fragment, except last one, will be filled
- * with CE_SENDLIST_ITEM_CTXT. ce_completed_send will return success for
- * each fragment done with send and the transfer context would be
- * CE_SENDLIST_ITEM_CTXT. Upper layer could use this to identify the
- * status of a send completion.
- */
-#define CE_SENDLIST_ITEM_CTXT	((void *)0xcecebeef)
-
 #define SR_BA_ADDRESS		0x0000
 #define SR_SIZE_ADDRESS		0x0004
 #define DR_BA_ADDRESS		0x0008
diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 2dd39a82ae99..76906d5a082e 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -53,7 +53,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 
 static void ath10k_send_suspend_complete(struct ath10k *ar)
 {
-	ath10k_dbg(ATH10K_DBG_CORE, "%s\n", __func__);
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot suspend complete\n");
 
 	ar->is_target_paused = true;
 	wake_up(&ar->event_queue);
@@ -101,7 +101,7 @@ static int ath10k_init_connect_htc(struct ath10k *ar)
 		goto timeout;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE, "core wmi ready\n");
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot wmi ready\n");
 	return 0;
 
 timeout:
@@ -203,8 +203,8 @@ static int ath10k_push_board_ext_data(struct ath10k *ar,
 		return ret;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE,
-		   "ath10k: Board extended Data download addr: 0x%x\n",
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot push board extended data addr 0x%x\n",
 		   board_ext_data_addr);
 
 	if (board_ext_data_addr == 0)
@@ -435,6 +435,13 @@ static int ath10k_init_uart(struct ath10k *ar)
 		return ret;
 	}
 
+	/* Set the UART baud rate to 19200. */
+	ret = ath10k_bmi_write32(ar, hi_desired_baud_rate, 19200);
+	if (ret) {
+		ath10k_warn("could not set the baud rate (%d)\n", ret);
+		return ret;
+	}
+
 	ath10k_info("UART prints enabled\n");
 	return 0;
 }
@@ -630,6 +637,10 @@ int ath10k_core_start(struct ath10k *ar)
 	if (status)
 		goto err_disconnect_htc;
 
+	status = ath10k_debug_start(ar);
+	if (status)
+		goto err_disconnect_htc;
+
 	ar->free_vdev_map = (1 << TARGET_NUM_VDEVS) - 1;
 
 	return 0;
@@ -647,6 +658,7 @@ EXPORT_SYMBOL(ath10k_core_start);
 
 void ath10k_core_stop(struct ath10k *ar)
 {
+	ath10k_debug_stop(ar);
 	ath10k_htc_stop(&ar->htc);
 	ath10k_htt_detach(&ar->htt);
 	ath10k_wmi_detach(ar);
@@ -710,6 +722,9 @@ static int ath10k_core_check_chip_id(struct ath10k *ar)
 {
 	u32 hw_revision = MS(ar->chip_id, SOC_CHIP_ID_REV);
 
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot chip_id 0x%08x hw_revision 0x%x\n",
+		   ar->chip_id, hw_revision);
+
 	/* Check that we are not using hw1.0 (some of them have same pci id
 	 * as hw2.0) before doing anything else as ath10k crashes horribly
 	 * due to missing hw1.0 workarounds. */
@@ -777,6 +792,7 @@ void ath10k_core_unregister(struct ath10k *ar)
 	 * Otherwise we will fail to submit commands to FW and mac80211 will be
 	 * unhappy about callback failures. */
 	ath10k_mac_unregister(ar);
+
 	ath10k_core_free_firmware_files(ar);
 }
 EXPORT_SYMBOL(ath10k_core_unregister);
diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 174c4b404223..292ad4577c98 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -52,18 +52,12 @@ struct ath10k_skb_cb {
 
 	struct {
 		u8 vdev_id;
-		u16 msdu_id;
 		u8 tid;
 		bool is_offchan;
-		bool is_conf;
-		bool discard;
-		bool no_ack;
-		u8 refcount;
-		struct sk_buff *txfrag;
-		struct sk_buff *msdu;
-	} __packed htt;
 
-	/* 4 bytes left on 64bit arch */
+		u8 frag_len;
+		u8 pad_len;
+	} __packed htt;
 } __packed;
 
 static inline struct ath10k_skb_cb *ATH10K_SKB_CB(struct sk_buff *skb)
@@ -112,11 +106,7 @@ struct ath10k_wmi {
 	enum ath10k_htc_ep_id eid;
 	struct completion service_ready;
 	struct completion unified_ready;
-	atomic_t pending_tx_count;
-	wait_queue_head_t wq;
-
-	struct sk_buff_head wmi_event_list;
-	struct work_struct wmi_event_work;
+	wait_queue_head_t tx_credits_wq;
 };
 
 struct ath10k_peer_stat {
@@ -203,6 +193,7 @@ struct ath10k_vif {
 	enum wmi_vdev_subtype vdev_subtype;
 	u32 beacon_interval;
 	u32 dtim_period;
+	struct sk_buff *beacon;
 
 	struct ath10k *ar;
 	struct ieee80211_vif *vif;
@@ -246,6 +237,9 @@ struct ath10k_debug {
 	u32 wmi_service_bitmap[WMI_SERVICE_BM_SIZE];
 
 	struct completion event_stats_compl;
+
+	unsigned long htt_stats_mask;
+	struct delayed_work htt_stats_dwork;
 };
 
 enum ath10k_state {
diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 09f535a1c767..59615c7f217e 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -21,6 +21,9 @@
 #include "core.h"
 #include "debug.h"
 
+/* ms */
+#define ATH10K_DEBUG_HTT_STATS_INTERVAL 1000
+
 static int ath10k_printk(const char *level, const char *fmt, ...)
 {
 	struct va_format vaf;
@@ -517,6 +520,117 @@ static const struct file_operations fops_chip_id = {
 	.llseek = default_llseek,
 };
 
+static int ath10k_debug_htt_stats_req(struct ath10k *ar)
+{
+	u64 cookie;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (ar->debug.htt_stats_mask == 0)
+		/* htt stats are disabled */
+		return 0;
+
+	if (ar->state != ATH10K_STATE_ON)
+		return 0;
+
+	cookie = get_jiffies_64();
+
+	ret = ath10k_htt_h2t_stats_req(&ar->htt, ar->debug.htt_stats_mask,
+				       cookie);
+	if (ret) {
+		ath10k_warn("failed to send htt stats request: %d\n", ret);
+		return ret;
+	}
+
+	queue_delayed_work(ar->workqueue, &ar->debug.htt_stats_dwork,
+			   msecs_to_jiffies(ATH10K_DEBUG_HTT_STATS_INTERVAL));
+
+	return 0;
+}
+
+static void ath10k_debug_htt_stats_dwork(struct work_struct *work)
+{
+	struct ath10k *ar = container_of(work, struct ath10k,
+					 debug.htt_stats_dwork.work);
+
+	mutex_lock(&ar->conf_mutex);
+
+	ath10k_debug_htt_stats_req(ar);
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
+static ssize_t ath10k_read_htt_stats_mask(struct file *file,
+					    char __user *user_buf,
+					    size_t count, loff_t *ppos)
+{
+	struct ath10k *ar = file->private_data;
+	char buf[32];
+	unsigned int len;
+
+	len = scnprintf(buf, sizeof(buf), "%lu\n", ar->debug.htt_stats_mask);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static ssize_t ath10k_write_htt_stats_mask(struct file *file,
+					     const char __user *user_buf,
+					     size_t count, loff_t *ppos)
+{
+	struct ath10k *ar = file->private_data;
+	unsigned long mask;
+	int ret;
+
+	ret = kstrtoul_from_user(user_buf, count, 0, &mask);
+	if (ret)
+		return ret;
+
+	/* max 8 bit masks (for now) */
+	if (mask > 0xff)
+		return -E2BIG;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ar->debug.htt_stats_mask = mask;
+
+	ret = ath10k_debug_htt_stats_req(ar);
+	if (ret)
+		goto out;
+
+	ret = count;
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static const struct file_operations fops_htt_stats_mask = {
+	.read = ath10k_read_htt_stats_mask,
+	.write = ath10k_write_htt_stats_mask,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+int ath10k_debug_start(struct ath10k *ar)
+{
+	int ret;
+
+	ret = ath10k_debug_htt_stats_req(ar);
+	if (ret)
+		/* continue normally anyway, this isn't serious */
+		ath10k_warn("failed to start htt stats workqueue: %d\n", ret);
+
+	return 0;
+}
+
+void ath10k_debug_stop(struct ath10k *ar)
+{
+	cancel_delayed_work_sync(&ar->debug.htt_stats_dwork);
+}
+
 int ath10k_debug_create(struct ath10k *ar)
 {
 	ar->debug.debugfs_phy = debugfs_create_dir("ath10k",
@@ -525,6 +639,9 @@ int ath10k_debug_create(struct ath10k *ar)
 	if (!ar->debug.debugfs_phy)
 		return -ENOMEM;
 
+	INIT_DELAYED_WORK(&ar->debug.htt_stats_dwork,
+			  ath10k_debug_htt_stats_dwork);
+
 	init_completion(&ar->debug.event_stats_compl);
 
 	debugfs_create_file("fw_stats", S_IRUSR, ar->debug.debugfs_phy, ar,
@@ -539,8 +656,12 @@ int ath10k_debug_create(struct ath10k *ar)
 	debugfs_create_file("chip_id", S_IRUSR, ar->debug.debugfs_phy,
 			    ar, &fops_chip_id);
 
+	debugfs_create_file("htt_stats_mask", S_IRUSR, ar->debug.debugfs_phy,
+			    ar, &fops_htt_stats_mask);
+
 	return 0;
 }
+
 #endif /* CONFIG_ATH10K_DEBUGFS */
 
 #ifdef CONFIG_ATH10K_DEBUG
diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h
index 168140c54028..fa581486626f 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -27,11 +27,12 @@ enum ath10k_debug_mask {
 	ATH10K_DBG_HTC		= 0x00000004,
 	ATH10K_DBG_HTT		= 0x00000008,
 	ATH10K_DBG_MAC		= 0x00000010,
-	ATH10K_DBG_CORE		= 0x00000020,
+	ATH10K_DBG_BOOT		= 0x00000020,
 	ATH10K_DBG_PCI_DUMP	= 0x00000040,
 	ATH10K_DBG_HTT_DUMP	= 0x00000080,
 	ATH10K_DBG_MGMT		= 0x00000100,
 	ATH10K_DBG_DATA		= 0x00000200,
+	ATH10K_DBG_BMI		= 0x00000400,
 	ATH10K_DBG_ANY		= 0xffffffff,
 };
 
@@ -42,6 +43,8 @@ extern __printf(1, 2) int ath10k_err(const char *fmt, ...);
 extern __printf(1, 2) int ath10k_warn(const char *fmt, ...);
 
 #ifdef CONFIG_ATH10K_DEBUGFS
+int ath10k_debug_start(struct ath10k *ar);
+void ath10k_debug_stop(struct ath10k *ar);
 int ath10k_debug_create(struct ath10k *ar);
 void ath10k_debug_read_service_map(struct ath10k *ar,
 				   void *service_map,
@@ -50,6 +53,15 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 				    struct wmi_stats_event *ev);
 
 #else
+static inline int ath10k_debug_start(struct ath10k *ar)
+{
+	return 0;
+}
+
+static inline void ath10k_debug_stop(struct ath10k *ar)
+{
+}
+
 static inline int ath10k_debug_create(struct ath10k *ar)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 7d445d3b849d..3118d7506734 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -103,10 +103,10 @@ static void ath10k_htc_prepare_tx_skb(struct ath10k_htc_ep *ep,
 	struct ath10k_htc_hdr *hdr;
 
 	hdr = (struct ath10k_htc_hdr *)skb->data;
-	memset(hdr, 0, sizeof(*hdr));
 
 	hdr->eid = ep->eid;
 	hdr->len = __cpu_to_le16(skb->len - sizeof(*hdr));
+	hdr->flags = 0;
 
 	spin_lock_bh(&ep->htc->tx_lock);
 	hdr->seq_no = ep->seq_no++;
@@ -117,134 +117,13 @@ static void ath10k_htc_prepare_tx_skb(struct ath10k_htc_ep *ep,
 	spin_unlock_bh(&ep->htc->tx_lock);
 }
 
-static int ath10k_htc_issue_skb(struct ath10k_htc *htc,
-				struct ath10k_htc_ep *ep,
-				struct sk_buff *skb,
-				u8 credits)
-{
-	struct ath10k_skb_cb *skb_cb = ATH10K_SKB_CB(skb);
-	int ret;
-
-	ath10k_dbg(ATH10K_DBG_HTC, "%s: ep %d skb %p\n", __func__,
-		   ep->eid, skb);
-
-	ath10k_htc_prepare_tx_skb(ep, skb);
-
-	ret = ath10k_skb_map(htc->ar->dev, skb);
-	if (ret)
-		goto err;
-
-	ret = ath10k_hif_send_head(htc->ar,
-				   ep->ul_pipe_id,
-				   ep->eid,
-				   skb->len,
-				   skb);
-	if (unlikely(ret))
-		goto err;
-
-	return 0;
-err:
-	ath10k_warn("HTC issue failed: %d\n", ret);
-
-	spin_lock_bh(&htc->tx_lock);
-	ep->tx_credits += credits;
-	spin_unlock_bh(&htc->tx_lock);
-
-	/* this is the simplest way to handle out-of-resources for non-credit
-	 * based endpoints. credit based endpoints can still get -ENOSR, but
-	 * this is highly unlikely as credit reservation should prevent that */
-	if (ret == -ENOSR) {
-		spin_lock_bh(&htc->tx_lock);
-		__skb_queue_head(&ep->tx_queue, skb);
-		spin_unlock_bh(&htc->tx_lock);
-
-		return ret;
-	}
-
-	skb_cb->is_aborted = true;
-	ath10k_htc_notify_tx_completion(ep, skb);
-
-	return ret;
-}
-
-static struct sk_buff *ath10k_htc_get_skb_credit_based(struct ath10k_htc *htc,
-						       struct ath10k_htc_ep *ep,
-						       u8 *credits)
-{
-	struct sk_buff *skb;
-	struct ath10k_skb_cb *skb_cb;
-	int credits_required;
-	int remainder;
-	unsigned int transfer_len;
-
-	lockdep_assert_held(&htc->tx_lock);
-
-	skb = __skb_dequeue(&ep->tx_queue);
-	if (!skb)
-		return NULL;
-
-	skb_cb = ATH10K_SKB_CB(skb);
-	transfer_len = skb->len;
-
-	if (likely(transfer_len <= htc->target_credit_size)) {
-		credits_required = 1;
-	} else {
-		/* figure out how many credits this message requires */
-		credits_required = transfer_len / htc->target_credit_size;
-		remainder = transfer_len % htc->target_credit_size;
-
-		if (remainder)
-			credits_required++;
-	}
-
-	ath10k_dbg(ATH10K_DBG_HTC, "Credits required %d got %d\n",
-		   credits_required, ep->tx_credits);
-
-	if (ep->tx_credits < credits_required) {
-		__skb_queue_head(&ep->tx_queue, skb);
-		return NULL;
-	}
-
-	ep->tx_credits -= credits_required;
-	*credits = credits_required;
-	return skb;
-}
-
-static void ath10k_htc_send_work(struct work_struct *work)
-{
-	struct ath10k_htc_ep *ep = container_of(work,
-					struct ath10k_htc_ep, send_work);
-	struct ath10k_htc *htc = ep->htc;
-	struct sk_buff *skb;
-	u8 credits = 0;
-	int ret;
-
-	while (true) {
-		if (ep->ul_is_polled)
-			ath10k_htc_send_complete_check(ep, 0);
-
-		spin_lock_bh(&htc->tx_lock);
-		if (ep->tx_credit_flow_enabled)
-			skb = ath10k_htc_get_skb_credit_based(htc, ep,
-							      &credits);
-		else
-			skb = __skb_dequeue(&ep->tx_queue);
-		spin_unlock_bh(&htc->tx_lock);
-
-		if (!skb)
-			break;
-
-		ret = ath10k_htc_issue_skb(htc, ep, skb, credits);
-		if (ret == -ENOSR)
-			break;
-	}
-}
-
 int ath10k_htc_send(struct ath10k_htc *htc,
 		    enum ath10k_htc_ep_id eid,
 		    struct sk_buff *skb)
 {
 	struct ath10k_htc_ep *ep = &htc->endpoint[eid];
+	int credits = 0;
+	int ret;
 
 	if (htc->ar->state == ATH10K_STATE_WEDGED)
 		return -ECOMM;
@@ -254,18 +133,55 @@ int ath10k_htc_send(struct ath10k_htc *htc,
 		return -ENOENT;
 	}
 
+	/* FIXME: This looks ugly, can we fix it? */
 	spin_lock_bh(&htc->tx_lock);
 	if (htc->stopped) {
 		spin_unlock_bh(&htc->tx_lock);
 		return -ESHUTDOWN;
 	}
+	spin_unlock_bh(&htc->tx_lock);
 
-	__skb_queue_tail(&ep->tx_queue, skb);
 	skb_push(skb, sizeof(struct ath10k_htc_hdr));
-	spin_unlock_bh(&htc->tx_lock);
 
-	queue_work(htc->ar->workqueue, &ep->send_work);
+	if (ep->tx_credit_flow_enabled) {
+		credits = DIV_ROUND_UP(skb->len, htc->target_credit_size);
+		spin_lock_bh(&htc->tx_lock);
+		if (ep->tx_credits < credits) {
+			spin_unlock_bh(&htc->tx_lock);
+			ret = -EAGAIN;
+			goto err_pull;
+		}
+		ep->tx_credits -= credits;
+		spin_unlock_bh(&htc->tx_lock);
+	}
+
+	ath10k_htc_prepare_tx_skb(ep, skb);
+
+	ret = ath10k_skb_map(htc->ar->dev, skb);
+	if (ret)
+		goto err_credits;
+
+	ret = ath10k_hif_send_head(htc->ar, ep->ul_pipe_id, ep->eid,
+				   skb->len, skb);
+	if (ret)
+		goto err_unmap;
+
 	return 0;
+
+err_unmap:
+	ath10k_skb_unmap(htc->ar->dev, skb);
+err_credits:
+	if (ep->tx_credit_flow_enabled) {
+		spin_lock_bh(&htc->tx_lock);
+		ep->tx_credits += credits;
+		spin_unlock_bh(&htc->tx_lock);
+
+		if (ep->ep_ops.ep_tx_credits)
+			ep->ep_ops.ep_tx_credits(htc->ar);
+	}
+err_pull:
+	skb_pull(skb, sizeof(struct ath10k_htc_hdr));
+	return ret;
 }
 
 static int ath10k_htc_tx_completion_handler(struct ath10k *ar,
@@ -278,39 +194,9 @@ static int ath10k_htc_tx_completion_handler(struct ath10k *ar,
 	ath10k_htc_notify_tx_completion(ep, skb);
 	/* the skb now belongs to the completion handler */
 
-	/* note: when using TX credit flow, the re-checking of queues happens
-	 * when credits flow back from the target.  in the non-TX credit case,
-	 * we recheck after the packet completes */
-	spin_lock_bh(&htc->tx_lock);
-	if (!ep->tx_credit_flow_enabled && !htc->stopped)
-		queue_work(ar->workqueue, &ep->send_work);
-	spin_unlock_bh(&htc->tx_lock);
-
 	return 0;
 }
 
-/* flush endpoint TX queue */
-static void ath10k_htc_flush_endpoint_tx(struct ath10k_htc *htc,
-					 struct ath10k_htc_ep *ep)
-{
-	struct sk_buff *skb;
-	struct ath10k_skb_cb *skb_cb;
-
-	spin_lock_bh(&htc->tx_lock);
-	for (;;) {
-		skb = __skb_dequeue(&ep->tx_queue);
-		if (!skb)
-			break;
-
-		skb_cb = ATH10K_SKB_CB(skb);
-		skb_cb->is_aborted = true;
-		ath10k_htc_notify_tx_completion(ep, skb);
-	}
-	spin_unlock_bh(&htc->tx_lock);
-
-	cancel_work_sync(&ep->send_work);
-}
-
 /***********/
 /* Receive */
 /***********/
@@ -340,8 +226,11 @@ ath10k_htc_process_credit_report(struct ath10k_htc *htc,
 		ep = &htc->endpoint[report->eid];
 		ep->tx_credits += report->credits;
 
-		if (ep->tx_credits && !skb_queue_empty(&ep->tx_queue))
-			queue_work(htc->ar->workqueue, &ep->send_work);
+		if (ep->ep_ops.ep_tx_credits) {
+			spin_unlock_bh(&htc->tx_lock);
+			ep->ep_ops.ep_tx_credits(htc->ar);
+			spin_lock_bh(&htc->tx_lock);
+		}
 	}
 	spin_unlock_bh(&htc->tx_lock);
 }
@@ -599,10 +488,8 @@ static void ath10k_htc_reset_endpoint_states(struct ath10k_htc *htc)
 		ep->max_ep_message_len = 0;
 		ep->max_tx_queue_depth = 0;
 		ep->eid = i;
-		skb_queue_head_init(&ep->tx_queue);
 		ep->htc = htc;
 		ep->tx_credit_flow_enabled = true;
-		INIT_WORK(&ep->send_work, ath10k_htc_send_work);
 	}
 }
 
@@ -752,8 +639,8 @@ int ath10k_htc_connect_service(struct ath10k_htc *htc,
 	tx_alloc = ath10k_htc_get_credit_allocation(htc,
 						    conn_req->service_id);
 	if (!tx_alloc)
-		ath10k_dbg(ATH10K_DBG_HTC,
-			   "HTC Service %s does not allocate target credits\n",
+		ath10k_dbg(ATH10K_DBG_BOOT,
+			   "boot htc service %s does not allocate target credits\n",
 			   htc_service_name(conn_req->service_id));
 
 	skb = ath10k_htc_build_tx_ctrl_skb(htc->ar);
@@ -873,19 +760,19 @@ setup:
 	if (status)
 		return status;
 
-	ath10k_dbg(ATH10K_DBG_HTC,
-		   "HTC service: %s UL pipe: %d DL pipe: %d eid: %d ready\n",
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot htc service '%s' ul pipe %d dl pipe %d eid %d ready\n",
 		   htc_service_name(ep->service_id), ep->ul_pipe_id,
 		   ep->dl_pipe_id, ep->eid);
 
-	ath10k_dbg(ATH10K_DBG_HTC,
-		   "EP %d UL polled: %d, DL polled: %d\n",
+	ath10k_dbg(ATH10K_DBG_BOOT,
+		   "boot htc ep %d ul polled %d dl polled %d\n",
 		   ep->eid, ep->ul_is_polled, ep->dl_is_polled);
 
 	if (disable_credit_flow_ctrl && ep->tx_credit_flow_enabled) {
 		ep->tx_credit_flow_enabled = false;
-		ath10k_dbg(ATH10K_DBG_HTC,
-			   "HTC service: %s eid: %d TX flow control disabled\n",
+		ath10k_dbg(ATH10K_DBG_BOOT,
+			   "boot htc service '%s' eid %d TX flow control disabled\n",
 			   htc_service_name(ep->service_id), assigned_eid);
 	}
 
@@ -945,18 +832,10 @@ int ath10k_htc_start(struct ath10k_htc *htc)
  */
 void ath10k_htc_stop(struct ath10k_htc *htc)
 {
-	int i;
-	struct ath10k_htc_ep *ep;
-
 	spin_lock_bh(&htc->tx_lock);
 	htc->stopped = true;
 	spin_unlock_bh(&htc->tx_lock);
 
-	for (i = ATH10K_HTC_EP_0; i < ATH10K_HTC_EP_COUNT; i++) {
-		ep = &htc->endpoint[i];
-		ath10k_htc_flush_endpoint_tx(htc, ep);
-	}
-
 	ath10k_hif_stop(htc->ar);
 }
 
diff --git a/drivers/net/wireless/ath/ath10k/htc.h b/drivers/net/wireless/ath/ath10k/htc.h
index e1dd8c761853..4716d331e6b6 100644
--- a/drivers/net/wireless/ath/ath10k/htc.h
+++ b/drivers/net/wireless/ath/ath10k/htc.h
@@ -276,6 +276,7 @@ struct ath10k_htc_ops {
 struct ath10k_htc_ep_ops {
 	void (*ep_tx_complete)(struct ath10k *, struct sk_buff *);
 	void (*ep_rx_complete)(struct ath10k *, struct sk_buff *);
+	void (*ep_tx_credits)(struct ath10k *);
 };
 
 /* service connection information */
@@ -315,15 +316,11 @@ struct ath10k_htc_ep {
 	int ul_is_polled; /* call HIF to get tx completions */
 	int dl_is_polled; /* call HIF to fetch rx (not implemented) */
 
-	struct sk_buff_head tx_queue;
-
 	u8 seq_no; /* for debugging */
 	int tx_credits;
 	int tx_credit_size;
 	int tx_credits_per_max_message;
 	bool tx_credit_flow_enabled;
-
-	struct work_struct send_work;
 };
 
 struct ath10k_htc_svc_tx_credits {
diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index 26c78a907f97..1a337e93b7e9 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -19,6 +19,7 @@
 #define _HTT_H_
 
 #include <linux/bug.h>
+#include <linux/interrupt.h>
 
 #include "htc.h"
 #include "rx_desc.h"
@@ -1268,6 +1269,7 @@ struct ath10k_htt {
 	/* set if host-fw communication goes haywire
 	 * used to avoid further failures */
 	bool rx_confused;
+	struct tasklet_struct rx_replenish_task;
 };
 
 #define RX_HTT_HDR_STATUS_LEN 64
@@ -1308,6 +1310,10 @@ struct htt_rx_desc {
 #define HTT_RX_BUF_SIZE 1920
 #define HTT_RX_MSDU_SIZE (HTT_RX_BUF_SIZE - (int)sizeof(struct htt_rx_desc))
 
+/* Refill a bunch of RX buffers for each refill round so that FW/HW can handle
+ * aggregated traffic more nicely. */
+#define ATH10K_HTT_MAX_NUM_REFILL 16
+
 /*
  * DMA_MAP expects the buffer to be an integral number of cache lines.
  * Rather than checking the actual cache line size, this code makes a
@@ -1327,6 +1333,7 @@ void ath10k_htt_rx_detach(struct ath10k_htt *htt);
 void ath10k_htt_htc_tx_complete(struct ath10k *ar, struct sk_buff *skb);
 void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb);
 int ath10k_htt_h2t_ver_req_msg(struct ath10k_htt *htt);
+int ath10k_htt_h2t_stats_req(struct ath10k_htt *htt, u8 mask, u64 cookie);
 int ath10k_htt_send_rx_ring_cfg_ll(struct ath10k_htt *htt);
 
 void __ath10k_htt_tx_dec_pending(struct ath10k_htt *htt);
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 9bb0ae89fdba..90d4f74c28d7 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -20,6 +20,7 @@
 #include "htt.h"
 #include "txrx.h"
 #include "debug.h"
+#include "trace.h"
 
 #include <linux/log2.h>
 
@@ -40,6 +41,10 @@
 /* when under memory pressure rx ring refill may fail and needs a retry */
 #define HTT_RX_RING_REFILL_RETRY_MS 50
 
+
+static int ath10k_htt_rx_get_csum_state(struct sk_buff *skb);
+
+
 static int ath10k_htt_rx_ring_size(struct ath10k_htt *htt)
 {
 	int size;
@@ -177,10 +182,27 @@ static int ath10k_htt_rx_ring_fill_n(struct ath10k_htt *htt, int num)
 
 static void ath10k_htt_rx_msdu_buff_replenish(struct ath10k_htt *htt)
 {
-	int ret, num_to_fill;
+	int ret, num_deficit, num_to_fill;
 
+	/* Refilling the whole RX ring buffer proves to be a bad idea. The
+	 * reason is RX may take up significant amount of CPU cycles and starve
+	 * other tasks, e.g. TX on an ethernet device while acting as a bridge
+	 * with ath10k wlan interface. This ended up with very poor performance
+	 * once CPU the host system was overwhelmed with RX on ath10k.
+	 *
+	 * By limiting the number of refills the replenishing occurs
+	 * progressively. This in turns makes use of the fact tasklets are
+	 * processed in FIFO order. This means actual RX processing can starve
+	 * out refilling. If there's not enough buffers on RX ring FW will not
+	 * report RX until it is refilled with enough buffers. This
+	 * automatically balances load wrt to CPU power.
+	 *
+	 * This probably comes at a cost of lower maximum throughput but
+	 * improves the avarage and stability. */
 	spin_lock_bh(&htt->rx_ring.lock);
-	num_to_fill = htt->rx_ring.fill_level - htt->rx_ring.fill_cnt;
+	num_deficit = htt->rx_ring.fill_level - htt->rx_ring.fill_cnt;
+	num_to_fill = min(ATH10K_HTT_MAX_NUM_REFILL, num_deficit);
+	num_deficit -= num_to_fill;
 	ret = ath10k_htt_rx_ring_fill_n(htt, num_to_fill);
 	if (ret == -ENOMEM) {
 		/*
@@ -191,6 +213,8 @@ static void ath10k_htt_rx_msdu_buff_replenish(struct ath10k_htt *htt)
 		 */
 		mod_timer(&htt->rx_ring.refill_retry_timer, jiffies +
 			  msecs_to_jiffies(HTT_RX_RING_REFILL_RETRY_MS));
+	} else if (num_deficit > 0) {
+		tasklet_schedule(&htt->rx_replenish_task);
 	}
 	spin_unlock_bh(&htt->rx_ring.lock);
 }
@@ -212,6 +236,7 @@ void ath10k_htt_rx_detach(struct ath10k_htt *htt)
 	int sw_rd_idx = htt->rx_ring.sw_rd_idx.msdu_payld;
 
 	del_timer_sync(&htt->rx_ring.refill_retry_timer);
+	tasklet_kill(&htt->rx_replenish_task);
 
 	while (sw_rd_idx != __le32_to_cpu(*(htt->rx_ring.alloc_idx.vaddr))) {
 		struct sk_buff *skb =
@@ -441,6 +466,12 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt,
 	return msdu_chaining;
 }
 
+static void ath10k_htt_rx_replenish_task(unsigned long ptr)
+{
+	struct ath10k_htt *htt = (struct ath10k_htt *)ptr;
+	ath10k_htt_rx_msdu_buff_replenish(htt);
+}
+
 int ath10k_htt_rx_attach(struct ath10k_htt *htt)
 {
 	dma_addr_t paddr;
@@ -501,7 +532,10 @@ int ath10k_htt_rx_attach(struct ath10k_htt *htt)
 	if (__ath10k_htt_rx_ring_fill_n(htt, htt->rx_ring.fill_level))
 		goto err_fill_ring;
 
-	ath10k_dbg(ATH10K_DBG_HTT, "HTT RX ring size: %d, fill_level: %d\n",
+	tasklet_init(&htt->rx_replenish_task, ath10k_htt_rx_replenish_task,
+		     (unsigned long)htt);
+
+	ath10k_dbg(ATH10K_DBG_BOOT, "htt rx ring size %d fill_level %d\n",
 		   htt->rx_ring.size, htt->rx_ring.fill_level);
 	return 0;
 
@@ -590,142 +624,144 @@ static bool ath10k_htt_rx_hdr_is_amsdu(struct ieee80211_hdr *hdr)
 	return false;
 }
 
-static int ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
-			struct htt_rx_info *info)
+struct rfc1042_hdr {
+	u8 llc_dsap;
+	u8 llc_ssap;
+	u8 llc_ctrl;
+	u8 snap_oui[3];
+	__be16 snap_type;
+} __packed;
+
+struct amsdu_subframe_hdr {
+	u8 dst[ETH_ALEN];
+	u8 src[ETH_ALEN];
+	__be16 len;
+} __packed;
+
+static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
+				struct htt_rx_info *info)
 {
 	struct htt_rx_desc *rxd;
-	struct sk_buff *amsdu;
 	struct sk_buff *first;
-	struct ieee80211_hdr *hdr;
 	struct sk_buff *skb = info->skb;
 	enum rx_msdu_decap_format fmt;
 	enum htt_rx_mpdu_encrypt_type enctype;
+	struct ieee80211_hdr *hdr;
+	u8 hdr_buf[64], addr[ETH_ALEN], *qos;
 	unsigned int hdr_len;
-	int crypto_len;
 
 	rxd = (void *)skb->data - sizeof(*rxd);
-	fmt = MS(__le32_to_cpu(rxd->msdu_start.info1),
-			RX_MSDU_START_INFO1_DECAP_FORMAT);
 	enctype = MS(__le32_to_cpu(rxd->mpdu_start.info0),
 			RX_MPDU_START_INFO0_ENCRYPT_TYPE);
 
-	/* FIXME: No idea what assumptions are safe here. Need logs */
-	if ((fmt == RX_MSDU_DECAP_RAW && skb->next)) {
-		ath10k_htt_rx_free_msdu_chain(skb->next);
-		skb->next = NULL;
-		return -ENOTSUPP;
-	}
+	hdr = (struct ieee80211_hdr *)rxd->rx_hdr_status;
+	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+	memcpy(hdr_buf, hdr, hdr_len);
+	hdr = (struct ieee80211_hdr *)hdr_buf;
 
-	/* A-MSDU max is a little less than 8K */
-	amsdu = dev_alloc_skb(8*1024);
-	if (!amsdu) {
-		ath10k_warn("A-MSDU allocation failed\n");
-		ath10k_htt_rx_free_msdu_chain(skb->next);
-		skb->next = NULL;
-		return -ENOMEM;
-	}
-
-	if (fmt >= RX_MSDU_DECAP_NATIVE_WIFI) {
-		int hdrlen;
-
-		hdr = (void *)rxd->rx_hdr_status;
-		hdrlen = ieee80211_hdrlen(hdr->frame_control);
-		memcpy(skb_put(amsdu, hdrlen), hdr, hdrlen);
-	}
+	/* FIXME: Hopefully this is a temporary measure.
+	 *
+	 * Reporting individual A-MSDU subframes means each reported frame
+	 * shares the same sequence number.
+	 *
+	 * mac80211 drops frames it recognizes as duplicates, i.e.
+	 * retransmission flag is set and sequence number matches sequence
+	 * number from a previous frame (as per IEEE 802.11-2012: 9.3.2.10
+	 * "Duplicate detection and recovery")
+	 *
+	 * To avoid frames being dropped clear retransmission flag for all
+	 * received A-MSDUs.
+	 *
+	 * Worst case: actual duplicate frames will be reported but this should
+	 * still be handled gracefully by other OSI/ISO layers. */
+	hdr->frame_control &= cpu_to_le16(~IEEE80211_FCTL_RETRY);
 
 	first = skb;
 	while (skb) {
 		void *decap_hdr;
-		int decap_len = 0;
+		int len;
 
 		rxd = (void *)skb->data - sizeof(*rxd);
 		fmt = MS(__le32_to_cpu(rxd->msdu_start.info1),
-				RX_MSDU_START_INFO1_DECAP_FORMAT);
+			 RX_MSDU_START_INFO1_DECAP_FORMAT);
 		decap_hdr = (void *)rxd->rx_hdr_status;
 
-		if (skb == first) {
-			/* We receive linked A-MSDU subframe skbuffs. The
-			 * first one contains the original 802.11 header (and
-			 * possible crypto param) in the RX descriptor. The
-			 * A-MSDU subframe header follows that. Each part is
-			 * aligned to 4 byte boundary. */
-
-			hdr = (void *)amsdu->data;
-			hdr_len = ieee80211_hdrlen(hdr->frame_control);
-			crypto_len = ath10k_htt_rx_crypto_param_len(enctype);
-
-			decap_hdr += roundup(hdr_len, 4);
-			decap_hdr += roundup(crypto_len, 4);
-		}
+		skb->ip_summed = ath10k_htt_rx_get_csum_state(skb);
 
-		/* When fmt == RX_MSDU_DECAP_8023_SNAP_LLC:
-		 *
-		 * SNAP 802.3 consists of:
-		 * [dst:6][src:6][len:2][dsap:1][ssap:1][ctl:1][snap:5]
-		 * [data][fcs:4].
-		 *
-		 * Since this overlaps with A-MSDU header (da, sa, len)
-		 * there's nothing extra to do. */
-
-		if (fmt == RX_MSDU_DECAP_ETHERNET2_DIX) {
-			/* Ethernet2 decap inserts ethernet header in place of
-			 * A-MSDU subframe header. */
-			skb_pull(skb, 6 + 6 + 2);
-
-			/* A-MSDU subframe header length */
-			decap_len += 6 + 6 + 2;
-
-			/* Ethernet2 decap also strips the LLC/SNAP so we need
-			 * to re-insert it. The LLC/SNAP follows A-MSDU
-			 * subframe header. */
-			/* FIXME: Not all LLCs are 8 bytes long */
-			decap_len += 8;
-
-			memcpy(skb_put(amsdu, decap_len), decap_hdr, decap_len);
+		/* First frame in an A-MSDU chain has more decapped data. */
+		if (skb == first) {
+			len = round_up(ieee80211_hdrlen(hdr->frame_control), 4);
+			len += round_up(ath10k_htt_rx_crypto_param_len(enctype),
+					4);
+			decap_hdr += len;
 		}
 
-		if (fmt == RX_MSDU_DECAP_NATIVE_WIFI) {
-			/* Native Wifi decap inserts regular 802.11 header
-			 * in place of A-MSDU subframe header. */
+		switch (fmt) {
+		case RX_MSDU_DECAP_RAW:
+			/* remove trailing FCS */
+			skb_trim(skb, skb->len - FCS_LEN);
+			break;
+		case RX_MSDU_DECAP_NATIVE_WIFI:
+			/* pull decapped header and copy DA */
 			hdr = (struct ieee80211_hdr *)skb->data;
-			skb_pull(skb, ieee80211_hdrlen(hdr->frame_control));
+			hdr_len = ieee80211_hdrlen(hdr->frame_control);
+			memcpy(addr, ieee80211_get_DA(hdr), ETH_ALEN);
+			skb_pull(skb, hdr_len);
 
-			/* A-MSDU subframe header length */
-			decap_len += 6 + 6 + 2;
+			/* push original 802.11 header */
+			hdr = (struct ieee80211_hdr *)hdr_buf;
+			hdr_len = ieee80211_hdrlen(hdr->frame_control);
+			memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
 
-			memcpy(skb_put(amsdu, decap_len), decap_hdr, decap_len);
-		}
+			/* original A-MSDU header has the bit set but we're
+			 * not including A-MSDU subframe header */
+			hdr = (struct ieee80211_hdr *)skb->data;
+			qos = ieee80211_get_qos_ctl(hdr);
+			qos[0] &= ~IEEE80211_QOS_CTL_A_MSDU_PRESENT;
 
-		if (fmt == RX_MSDU_DECAP_RAW)
-			skb_trim(skb, skb->len - 4); /* remove FCS */
+			/* original 802.11 header has a different DA */
+			memcpy(ieee80211_get_DA(hdr), addr, ETH_ALEN);
+			break;
+		case RX_MSDU_DECAP_ETHERNET2_DIX:
+			/* strip ethernet header and insert decapped 802.11
+			 * header, amsdu subframe header and rfc1042 header */
 
-		memcpy(skb_put(amsdu, skb->len), skb->data, skb->len);
+			len = 0;
+			len += sizeof(struct rfc1042_hdr);
+			len += sizeof(struct amsdu_subframe_hdr);
 
-		/* A-MSDU subframes are padded to 4bytes
-		 * but relative to first subframe, not the whole MPDU */
-		if (skb->next && ((decap_len + skb->len) & 3)) {
-			int padlen = 4 - ((decap_len + skb->len) & 3);
-			memset(skb_put(amsdu, padlen), 0, padlen);
+			skb_pull(skb, sizeof(struct ethhdr));
+			memcpy(skb_push(skb, len), decap_hdr, len);
+			memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
+			break;
+		case RX_MSDU_DECAP_8023_SNAP_LLC:
+			/* insert decapped 802.11 header making a singly
+			 * A-MSDU */
+			memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
+			break;
 		}
 
+		info->skb = skb;
+		info->encrypt_type = enctype;
 		skb = skb->next;
-	}
-
-	info->skb = amsdu;
-	info->encrypt_type = enctype;
+		info->skb->next = NULL;
 
-	ath10k_htt_rx_free_msdu_chain(first);
+		ath10k_process_rx(htt->ar, info);
+	}
 
-	return 0;
+	/* FIXME: It might be nice to re-assemble the A-MSDU when there's a
+	 * monitor interface active for sniffing purposes. */
 }
 
-static int ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
+static void ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 {
 	struct sk_buff *skb = info->skb;
 	struct htt_rx_desc *rxd;
 	struct ieee80211_hdr *hdr;
 	enum rx_msdu_decap_format fmt;
 	enum htt_rx_mpdu_encrypt_type enctype;
+	int hdr_len;
+	void *rfc1042;
 
 	/* This shouldn't happen. If it does than it may be a FW bug. */
 	if (skb->next) {
@@ -739,49 +775,53 @@ static int ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 			RX_MSDU_START_INFO1_DECAP_FORMAT);
 	enctype = MS(__le32_to_cpu(rxd->mpdu_start.info0),
 			RX_MPDU_START_INFO0_ENCRYPT_TYPE);
-	hdr = (void *)skb->data - RX_HTT_HDR_STATUS_LEN;
+	hdr = (struct ieee80211_hdr *)rxd->rx_hdr_status;
+	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+
+	skb->ip_summed = ath10k_htt_rx_get_csum_state(skb);
 
 	switch (fmt) {
 	case RX_MSDU_DECAP_RAW:
 		/* remove trailing FCS */
-		skb_trim(skb, skb->len - 4);
+		skb_trim(skb, skb->len - FCS_LEN);
 		break;
 	case RX_MSDU_DECAP_NATIVE_WIFI:
-		/* nothing to do here */
+		/* Pull decapped header */
+		hdr = (struct ieee80211_hdr *)skb->data;
+		hdr_len = ieee80211_hdrlen(hdr->frame_control);
+		skb_pull(skb, hdr_len);
+
+		/* Push original header */
+		hdr = (struct ieee80211_hdr *)rxd->rx_hdr_status;
+		hdr_len = ieee80211_hdrlen(hdr->frame_control);
+		memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
 		break;
 	case RX_MSDU_DECAP_ETHERNET2_DIX:
-		/* macaddr[6] + macaddr[6] + ethertype[2] */
-		skb_pull(skb, 6 + 6 + 2);
-		break;
-	case RX_MSDU_DECAP_8023_SNAP_LLC:
-		/* macaddr[6] + macaddr[6] + len[2] */
-		/* we don't need this for non-A-MSDU */
-		skb_pull(skb, 6 + 6 + 2);
-		break;
-	}
-
-	if (fmt == RX_MSDU_DECAP_ETHERNET2_DIX) {
-		void *llc;
-		int llclen;
+		/* strip ethernet header and insert decapped 802.11 header and
+		 * rfc1042 header */
 
-		llclen = 8;
-		llc  = hdr;
-		llc += roundup(ieee80211_hdrlen(hdr->frame_control), 4);
-		llc += roundup(ath10k_htt_rx_crypto_param_len(enctype), 4);
+		rfc1042 = hdr;
+		rfc1042 += roundup(hdr_len, 4);
+		rfc1042 += roundup(ath10k_htt_rx_crypto_param_len(enctype), 4);
 
-		skb_push(skb, llclen);
-		memcpy(skb->data, llc, llclen);
-	}
+		skb_pull(skb, sizeof(struct ethhdr));
+		memcpy(skb_push(skb, sizeof(struct rfc1042_hdr)),
+		       rfc1042, sizeof(struct rfc1042_hdr));
+		memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
+		break;
+	case RX_MSDU_DECAP_8023_SNAP_LLC:
+		/* remove A-MSDU subframe header and insert
+		 * decapped 802.11 header. rfc1042 header is already there */
 
-	if (fmt >= RX_MSDU_DECAP_ETHERNET2_DIX) {
-		int len = ieee80211_hdrlen(hdr->frame_control);
-		skb_push(skb, len);
-		memcpy(skb->data, hdr, len);
+		skb_pull(skb, sizeof(struct amsdu_subframe_hdr));
+		memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
+		break;
 	}
 
 	info->skb = skb;
 	info->encrypt_type = enctype;
-	return 0;
+
+	ath10k_process_rx(htt->ar, info);
 }
 
 static bool ath10k_htt_rx_has_decrypt_err(struct sk_buff *skb)
@@ -853,8 +893,6 @@ static void ath10k_htt_rx_handler(struct ath10k_htt *htt,
 	int fw_desc_len;
 	u8 *fw_desc;
 	int i, j;
-	int ret;
-	int ip_summed;
 
 	memset(&info, 0, sizeof(info));
 
@@ -929,11 +967,6 @@ static void ath10k_htt_rx_handler(struct ath10k_htt *htt,
 				continue;
 			}
 
-			/* The skb is not yet processed and it may be
-			 * reallocated. Since the offload is in the original
-			 * skb extract the checksum now and assign it later */
-			ip_summed = ath10k_htt_rx_get_csum_state(msdu_head);
-
 			info.skb     = msdu_head;
 			info.fcs_err = ath10k_htt_rx_has_fcs_err(msdu_head);
 			info.signal  = ATH10K_DEFAULT_NOISE_FLOOR;
@@ -946,28 +979,13 @@ static void ath10k_htt_rx_handler(struct ath10k_htt *htt,
 			hdr = ath10k_htt_rx_skb_get_hdr(msdu_head);
 
 			if (ath10k_htt_rx_hdr_is_amsdu(hdr))
-				ret = ath10k_htt_rx_amsdu(htt, &info);
+				ath10k_htt_rx_amsdu(htt, &info);
 			else
-				ret = ath10k_htt_rx_msdu(htt, &info);
-
-			if (ret && !info.fcs_err) {
-				ath10k_warn("error processing msdus %d\n", ret);
-				dev_kfree_skb_any(info.skb);
-				continue;
-			}
-
-			if (ath10k_htt_rx_hdr_is_amsdu((void *)info.skb->data))
-				ath10k_dbg(ATH10K_DBG_HTT, "htt mpdu is amsdu\n");
-
-			info.skb->ip_summed = ip_summed;
-
-			ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "htt mpdu: ",
-					info.skb->data, info.skb->len);
-			ath10k_process_rx(htt->ar, &info);
+				ath10k_htt_rx_msdu(htt, &info);
 		}
 	}
 
-	ath10k_htt_rx_msdu_buff_replenish(htt);
+	tasklet_schedule(&htt->rx_replenish_task);
 }
 
 static void ath10k_htt_rx_frag_handler(struct ath10k_htt *htt,
@@ -1139,7 +1157,7 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 			break;
 		}
 
-		ath10k_txrx_tx_completed(htt, &tx_done);
+		ath10k_txrx_tx_unref(htt, &tx_done);
 		break;
 	}
 	case HTT_T2H_MSG_TYPE_TX_COMPL_IND: {
@@ -1173,7 +1191,7 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		for (i = 0; i < resp->data_tx_completion.num_msdus; i++) {
 			msdu_id = resp->data_tx_completion.msdus[i];
 			tx_done.msdu_id = __le16_to_cpu(msdu_id);
-			ath10k_txrx_tx_completed(htt, &tx_done);
+			ath10k_txrx_tx_unref(htt, &tx_done);
 		}
 		break;
 	}
@@ -1198,8 +1216,10 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 	case HTT_T2H_MSG_TYPE_TEST:
 		/* FIX THIS */
 		break;
-	case HTT_T2H_MSG_TYPE_TX_INSPECT_IND:
 	case HTT_T2H_MSG_TYPE_STATS_CONF:
+		trace_ath10k_htt_stats(skb->data, skb->len);
+		break;
+	case HTT_T2H_MSG_TYPE_TX_INSPECT_IND:
 	case HTT_T2H_MSG_TYPE_RX_ADDBA:
 	case HTT_T2H_MSG_TYPE_RX_DELBA:
 	case HTT_T2H_MSG_TYPE_RX_FLUSH:
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index d4fb3875c918..3b93c6a01c6c 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -96,7 +96,7 @@ int ath10k_htt_tx_attach(struct ath10k_htt *htt)
 	htt->max_num_pending_tx = ath10k_hif_get_free_queue_number(htt->ar,
 								   pipe);
 
-	ath10k_dbg(ATH10K_DBG_HTT, "htt tx max num pending tx %d\n",
+	ath10k_dbg(ATH10K_DBG_BOOT, "htt tx max num pending tx %d\n",
 		   htt->max_num_pending_tx);
 
 	htt->pending_tx = kzalloc(sizeof(*htt->pending_tx) *
@@ -117,7 +117,7 @@ int ath10k_htt_tx_attach(struct ath10k_htt *htt)
 
 static void ath10k_htt_tx_cleanup_pending(struct ath10k_htt *htt)
 {
-	struct sk_buff *txdesc;
+	struct htt_tx_done tx_done = {0};
 	int msdu_id;
 
 	/* No locks needed. Called after communication with the device has
@@ -127,18 +127,13 @@ static void ath10k_htt_tx_cleanup_pending(struct ath10k_htt *htt)
 		if (!test_bit(msdu_id, htt->used_msdu_ids))
 			continue;
 
-		txdesc = htt->pending_tx[msdu_id];
-		if (!txdesc)
-			continue;
-
 		ath10k_dbg(ATH10K_DBG_HTT, "force cleanup msdu_id %hu\n",
 			   msdu_id);
 
-		if (ATH10K_SKB_CB(txdesc)->htt.refcount > 0)
-			ATH10K_SKB_CB(txdesc)->htt.refcount = 1;
+		tx_done.discard = 1;
+		tx_done.msdu_id = msdu_id;
 
-		ATH10K_SKB_CB(txdesc)->htt.discard = true;
-		ath10k_txrx_tx_unref(htt, txdesc);
+		ath10k_txrx_tx_unref(htt, &tx_done);
 	}
 }
 
@@ -152,26 +147,7 @@ void ath10k_htt_tx_detach(struct ath10k_htt *htt)
 
 void ath10k_htt_htc_tx_complete(struct ath10k *ar, struct sk_buff *skb)
 {
-	struct ath10k_skb_cb *skb_cb = ATH10K_SKB_CB(skb);
-	struct ath10k_htt *htt = &ar->htt;
-
-	if (skb_cb->htt.is_conf) {
-		dev_kfree_skb_any(skb);
-		return;
-	}
-
-	if (skb_cb->is_aborted) {
-		skb_cb->htt.discard = true;
-
-		/* if the skbuff is aborted we need to make sure we'll free up
-		 * the tx resources, we can't simply run tx_unref() 2 times
-		 * because if htt tx completion came in earlier we'd access
-		 * unallocated memory */
-		if (skb_cb->htt.refcount > 1)
-			skb_cb->htt.refcount = 1;
-	}
-
-	ath10k_txrx_tx_unref(htt, skb);
+	dev_kfree_skb_any(skb);
 }
 
 int ath10k_htt_h2t_ver_req_msg(struct ath10k_htt *htt)
@@ -192,10 +168,48 @@ int ath10k_htt_h2t_ver_req_msg(struct ath10k_htt *htt)
 	cmd = (struct htt_cmd *)skb->data;
 	cmd->hdr.msg_type = HTT_H2T_MSG_TYPE_VERSION_REQ;
 
-	ATH10K_SKB_CB(skb)->htt.is_conf = true;
+	ret = ath10k_htc_send(&htt->ar->htc, htt->eid, skb);
+	if (ret) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	return 0;
+}
+
+int ath10k_htt_h2t_stats_req(struct ath10k_htt *htt, u8 mask, u64 cookie)
+{
+	struct htt_stats_req *req;
+	struct sk_buff *skb;
+	struct htt_cmd *cmd;
+	int len = 0, ret;
+
+	len += sizeof(cmd->hdr);
+	len += sizeof(cmd->stats_req);
+
+	skb = ath10k_htc_alloc_skb(len);
+	if (!skb)
+		return -ENOMEM;
+
+	skb_put(skb, len);
+	cmd = (struct htt_cmd *)skb->data;
+	cmd->hdr.msg_type = HTT_H2T_MSG_TYPE_STATS_REQ;
+
+	req = &cmd->stats_req;
+
+	memset(req, 0, sizeof(*req));
+
+	/* currently we support only max 8 bit masks so no need to worry
+	 * about endian support */
+	req->upload_types[0] = mask;
+	req->reset_types[0] = mask;
+	req->stat_type = HTT_STATS_REQ_CFG_STAT_TYPE_INVALID;
+	req->cookie_lsb = cpu_to_le32(cookie & 0xffffffff);
+	req->cookie_msb = cpu_to_le32((cookie & 0xffffffff00000000ULL) >> 32);
 
 	ret = ath10k_htc_send(&htt->ar->htc, htt->eid, skb);
 	if (ret) {
+		ath10k_warn("failed to send htt type stats request: %d", ret);
 		dev_kfree_skb_any(skb);
 		return ret;
 	}
@@ -279,8 +293,6 @@ int ath10k_htt_send_rx_ring_cfg_ll(struct ath10k_htt *htt)
 
 #undef desc_offset
 
-	ATH10K_SKB_CB(skb)->htt.is_conf = true;
-
 	ret = ath10k_htc_send(&htt->ar->htc, htt->eid, skb);
 	if (ret) {
 		dev_kfree_skb_any(skb);
@@ -293,10 +305,10 @@ int ath10k_htt_send_rx_ring_cfg_ll(struct ath10k_htt *htt)
 int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 {
 	struct device *dev = htt->ar->dev;
-	struct ath10k_skb_cb *skb_cb;
 	struct sk_buff *txdesc = NULL;
 	struct htt_cmd *cmd;
-	u8 vdev_id = ATH10K_SKB_CB(msdu)->htt.vdev_id;
+	struct ath10k_skb_cb *skb_cb = ATH10K_SKB_CB(msdu);
+	u8 vdev_id = skb_cb->htt.vdev_id;
 	int len = 0;
 	int msdu_id = -1;
 	int res;
@@ -304,30 +316,30 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 	res = ath10k_htt_tx_inc_pending(htt);
 	if (res)
-		return res;
+		goto err;
 
 	len += sizeof(cmd->hdr);
 	len += sizeof(cmd->mgmt_tx);
 
-	txdesc = ath10k_htc_alloc_skb(len);
-	if (!txdesc) {
-		res = -ENOMEM;
-		goto err;
-	}
-
 	spin_lock_bh(&htt->tx_lock);
-	msdu_id = ath10k_htt_tx_alloc_msdu_id(htt);
-	if (msdu_id < 0) {
+	res = ath10k_htt_tx_alloc_msdu_id(htt);
+	if (res < 0) {
 		spin_unlock_bh(&htt->tx_lock);
-		res = msdu_id;
-		goto err;
+		goto err_tx_dec;
 	}
-	htt->pending_tx[msdu_id] = txdesc;
+	msdu_id = res;
+	htt->pending_tx[msdu_id] = msdu;
 	spin_unlock_bh(&htt->tx_lock);
 
+	txdesc = ath10k_htc_alloc_skb(len);
+	if (!txdesc) {
+		res = -ENOMEM;
+		goto err_free_msdu_id;
+	}
+
 	res = ath10k_skb_map(dev, msdu);
 	if (res)
-		goto err;
+		goto err_free_txdesc;
 
 	skb_put(txdesc, len);
 	cmd = (struct htt_cmd *)txdesc->data;
@@ -339,31 +351,27 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	memcpy(cmd->mgmt_tx.hdr, msdu->data,
 	       min_t(int, msdu->len, HTT_MGMT_FRM_HDR_DOWNLOAD_LEN));
 
-	/* refcount is decremented by HTC and HTT completions until it reaches
-	 * zero and is freed */
-	skb_cb = ATH10K_SKB_CB(txdesc);
-	skb_cb->htt.msdu_id = msdu_id;
-	skb_cb->htt.refcount = 2;
-	skb_cb->htt.msdu = msdu;
+	skb_cb->htt.frag_len = 0;
+	skb_cb->htt.pad_len = 0;
 
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
-		goto err;
+		goto err_unmap_msdu;
 
 	return 0;
 
-err:
+err_unmap_msdu:
 	ath10k_skb_unmap(dev, msdu);
-
-	if (txdesc)
-		dev_kfree_skb_any(txdesc);
-	if (msdu_id >= 0) {
-		spin_lock_bh(&htt->tx_lock);
-		htt->pending_tx[msdu_id] = NULL;
-		ath10k_htt_tx_free_msdu_id(htt, msdu_id);
-		spin_unlock_bh(&htt->tx_lock);
-	}
+err_free_txdesc:
+	dev_kfree_skb_any(txdesc);
+err_free_msdu_id:
+	spin_lock_bh(&htt->tx_lock);
+	htt->pending_tx[msdu_id] = NULL;
+	ath10k_htt_tx_free_msdu_id(htt, msdu_id);
+	spin_unlock_bh(&htt->tx_lock);
+err_tx_dec:
 	ath10k_htt_tx_dec_pending(htt);
+err:
 	return res;
 }
 
@@ -373,13 +381,12 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	struct htt_cmd *cmd;
 	struct htt_data_tx_desc_frag *tx_frags;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)msdu->data;
-	struct ath10k_skb_cb *skb_cb;
+	struct ath10k_skb_cb *skb_cb = ATH10K_SKB_CB(msdu);
 	struct sk_buff *txdesc = NULL;
-	struct sk_buff *txfrag = NULL;
+	bool use_frags;
 	u8 vdev_id = ATH10K_SKB_CB(msdu)->htt.vdev_id;
 	u8 tid;
-	int prefetch_len, desc_len, frag_len;
-	dma_addr_t frags_paddr;
+	int prefetch_len, desc_len;
 	int msdu_id = -1;
 	int res;
 	u8 flags0;
@@ -387,73 +394,73 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 	res = ath10k_htt_tx_inc_pending(htt);
 	if (res)
-		return res;
+		goto err;
+
+	spin_lock_bh(&htt->tx_lock);
+	res = ath10k_htt_tx_alloc_msdu_id(htt);
+	if (res < 0) {
+		spin_unlock_bh(&htt->tx_lock);
+		goto err_tx_dec;
+	}
+	msdu_id = res;
+	htt->pending_tx[msdu_id] = msdu;
+	spin_unlock_bh(&htt->tx_lock);
 
 	prefetch_len = min(htt->prefetch_len, msdu->len);
 	prefetch_len = roundup(prefetch_len, 4);
 
 	desc_len = sizeof(cmd->hdr) + sizeof(cmd->data_tx) + prefetch_len;
-	frag_len = sizeof(*tx_frags) * 2;
 
 	txdesc = ath10k_htc_alloc_skb(desc_len);
 	if (!txdesc) {
 		res = -ENOMEM;
-		goto err;
+		goto err_free_msdu_id;
 	}
 
 	/* Since HTT 3.0 there is no separate mgmt tx command. However in case
 	 * of mgmt tx using TX_FRM there is not tx fragment list. Instead of tx
 	 * fragment list host driver specifies directly frame pointer. */
-	if (htt->target_version_major < 3 ||
-	    !ieee80211_is_mgmt(hdr->frame_control)) {
-		txfrag = dev_alloc_skb(frag_len);
-		if (!txfrag) {
-			res = -ENOMEM;
-			goto err;
-		}
-	}
+	use_frags = htt->target_version_major < 3 ||
+		    !ieee80211_is_mgmt(hdr->frame_control);
 
 	if (!IS_ALIGNED((unsigned long)txdesc->data, 4)) {
 		ath10k_warn("htt alignment check failed. dropping packet.\n");
 		res = -EIO;
-		goto err;
+		goto err_free_txdesc;
 	}
 
-	spin_lock_bh(&htt->tx_lock);
-	msdu_id = ath10k_htt_tx_alloc_msdu_id(htt);
-	if (msdu_id < 0) {
-		spin_unlock_bh(&htt->tx_lock);
-		res = msdu_id;
-		goto err;
+	if (use_frags) {
+		skb_cb->htt.frag_len = sizeof(*tx_frags) * 2;
+		skb_cb->htt.pad_len = (unsigned long)msdu->data -
+				      round_down((unsigned long)msdu->data, 4);
+
+		skb_push(msdu, skb_cb->htt.frag_len + skb_cb->htt.pad_len);
+	} else {
+		skb_cb->htt.frag_len = 0;
+		skb_cb->htt.pad_len = 0;
 	}
-	htt->pending_tx[msdu_id] = txdesc;
-	spin_unlock_bh(&htt->tx_lock);
 
 	res = ath10k_skb_map(dev, msdu);
 	if (res)
-		goto err;
+		goto err_pull_txfrag;
+
+	if (use_frags) {
+		dma_sync_single_for_cpu(dev, skb_cb->paddr, msdu->len,
+					DMA_TO_DEVICE);
 
-	/* Since HTT 3.0 there is no separate mgmt tx command. However in case
-	 * of mgmt tx using TX_FRM there is not tx fragment list. Instead of tx
-	 * fragment list host driver specifies directly frame pointer. */
-	if (htt->target_version_major < 3 ||
-	    !ieee80211_is_mgmt(hdr->frame_control)) {
 		/* tx fragment list must be terminated with zero-entry */
-		skb_put(txfrag, frag_len);
-		tx_frags = (struct htt_data_tx_desc_frag *)txfrag->data;
-		tx_frags[0].paddr = __cpu_to_le32(ATH10K_SKB_CB(msdu)->paddr);
-		tx_frags[0].len   = __cpu_to_le32(msdu->len);
+		tx_frags = (struct htt_data_tx_desc_frag *)msdu->data;
+		tx_frags[0].paddr = __cpu_to_le32(skb_cb->paddr +
+						  skb_cb->htt.frag_len +
+						  skb_cb->htt.pad_len);
+		tx_frags[0].len   = __cpu_to_le32(msdu->len -
+						  skb_cb->htt.frag_len -
+						  skb_cb->htt.pad_len);
 		tx_frags[1].paddr = __cpu_to_le32(0);
 		tx_frags[1].len   = __cpu_to_le32(0);
 
-		res = ath10k_skb_map(dev, txfrag);
-		if (res)
-			goto err;
-
-		ath10k_dbg(ATH10K_DBG_HTT, "txfrag 0x%llx\n",
-			   (unsigned long long) ATH10K_SKB_CB(txfrag)->paddr);
-		ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "txfrag: ",
-				txfrag->data, frag_len);
+		dma_sync_single_for_device(dev, skb_cb->paddr, msdu->len,
+					   DMA_TO_DEVICE);
 	}
 
 	ath10k_dbg(ATH10K_DBG_HTT, "msdu 0x%llx\n",
@@ -463,7 +470,6 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 	skb_put(txdesc, desc_len);
 	cmd = (struct htt_cmd *)txdesc->data;
-	memset(cmd, 0, desc_len);
 
 	tid = ATH10K_SKB_CB(msdu)->htt.tid;
 
@@ -474,15 +480,11 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 		flags0 |= HTT_DATA_TX_DESC_FLAGS0_NO_ENCRYPT;
 	flags0 |= HTT_DATA_TX_DESC_FLAGS0_MAC_HDR_PRESENT;
 
-	/* Since HTT 3.0 there is no separate mgmt tx command. However in case
-	 * of mgmt tx using TX_FRM there is not tx fragment list. Instead of tx
-	 * fragment list host driver specifies directly frame pointer. */
-	if (htt->target_version_major >= 3 &&
-	    ieee80211_is_mgmt(hdr->frame_control))
-		flags0 |= SM(ATH10K_HW_TXRX_MGMT,
+	if (use_frags)
+		flags0 |= SM(ATH10K_HW_TXRX_NATIVE_WIFI,
 			     HTT_DATA_TX_DESC_FLAGS0_PKT_TYPE);
 	else
-		flags0 |= SM(ATH10K_HW_TXRX_NATIVE_WIFI,
+		flags0 |= SM(ATH10K_HW_TXRX_MGMT,
 			     HTT_DATA_TX_DESC_FLAGS0_PKT_TYPE);
 
 	flags1  = 0;
@@ -491,52 +493,37 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	flags1 |= HTT_DATA_TX_DESC_FLAGS1_CKSUM_L3_OFFLOAD;
 	flags1 |= HTT_DATA_TX_DESC_FLAGS1_CKSUM_L4_OFFLOAD;
 
-	/* Since HTT 3.0 there is no separate mgmt tx command. However in case
-	 * of mgmt tx using TX_FRM there is not tx fragment list. Instead of tx
-	 * fragment list host driver specifies directly frame pointer. */
-	if (htt->target_version_major >= 3 &&
-	    ieee80211_is_mgmt(hdr->frame_control))
-		frags_paddr = ATH10K_SKB_CB(msdu)->paddr;
-	else
-		frags_paddr = ATH10K_SKB_CB(txfrag)->paddr;
-
 	cmd->hdr.msg_type        = HTT_H2T_MSG_TYPE_TX_FRM;
 	cmd->data_tx.flags0      = flags0;
 	cmd->data_tx.flags1      = __cpu_to_le16(flags1);
-	cmd->data_tx.len         = __cpu_to_le16(msdu->len);
+	cmd->data_tx.len         = __cpu_to_le16(msdu->len -
+						 skb_cb->htt.frag_len -
+						 skb_cb->htt.pad_len);
 	cmd->data_tx.id          = __cpu_to_le16(msdu_id);
-	cmd->data_tx.frags_paddr = __cpu_to_le32(frags_paddr);
+	cmd->data_tx.frags_paddr = __cpu_to_le32(skb_cb->paddr);
 	cmd->data_tx.peerid      = __cpu_to_le32(HTT_INVALID_PEERID);
 
-	memcpy(cmd->data_tx.prefetch, msdu->data, prefetch_len);
-
-	/* refcount is decremented by HTC and HTT completions until it reaches
-	 * zero and is freed */
-	skb_cb = ATH10K_SKB_CB(txdesc);
-	skb_cb->htt.msdu_id = msdu_id;
-	skb_cb->htt.refcount = 2;
-	skb_cb->htt.txfrag = txfrag;
-	skb_cb->htt.msdu = msdu;
+	memcpy(cmd->data_tx.prefetch, hdr, prefetch_len);
 
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
-		goto err;
+		goto err_unmap_msdu;
 
 	return 0;
-err:
-	if (txfrag)
-		ath10k_skb_unmap(dev, txfrag);
-	if (txdesc)
-		dev_kfree_skb_any(txdesc);
-	if (txfrag)
-		dev_kfree_skb_any(txfrag);
-	if (msdu_id >= 0) {
-		spin_lock_bh(&htt->tx_lock);
-		htt->pending_tx[msdu_id] = NULL;
-		ath10k_htt_tx_free_msdu_id(htt, msdu_id);
-		spin_unlock_bh(&htt->tx_lock);
-	}
-	ath10k_htt_tx_dec_pending(htt);
+
+err_unmap_msdu:
 	ath10k_skb_unmap(dev, msdu);
+err_pull_txfrag:
+	skb_pull(msdu, skb_cb->htt.frag_len + skb_cb->htt.pad_len);
+err_free_txdesc:
+	dev_kfree_skb_any(txdesc);
+err_free_msdu_id:
+	spin_lock_bh(&htt->tx_lock);
+	htt->pending_tx[msdu_id] = NULL;
+	ath10k_htt_tx_free_msdu_id(htt, msdu_id);
+	spin_unlock_bh(&htt->tx_lock);
+err_tx_dec:
+	ath10k_htt_tx_dec_pending(htt);
+err:
 	return res;
 }
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 643f0c9f052b..8c1be7685922 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -74,7 +74,11 @@ enum ath10k_mcast2ucast_mode {
 #define TARGET_RX_CHAIN_MASK			(BIT(0) | BIT(1) | BIT(2))
 #define TARGET_RX_TIMEOUT_LO_PRI		100
 #define TARGET_RX_TIMEOUT_HI_PRI		40
-#define TARGET_RX_DECAP_MODE			ATH10K_HW_TXRX_ETHERNET
+
+/* Native Wifi decap mode is used to align IP frames to 4-byte boundaries and
+ * avoid a very expensive re-alignment in mac80211. */
+#define TARGET_RX_DECAP_MODE			ATH10K_HW_TXRX_NATIVE_WIFI
+
 #define TARGET_SCAN_MAX_PENDING_REQS		4
 #define TARGET_BMISS_OFFLOAD_MAX_VDEV		3
 #define TARGET_ROAM_OFFLOAD_MAX_VDEV		3
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 248248b5c196..99a9bad3f398 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -460,6 +460,11 @@ static int ath10k_vdev_start(struct ath10k_vif *arvif)
 		arg.ssid_len = arvif->vif->bss_conf.ssid_len;
 	}
 
+	ath10k_dbg(ATH10K_DBG_MAC,
+		   "mac vdev %d start center_freq %d phymode %s\n",
+		   arg.vdev_id, arg.channel.freq,
+		   ath10k_wmi_phymode_str(arg.channel.mode));
+
 	ret = ath10k_wmi_vdev_start(ar, &arg);
 	if (ret) {
 		ath10k_warn("WMI vdev start failed: ret %d\n", ret);
@@ -604,7 +609,7 @@ static int ath10k_monitor_create(struct ath10k *ar)
 		goto vdev_fail;
 	}
 
-	ath10k_dbg(ATH10K_DBG_MAC, "Monitor interface created, vdev id: %d\n",
+	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor vdev %d created\n",
 		   ar->monitor_vdev_id);
 
 	ar->monitor_present = true;
@@ -636,7 +641,7 @@ static int ath10k_monitor_destroy(struct ath10k *ar)
 	ar->free_vdev_map |= 1 << (ar->monitor_vdev_id);
 	ar->monitor_present = false;
 
-	ath10k_dbg(ATH10K_DBG_MAC, "Monitor interface destroyed, vdev id: %d\n",
+	ath10k_dbg(ATH10K_DBG_MAC, "mac monitor vdev %d deleted\n",
 		   ar->monitor_vdev_id);
 	return ret;
 }
@@ -665,7 +670,7 @@ static void ath10k_control_beaconing(struct ath10k_vif *arvif,
 			    arvif->vdev_id);
 		return;
 	}
-	ath10k_dbg(ATH10K_DBG_MAC, "VDEV: %d up\n", arvif->vdev_id);
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d up\n", arvif->vdev_id);
 }
 
 static void ath10k_control_ibss(struct ath10k_vif *arvif,
@@ -749,14 +754,14 @@ static void ath10k_ps_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
 		psmode = WMI_STA_PS_MODE_DISABLED;
 	}
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d psmode %s\n",
+		   arvif->vdev_id, psmode ? "enable" : "disable");
+
 	ar_iter->ret = ath10k_wmi_set_psmode(ar_iter->ar, arvif->vdev_id,
 					     psmode);
 	if (ar_iter->ret)
 		ath10k_warn("Failed to set PS Mode: %d for VDEV: %d\n",
 			    psmode, arvif->vdev_id);
-	else
-		ath10k_dbg(ATH10K_DBG_MAC, "Set PS Mode: %d for VDEV: %d\n",
-			   psmode, arvif->vdev_id);
 }
 
 /**********************/
@@ -946,7 +951,8 @@ static void ath10k_peer_assoc_h_ht(struct ath10k *ar,
 	arg->peer_ht_rates.num_rates = n;
 	arg->peer_num_spatial_streams = max((n+7) / 8, 1);
 
-	ath10k_dbg(ATH10K_DBG_MAC, "mcs cnt %d nss %d\n",
+	ath10k_dbg(ATH10K_DBG_MAC, "mac ht peer %pM mcs cnt %d nss %d\n",
+		   arg->addr,
 		   arg->peer_ht_rates.num_rates,
 		   arg->peer_num_spatial_streams);
 }
@@ -966,7 +972,7 @@ static void ath10k_peer_assoc_h_qos_ap(struct ath10k *ar,
 		arg->peer_flags |= WMI_PEER_QOS;
 
 	if (sta->wme && sta->uapsd_queues) {
-		ath10k_dbg(ATH10K_DBG_MAC, "uapsd_queues: 0x%X, max_sp: %d\n",
+		ath10k_dbg(ATH10K_DBG_MAC, "mac uapsd_queues 0x%x max_sp %d\n",
 			   sta->uapsd_queues, sta->max_sp);
 
 		arg->peer_flags |= WMI_PEER_APSD;
@@ -1045,7 +1051,8 @@ static void ath10k_peer_assoc_h_vht(struct ath10k *ar,
 	arg->peer_vht_rates.tx_mcs_set =
 		__le16_to_cpu(vht_cap->vht_mcs.tx_mcs_map);
 
-	ath10k_dbg(ATH10K_DBG_MAC, "mac vht peer\n");
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vht peer %pM max_mpdu %d flags 0x%x\n",
+		   sta->addr, arg->peer_max_mpdu, arg->peer_flags);
 }
 
 static void ath10k_peer_assoc_h_qos(struct ath10k *ar,
@@ -1073,8 +1080,6 @@ static void ath10k_peer_assoc_h_phymode(struct ath10k *ar,
 {
 	enum wmi_phy_mode phymode = MODE_UNKNOWN;
 
-	/* FIXME: add VHT */
-
 	switch (ar->hw->conf.chandef.chan->band) {
 	case IEEE80211_BAND_2GHZ:
 		if (sta->ht_cap.ht_supported) {
@@ -1088,7 +1093,17 @@ static void ath10k_peer_assoc_h_phymode(struct ath10k *ar,
 
 		break;
 	case IEEE80211_BAND_5GHZ:
-		if (sta->ht_cap.ht_supported) {
+		/*
+		 * Check VHT first.
+		 */
+		if (sta->vht_cap.vht_supported) {
+			if (sta->bandwidth == IEEE80211_STA_RX_BW_80)
+				phymode = MODE_11AC_VHT80;
+			else if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
+				phymode = MODE_11AC_VHT40;
+			else if (sta->bandwidth == IEEE80211_STA_RX_BW_20)
+				phymode = MODE_11AC_VHT20;
+		} else if (sta->ht_cap.ht_supported) {
 			if (sta->bandwidth == IEEE80211_STA_RX_BW_40)
 				phymode = MODE_11NA_HT40;
 			else
@@ -1102,6 +1117,9 @@ static void ath10k_peer_assoc_h_phymode(struct ath10k *ar,
 		break;
 	}
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac peer %pM phymode %s\n",
+		   sta->addr, ath10k_wmi_phymode_str(phymode));
+
 	arg->peer_phymode = phymode;
 	WARN_ON(phymode == MODE_UNKNOWN);
 }
@@ -1159,15 +1177,15 @@ static void ath10k_bss_assoc(struct ieee80211_hw *hw,
 
 	rcu_read_unlock();
 
+	ath10k_dbg(ATH10K_DBG_MAC,
+		   "mac vdev %d up (associated) bssid %pM aid %d\n",
+		   arvif->vdev_id, bss_conf->bssid, bss_conf->aid);
+
 	ret = ath10k_wmi_vdev_up(ar, arvif->vdev_id, bss_conf->aid,
 				 bss_conf->bssid);
 	if (ret)
 		ath10k_warn("VDEV: %d up failed: ret %d\n",
 			    arvif->vdev_id, ret);
-	else
-		ath10k_dbg(ATH10K_DBG_MAC,
-			   "VDEV: %d associated, BSSID: %pM, AID: %d\n",
-			   arvif->vdev_id, bss_conf->bssid, bss_conf->aid);
 }
 
 /*
@@ -1188,10 +1206,11 @@ static void ath10k_bss_disassoc(struct ieee80211_hw *hw,
 	 * No idea why this happens, even though VDEV-DOWN is supposed
 	 * to be analogous to link down, so just stop the VDEV.
 	 */
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d stop (disassociated\n",
+		   arvif->vdev_id);
+
+	/* FIXME: check return value */
 	ret = ath10k_vdev_stop(arvif);
-	if (!ret)
-		ath10k_dbg(ATH10K_DBG_MAC, "VDEV: %d stopped\n",
-			   arvif->vdev_id);
 
 	/*
 	 * If we don't call VDEV-DOWN after VDEV-STOP FW will remain active and
@@ -1200,12 +1219,10 @@ static void ath10k_bss_disassoc(struct ieee80211_hw *hw,
 	 * interfaces as it expects there is no rx when no interface is
 	 * running.
 	 */
-	ret = ath10k_wmi_vdev_down(ar, arvif->vdev_id);
-	if (ret)
-		ath10k_dbg(ATH10K_DBG_MAC, "VDEV: %d ath10k_wmi_vdev_down failed (%d)\n",
-			   arvif->vdev_id, ret);
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d down\n", arvif->vdev_id);
 
-	ath10k_wmi_flush_tx(ar);
+	/* FIXME: why don't we print error if wmi call fails? */
+	ret = ath10k_wmi_vdev_down(ar, arvif->vdev_id);
 
 	arvif->def_wep_key_index = 0;
 }
@@ -1330,8 +1347,8 @@ static int ath10k_update_channel_list(struct ath10k *ar)
 				continue;
 
 			ath10k_dbg(ATH10K_DBG_WMI,
-				   "%s: [%zd/%d] freq %d maxpower %d regpower %d antenna %d mode %d\n",
-				   __func__, ch - arg.channels, arg.n_channels,
+				   "mac channel [%zd/%d] freq %d maxpower %d regpower %d antenna %d mode %d\n",
+				    ch - arg.channels, arg.n_channels,
 				   ch->freq, ch->max_power, ch->max_reg_power,
 				   ch->max_antenna_gain, ch->mode);
 
@@ -1431,7 +1448,8 @@ static void ath10k_tx_h_update_wep_key(struct sk_buff *skb)
 	if (key->keyidx == arvif->def_wep_key_index)
 		return;
 
-	ath10k_dbg(ATH10K_DBG_MAC, "new wep keyidx will be %d\n", key->keyidx);
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d keyidx %d\n",
+		   arvif->vdev_id, key->keyidx);
 
 	ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 					WMI_VDEV_PARAM_DEF_KEYID,
@@ -1534,7 +1552,7 @@ void ath10k_offchan_tx_work(struct work_struct *work)
 
 		mutex_lock(&ar->conf_mutex);
 
-		ath10k_dbg(ATH10K_DBG_MAC, "processing offchannel skb %p\n",
+		ath10k_dbg(ATH10K_DBG_MAC, "mac offchannel skb %p\n",
 			   skb);
 
 		hdr = (struct ieee80211_hdr *)skb->data;
@@ -1546,6 +1564,7 @@ void ath10k_offchan_tx_work(struct work_struct *work)
 		spin_unlock_bh(&ar->data_lock);
 
 		if (peer)
+			/* FIXME: should this use ath10k_warn()? */
 			ath10k_dbg(ATH10K_DBG_MAC, "peer %pM on vdev %d already present\n",
 				   peer_addr, vdev_id);
 
@@ -1643,8 +1662,6 @@ static int ath10k_abort_scan(struct ath10k *ar)
 		return -EIO;
 	}
 
-	ath10k_wmi_flush_tx(ar);
-
 	ret = wait_for_completion_timeout(&ar->scan.completed, 3*HZ);
 	if (ret == 0)
 		ath10k_warn("timed out while waiting for scan to stop\n");
@@ -1678,10 +1695,6 @@ static int ath10k_start_scan(struct ath10k *ar,
 	if (ret)
 		return ret;
 
-	/* make sure we submit the command so the completion
-	* timeout makes sense */
-	ath10k_wmi_flush_tx(ar);
-
 	ret = wait_for_completion_timeout(&ar->scan.started, 1*HZ);
 	if (ret == 0) {
 		ath10k_abort_scan(ar);
@@ -1744,7 +1757,7 @@ static void ath10k_tx(struct ieee80211_hw *hw,
 		ath10k_tx_h_seq_no(skb);
 	}
 
-	memset(ATH10K_SKB_CB(skb), 0, sizeof(*ATH10K_SKB_CB(skb)));
+	ATH10K_SKB_CB(skb)->htt.is_offchan = false;
 	ATH10K_SKB_CB(skb)->htt.vdev_id = vdev_id;
 	ATH10K_SKB_CB(skb)->htt.tid = tid;
 
@@ -1886,7 +1899,7 @@ static int ath10k_config(struct ieee80211_hw *hw, u32 changed)
 	mutex_lock(&ar->conf_mutex);
 
 	if (changed & IEEE80211_CONF_CHANGE_CHANNEL) {
-		ath10k_dbg(ATH10K_DBG_MAC, "Config channel %d mhz\n",
+		ath10k_dbg(ATH10K_DBG_MAC, "mac config channel %d mhz\n",
 			   conf->chandef.chan->center_freq);
 		spin_lock_bh(&ar->data_lock);
 		ar->rx_channel = conf->chandef.chan;
@@ -1903,7 +1916,6 @@ static int ath10k_config(struct ieee80211_hw *hw, u32 changed)
 			ret = ath10k_monitor_destroy(ar);
 	}
 
-	ath10k_wmi_flush_tx(ar);
 	mutex_unlock(&ar->conf_mutex);
 	return ret;
 }
@@ -1975,7 +1987,7 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		break;
 	}
 
-	ath10k_dbg(ATH10K_DBG_MAC, "Add interface: id %d type %d subtype %d\n",
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev create %d (add interface) type %d subtype %d\n",
 		   arvif->vdev_id, arvif->vdev_type, arvif->vdev_subtype);
 
 	ret = ath10k_wmi_vdev_create(ar, arvif->vdev_id, arvif->vdev_type,
@@ -2054,7 +2066,12 @@ static void ath10k_remove_interface(struct ieee80211_hw *hw,
 
 	mutex_lock(&ar->conf_mutex);
 
-	ath10k_dbg(ATH10K_DBG_MAC, "Remove interface: id %d\n", arvif->vdev_id);
+	spin_lock_bh(&ar->data_lock);
+	if (arvif->beacon) {
+		dev_kfree_skb_any(arvif->beacon);
+		arvif->beacon = NULL;
+	}
+	spin_unlock_bh(&ar->data_lock);
 
 	ar->free_vdev_map |= 1 << (arvif->vdev_id);
 
@@ -2066,6 +2083,9 @@ static void ath10k_remove_interface(struct ieee80211_hw *hw,
 		kfree(arvif->u.ap.noa_data);
 	}
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev delete %d (remove interface)\n",
+		   arvif->vdev_id);
+
 	ret = ath10k_wmi_vdev_delete(ar, arvif->vdev_id);
 	if (ret)
 		ath10k_warn("WMI vdev delete failed: %d\n", ret);
@@ -2107,18 +2127,20 @@ static void ath10k_configure_filter(struct ieee80211_hw *hw,
 
 	if ((ar->filter_flags & FIF_PROMISC_IN_BSS) &&
 	    !ar->monitor_enabled) {
+		ath10k_dbg(ATH10K_DBG_MAC, "mac monitor %d start\n",
+			   ar->monitor_vdev_id);
+
 		ret = ath10k_monitor_start(ar, ar->monitor_vdev_id);
 		if (ret)
 			ath10k_warn("Unable to start monitor mode\n");
-		else
-			ath10k_dbg(ATH10K_DBG_MAC, "Monitor mode started\n");
 	} else if (!(ar->filter_flags & FIF_PROMISC_IN_BSS) &&
 		   ar->monitor_enabled) {
+		ath10k_dbg(ATH10K_DBG_MAC, "mac monitor %d stop\n",
+			   ar->monitor_vdev_id);
+
 		ret = ath10k_monitor_stop(ar);
 		if (ret)
 			ath10k_warn("Unable to stop monitor mode\n");
-		else
-			ath10k_dbg(ATH10K_DBG_MAC, "Monitor mode stopped\n");
 	}
 
 	mutex_unlock(&ar->conf_mutex);
@@ -2143,41 +2165,41 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 						WMI_VDEV_PARAM_BEACON_INTERVAL,
 						arvif->beacon_interval);
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac vdev %d beacon_interval %d\n",
+			   arvif->vdev_id, arvif->beacon_interval);
+
 		if (ret)
 			ath10k_warn("Failed to set beacon interval for VDEV: %d\n",
 				    arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Beacon interval: %d set for VDEV: %d\n",
-				   arvif->beacon_interval, arvif->vdev_id);
 	}
 
 	if (changed & BSS_CHANGED_BEACON) {
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "vdev %d set beacon tx mode to staggered\n",
+			   arvif->vdev_id);
+
 		ret = ath10k_wmi_pdev_set_param(ar,
 						WMI_PDEV_PARAM_BEACON_TX_MODE,
 						WMI_BEACON_STAGGERED_MODE);
 		if (ret)
 			ath10k_warn("Failed to set beacon mode for VDEV: %d\n",
 				    arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Set staggered beacon mode for VDEV: %d\n",
-				   arvif->vdev_id);
 	}
 
 	if (changed & BSS_CHANGED_BEACON_INFO) {
 		arvif->dtim_period = info->dtim_period;
 
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac vdev %d dtim_period %d\n",
+			   arvif->vdev_id, arvif->dtim_period);
+
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 						WMI_VDEV_PARAM_DTIM_PERIOD,
 						arvif->dtim_period);
 		if (ret)
 			ath10k_warn("Failed to set dtim period for VDEV: %d\n",
 				    arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Set dtim period: %d for VDEV: %d\n",
-				   arvif->dtim_period, arvif->vdev_id);
 	}
 
 	if (changed & BSS_CHANGED_SSID &&
@@ -2190,16 +2212,15 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 
 	if (changed & BSS_CHANGED_BSSID) {
 		if (!is_zero_ether_addr(info->bssid)) {
+			ath10k_dbg(ATH10K_DBG_MAC,
+				   "mac vdev %d create peer %pM\n",
+				   arvif->vdev_id, info->bssid);
+
 			ret = ath10k_peer_create(ar, arvif->vdev_id,
 						 info->bssid);
 			if (ret)
 				ath10k_warn("Failed to add peer: %pM for VDEV: %d\n",
 					    info->bssid, arvif->vdev_id);
-			else
-				ath10k_dbg(ATH10K_DBG_MAC,
-					   "Added peer: %pM for VDEV: %d\n",
-					   info->bssid, arvif->vdev_id);
-
 
 			if (vif->type == NL80211_IFTYPE_STATION) {
 				/*
@@ -2209,11 +2230,12 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 				memcpy(arvif->u.sta.bssid, info->bssid,
 				       ETH_ALEN);
 
+				ath10k_dbg(ATH10K_DBG_MAC,
+					   "mac vdev %d start %pM\n",
+					   arvif->vdev_id, info->bssid);
+
+				/* FIXME: check return value */
 				ret = ath10k_vdev_start(arvif);
-				if (!ret)
-					ath10k_dbg(ATH10K_DBG_MAC,
-						   "VDEV: %d started with BSSID: %pM\n",
-						   arvif->vdev_id, info->bssid);
 			}
 
 			/*
@@ -2237,16 +2259,15 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		else
 			cts_prot = 0;
 
+		ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d cts_prot %d\n",
+			   arvif->vdev_id, cts_prot);
+
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 						WMI_VDEV_PARAM_ENABLE_RTSCTS,
 						cts_prot);
 		if (ret)
 			ath10k_warn("Failed to set CTS prot for VDEV: %d\n",
 				    arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Set CTS prot: %d for VDEV: %d\n",
-				   cts_prot, arvif->vdev_id);
 	}
 
 	if (changed & BSS_CHANGED_ERP_SLOT) {
@@ -2257,16 +2278,15 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		else
 			slottime = WMI_VDEV_SLOT_TIME_LONG; /* 20us */
 
+		ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d slot_time %d\n",
+			   arvif->vdev_id, slottime);
+
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 						WMI_VDEV_PARAM_SLOT_TIME,
 						slottime);
 		if (ret)
 			ath10k_warn("Failed to set erp slot for VDEV: %d\n",
 				    arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Set slottime: %d for VDEV: %d\n",
-				   slottime, arvif->vdev_id);
 	}
 
 	if (changed & BSS_CHANGED_ERP_PREAMBLE) {
@@ -2276,16 +2296,16 @@ static void ath10k_bss_info_changed(struct ieee80211_hw *hw,
 		else
 			preamble = WMI_VDEV_PREAMBLE_LONG;
 
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac vdev %d preamble %dn",
+			   arvif->vdev_id, preamble);
+
 		ret = ath10k_wmi_vdev_set_param(ar, arvif->vdev_id,
 						WMI_VDEV_PARAM_PREAMBLE,
 						preamble);
 		if (ret)
 			ath10k_warn("Failed to set preamble for VDEV: %d\n",
 				    arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Set preamble: %d for VDEV: %d\n",
-				   preamble, arvif->vdev_id);
 	}
 
 	if (changed & BSS_CHANGED_ASSOC) {
@@ -2476,27 +2496,26 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 		/*
 		 * New station addition.
 		 */
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac vdev %d peer create %pM (new sta)\n",
+			   arvif->vdev_id, sta->addr);
+
 		ret = ath10k_peer_create(ar, arvif->vdev_id, sta->addr);
 		if (ret)
 			ath10k_warn("Failed to add peer: %pM for VDEV: %d\n",
 				    sta->addr, arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Added peer: %pM for VDEV: %d\n",
-				   sta->addr, arvif->vdev_id);
 	} else if ((old_state == IEEE80211_STA_NONE &&
 		    new_state == IEEE80211_STA_NOTEXIST)) {
 		/*
 		 * Existing station deletion.
 		 */
+		ath10k_dbg(ATH10K_DBG_MAC,
+			   "mac vdev %d peer delete %pM (sta gone)\n",
+			   arvif->vdev_id, sta->addr);
 		ret = ath10k_peer_delete(ar, arvif->vdev_id, sta->addr);
 		if (ret)
 			ath10k_warn("Failed to delete peer: %pM for VDEV: %d\n",
 				    sta->addr, arvif->vdev_id);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Removed peer: %pM for VDEV: %d\n",
-				   sta->addr, arvif->vdev_id);
 
 		if (vif->type == NL80211_IFTYPE_STATION)
 			ath10k_bss_disassoc(hw, vif);
@@ -2507,14 +2526,13 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 		/*
 		 * New association.
 		 */
+		ath10k_dbg(ATH10K_DBG_MAC, "mac sta %pM associated\n",
+			   sta->addr);
+
 		ret = ath10k_station_assoc(ar, arvif, sta);
 		if (ret)
 			ath10k_warn("Failed to associate station: %pM\n",
 				    sta->addr);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Station %pM moved to assoc state\n",
-				   sta->addr);
 	} else if (old_state == IEEE80211_STA_ASSOC &&
 		   new_state == IEEE80211_STA_AUTH &&
 		   (vif->type == NL80211_IFTYPE_AP ||
@@ -2522,14 +2540,13 @@ static int ath10k_sta_state(struct ieee80211_hw *hw,
 		/*
 		 * Disassociation.
 		 */
+		ath10k_dbg(ATH10K_DBG_MAC, "mac sta %pM disassociated\n",
+			   sta->addr);
+
 		ret = ath10k_station_disassoc(ar, arvif, sta);
 		if (ret)
 			ath10k_warn("Failed to disassociate station: %pM\n",
 				    sta->addr);
-		else
-			ath10k_dbg(ATH10K_DBG_MAC,
-				   "Station %pM moved to disassociated state\n",
-				   sta->addr);
 	}
 
 	mutex_unlock(&ar->conf_mutex);
@@ -2749,14 +2766,13 @@ static void ath10k_set_rts_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
 	if (ar_iter->ar->state == ATH10K_STATE_RESTARTED)
 		return;
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d rts_threshold %d\n",
+		   arvif->vdev_id, rts);
+
 	ar_iter->ret = ath10k_mac_set_rts(arvif, rts);
 	if (ar_iter->ret)
 		ath10k_warn("Failed to set RTS threshold for VDEV: %d\n",
 			    arvif->vdev_id);
-	else
-		ath10k_dbg(ATH10K_DBG_MAC,
-			   "Set RTS threshold: %d for VDEV: %d\n",
-			   rts, arvif->vdev_id);
 }
 
 static int ath10k_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
@@ -2791,14 +2807,13 @@ static void ath10k_set_frag_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
 	if (ar_iter->ar->state == ATH10K_STATE_RESTARTED)
 		return;
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d fragmentation_threshold %d\n",
+		   arvif->vdev_id, frag);
+
 	ar_iter->ret = ath10k_mac_set_frag(arvif, frag);
 	if (ar_iter->ret)
 		ath10k_warn("Failed to set frag threshold for VDEV: %d\n",
 			    arvif->vdev_id);
-	else
-		ath10k_dbg(ATH10K_DBG_MAC,
-			   "Set frag threshold: %d for VDEV: %d\n",
-			   frag, arvif->vdev_id);
 }
 
 static int ath10k_set_frag_threshold(struct ieee80211_hw *hw, u32 value)
@@ -2838,8 +2853,7 @@ static void ath10k_flush(struct ieee80211_hw *hw, u32 queues, bool drop)
 			bool empty;
 
 			spin_lock_bh(&ar->htt.tx_lock);
-			empty = bitmap_empty(ar->htt.used_msdu_ids,
-					     ar->htt.max_num_pending_tx);
+			empty = (ar->htt.num_pending_tx == 0);
 			spin_unlock_bh(&ar->htt.tx_lock);
 
 			skip = (ar->state == ATH10K_STATE_WEDGED);
@@ -3328,6 +3342,10 @@ int ath10k_mac_register(struct ath10k *ar)
 			IEEE80211_HW_WANT_MONITOR_VIF |
 			IEEE80211_HW_AP_LINK_PS;
 
+	/* MSDU can have HTT TX fragment pushed in front. The additional 4
+	 * bytes is used for padding/alignment if necessary. */
+	ar->hw->extra_tx_headroom += sizeof(struct htt_data_tx_desc_frag)*2 + 4;
+
 	if (ar->ht_cap_info & WMI_HT_CAP_DYNAMIC_SMPS)
 		ar->hw->flags |= IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS;
 
diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 56501f47ab9c..dff23d97bed0 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -612,31 +612,20 @@ exit:
 }
 
 /* Called by lower (CE) layer when a send to Target completes. */
-static void ath10k_pci_ce_send_done(struct ath10k_ce_pipe *ce_state,
-				    void *transfer_context,
-				    u32 ce_data,
-				    unsigned int nbytes,
-				    unsigned int transfer_id)
+static void ath10k_pci_ce_send_done(struct ath10k_ce_pipe *ce_state)
 {
 	struct ath10k *ar = ce_state->ar;
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pipe_info =  &ar_pci->pipe_info[ce_state->id];
 	struct ath10k_pci_compl *compl;
-	bool process = false;
-
-	do {
-		/*
-		 * For the send completion of an item in sendlist, just
-		 * increment num_sends_allowed. The upper layer callback will
-		 * be triggered when last fragment is done with send.
-		 */
-		if (transfer_context == CE_SENDLIST_ITEM_CTXT) {
-			spin_lock_bh(&pipe_info->pipe_lock);
-			pipe_info->num_sends_allowed++;
-			spin_unlock_bh(&pipe_info->pipe_lock);
-			continue;
-		}
+	void *transfer_context;
+	u32 ce_data;
+	unsigned int nbytes;
+	unsigned int transfer_id;
 
+	while (ath10k_ce_completed_send_next(ce_state, &transfer_context,
+					     &ce_data, &nbytes,
+					     &transfer_id) == 0) {
 		compl = get_free_compl(pipe_info);
 		if (!compl)
 			break;
@@ -655,38 +644,28 @@ static void ath10k_pci_ce_send_done(struct ath10k_ce_pipe *ce_state,
 		spin_lock_bh(&ar_pci->compl_lock);
 		list_add_tail(&compl->list, &ar_pci->compl_process);
 		spin_unlock_bh(&ar_pci->compl_lock);
-
-		process = true;
-	} while (ath10k_ce_completed_send_next(ce_state,
-							   &transfer_context,
-							   &ce_data, &nbytes,
-							   &transfer_id) == 0);
-
-	/*
-	 * If only some of the items within a sendlist have completed,
-	 * don't invoke completion processing until the entire sendlist
-	 * has been sent.
-	 */
-	if (!process)
-		return;
+	}
 
 	ath10k_pci_process_ce(ar);
 }
 
 /* Called by lower (CE) layer when data is received from the Target. */
-static void ath10k_pci_ce_recv_data(struct ath10k_ce_pipe *ce_state,
-				    void *transfer_context, u32 ce_data,
-				    unsigned int nbytes,
-				    unsigned int transfer_id,
-				    unsigned int flags)
+static void ath10k_pci_ce_recv_data(struct ath10k_ce_pipe *ce_state)
 {
 	struct ath10k *ar = ce_state->ar;
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pipe_info =  &ar_pci->pipe_info[ce_state->id];
 	struct ath10k_pci_compl *compl;
 	struct sk_buff *skb;
+	void *transfer_context;
+	u32 ce_data;
+	unsigned int nbytes;
+	unsigned int transfer_id;
+	unsigned int flags;
 
-	do {
+	while (ath10k_ce_completed_recv_next(ce_state, &transfer_context,
+					     &ce_data, &nbytes, &transfer_id,
+					     &flags) == 0) {
 		compl = get_free_compl(pipe_info);
 		if (!compl)
 			break;
@@ -709,12 +688,7 @@ static void ath10k_pci_ce_recv_data(struct ath10k_ce_pipe *ce_state,
 		spin_lock_bh(&ar_pci->compl_lock);
 		list_add_tail(&compl->list, &ar_pci->compl_process);
 		spin_unlock_bh(&ar_pci->compl_lock);
-
-	} while (ath10k_ce_completed_recv_next(ce_state,
-							   &transfer_context,
-							   &ce_data, &nbytes,
-							   &transfer_id,
-							   &flags) == 0);
+	}
 
 	ath10k_pci_process_ce(ar);
 }
@@ -728,13 +702,10 @@ static int ath10k_pci_hif_send_head(struct ath10k *ar, u8 pipe_id,
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pipe_info = &(ar_pci->pipe_info[pipe_id]);
 	struct ath10k_ce_pipe *ce_hdl = pipe_info->ce_hdl;
-	struct ce_sendlist sendlist;
 	unsigned int len;
 	u32 flags = 0;
 	int ret;
 
-	memset(&sendlist, 0, sizeof(struct ce_sendlist));
-
 	len = min(bytes, nbuf->len);
 	bytes -= len;
 
@@ -749,8 +720,6 @@ static int ath10k_pci_hif_send_head(struct ath10k *ar, u8 pipe_id,
 			"ath10k tx: data: ",
 			nbuf->data, nbuf->len);
 
-	ath10k_ce_sendlist_buf_add(&sendlist, skb_cb->paddr, len, flags);
-
 	/* Make sure we have resources to handle this request */
 	spin_lock_bh(&pipe_info->pipe_lock);
 	if (!pipe_info->num_sends_allowed) {
@@ -761,7 +730,8 @@ static int ath10k_pci_hif_send_head(struct ath10k *ar, u8 pipe_id,
 	pipe_info->num_sends_allowed--;
 	spin_unlock_bh(&pipe_info->pipe_lock);
 
-	ret = ath10k_ce_sendlist_send(ce_hdl, nbuf, &sendlist, transfer_id);
+	ret = ath10k_ce_sendlist_send(ce_hdl, nbuf, transfer_id,
+				      skb_cb->paddr, len, flags);
 	if (ret)
 		ath10k_warn("CE send failed: %p\n", nbuf);
 
@@ -1316,15 +1286,14 @@ static void ath10k_pci_tx_pipe_cleanup(struct ath10k_pci_pipe *pipe_info)
 
 	while (ath10k_ce_cancel_send_next(ce_hdl, (void **)&netbuf,
 					  &ce_data, &nbytes, &id) == 0) {
-		if (netbuf != CE_SENDLIST_ITEM_CTXT)
-			/*
-			 * Indicate the completion to higer layer to free
-			 * the buffer
-			 */
-			ATH10K_SKB_CB(netbuf)->is_aborted = true;
-			ar_pci->msg_callbacks_current.tx_completion(ar,
-								    netbuf,
-								    id);
+		/*
+		 * Indicate the completion to higer layer to free
+		 * the buffer
+		 */
+		ATH10K_SKB_CB(netbuf)->is_aborted = true;
+		ar_pci->msg_callbacks_current.tx_completion(ar,
+							    netbuf,
+							    id);
 	}
 }
 
@@ -1490,13 +1459,16 @@ err_dma:
 	return ret;
 }
 
-static void ath10k_pci_bmi_send_done(struct ath10k_ce_pipe *ce_state,
-				     void *transfer_context,
-				     u32 data,
-				     unsigned int nbytes,
-				     unsigned int transfer_id)
+static void ath10k_pci_bmi_send_done(struct ath10k_ce_pipe *ce_state)
 {
-	struct bmi_xfer *xfer = transfer_context;
+	struct bmi_xfer *xfer;
+	u32 ce_data;
+	unsigned int nbytes;
+	unsigned int transfer_id;
+
+	if (ath10k_ce_completed_send_next(ce_state, (void **)&xfer, &ce_data,
+					  &nbytes, &transfer_id))
+		return;
 
 	if (xfer->wait_for_resp)
 		return;
@@ -1504,14 +1476,17 @@ static void ath10k_pci_bmi_send_done(struct ath10k_ce_pipe *ce_state,
 	complete(&xfer->done);
 }
 
-static void ath10k_pci_bmi_recv_data(struct ath10k_ce_pipe *ce_state,
-				     void *transfer_context,
-				     u32 data,
-				     unsigned int nbytes,
-				     unsigned int transfer_id,
-				     unsigned int flags)
+static void ath10k_pci_bmi_recv_data(struct ath10k_ce_pipe *ce_state)
 {
-	struct bmi_xfer *xfer = transfer_context;
+	struct bmi_xfer *xfer;
+	u32 ce_data;
+	unsigned int nbytes;
+	unsigned int transfer_id;
+	unsigned int flags;
+
+	if (ath10k_ce_completed_recv_next(ce_state, (void **)&xfer, &ce_data,
+					  &nbytes, &transfer_id, &flags))
+		return;
 
 	if (!xfer->wait_for_resp) {
 		ath10k_warn("unexpected: BMI data received; ignoring\n");
@@ -2374,10 +2349,10 @@ static void ath10k_pci_dump_features(struct ath10k_pci *ar_pci)
 
 		switch (i) {
 		case ATH10K_PCI_FEATURE_MSI_X:
-			ath10k_dbg(ATH10K_DBG_PCI, "device supports MSI-X\n");
+			ath10k_dbg(ATH10K_DBG_BOOT, "device supports MSI-X\n");
 			break;
 		case ATH10K_PCI_FEATURE_SOC_POWER_SAVE:
-			ath10k_dbg(ATH10K_DBG_PCI, "QCA98XX SoC power save enabled\n");
+			ath10k_dbg(ATH10K_DBG_BOOT, "QCA98XX SoC power save enabled\n");
 			break;
 		}
 	}
@@ -2503,6 +2478,8 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 
 	ath10k_do_pci_sleep(ar);
 
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot pci_mem 0x%p\n", ar_pci->mem);
+
 	ret = ath10k_core_register(ar, chip_id);
 	if (ret) {
 		ath10k_err("could not register driver core (%d)\n", ret);
diff --git a/drivers/net/wireless/ath/ath10k/rx_desc.h b/drivers/net/wireless/ath/ath10k/rx_desc.h
index bfec6c8f2ecb..1c584c4b019c 100644
--- a/drivers/net/wireless/ath/ath10k/rx_desc.h
+++ b/drivers/net/wireless/ath/ath10k/rx_desc.h
@@ -422,10 +422,30 @@ struct rx_mpdu_end {
 #define RX_MSDU_START_INFO1_IP_FRAG             (1 << 14)
 #define RX_MSDU_START_INFO1_TCP_ONLY_ACK        (1 << 15)
 
+/* The decapped header (rx_hdr_status) contains the following:
+ *  a) 802.11 header
+ *  [padding to 4 bytes]
+ *  b) HW crypto parameter
+ *     - 0 bytes for no security
+ *     - 4 bytes for WEP
+ *     - 8 bytes for TKIP, AES
+ *  [padding to 4 bytes]
+ *  c) A-MSDU subframe header (14 bytes) if appliable
+ *  d) LLC/SNAP (RFC1042, 8 bytes)
+ *
+ * In case of A-MSDU only first frame in sequence contains (a) and (b). */
 enum rx_msdu_decap_format {
-	RX_MSDU_DECAP_RAW           = 0,
-	RX_MSDU_DECAP_NATIVE_WIFI   = 1,
+	RX_MSDU_DECAP_RAW = 0,
+
+	/* Note: QoS frames are reported as non-QoS. The rx_hdr_status in
+	 * htt_rx_desc contains the original decapped 802.11 header. */
+	RX_MSDU_DECAP_NATIVE_WIFI = 1,
+
+	/* Payload contains an ethernet header (struct ethhdr). */
 	RX_MSDU_DECAP_ETHERNET2_DIX = 2,
+
+	/* Payload contains two 48-bit addresses and 2-byte length (14 bytes
+	 * total), followed by an RFC1042 header (8 bytes). */
 	RX_MSDU_DECAP_8023_SNAP_LLC = 3
 };
 
diff --git a/drivers/net/wireless/ath/ath10k/trace.h b/drivers/net/wireless/ath/ath10k/trace.h
index 85e806bf7257..90817ddc92ba 100644
--- a/drivers/net/wireless/ath/ath10k/trace.h
+++ b/drivers/net/wireless/ath/ath10k/trace.h
@@ -111,26 +111,29 @@ TRACE_EVENT(ath10k_log_dbg_dump,
 );
 
 TRACE_EVENT(ath10k_wmi_cmd,
-	TP_PROTO(int id, void *buf, size_t buf_len),
+	TP_PROTO(int id, void *buf, size_t buf_len, int ret),
 
-	TP_ARGS(id, buf, buf_len),
+	TP_ARGS(id, buf, buf_len, ret),
 
 	TP_STRUCT__entry(
 		__field(unsigned int, id)
 		__field(size_t, buf_len)
 		__dynamic_array(u8, buf, buf_len)
+		__field(int, ret)
 	),
 
 	TP_fast_assign(
 		__entry->id = id;
 		__entry->buf_len = buf_len;
+		__entry->ret = ret;
 		memcpy(__get_dynamic_array(buf), buf, buf_len);
 	),
 
 	TP_printk(
-		"id %d len %zu",
+		"id %d len %zu ret %d",
 		__entry->id,
-		__entry->buf_len
+		__entry->buf_len,
+		__entry->ret
 	)
 );
 
@@ -158,6 +161,27 @@ TRACE_EVENT(ath10k_wmi_event,
 	)
 );
 
+TRACE_EVENT(ath10k_htt_stats,
+	TP_PROTO(void *buf, size_t buf_len),
+
+	TP_ARGS(buf, buf_len),
+
+	TP_STRUCT__entry(
+		__field(size_t, buf_len)
+		__dynamic_array(u8, buf, buf_len)
+	),
+
+	TP_fast_assign(
+		__entry->buf_len = buf_len;
+		memcpy(__get_dynamic_array(buf), buf, buf_len);
+	),
+
+	TP_printk(
+		"len %zu",
+		__entry->buf_len
+	)
+);
+
 #endif /* _TRACE_H_ || TRACE_HEADER_MULTI_READ*/
 
 /* we don't want to use include/trace/events */
diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index 68b6faefd1d8..5ae373a1e294 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -44,40 +44,39 @@ out:
 	spin_unlock_bh(&ar->data_lock);
 }
 
-void ath10k_txrx_tx_unref(struct ath10k_htt *htt, struct sk_buff *txdesc)
+void ath10k_txrx_tx_unref(struct ath10k_htt *htt,
+			  const struct htt_tx_done *tx_done)
 {
 	struct device *dev = htt->ar->dev;
 	struct ieee80211_tx_info *info;
-	struct sk_buff *txfrag = ATH10K_SKB_CB(txdesc)->htt.txfrag;
-	struct sk_buff *msdu = ATH10K_SKB_CB(txdesc)->htt.msdu;
+	struct ath10k_skb_cb *skb_cb;
+	struct sk_buff *msdu;
 	int ret;
 
-	if (ATH10K_SKB_CB(txdesc)->htt.refcount == 0)
-		return;
-
-	ATH10K_SKB_CB(txdesc)->htt.refcount--;
+	ath10k_dbg(ATH10K_DBG_HTT, "htt tx completion msdu_id %u discard %d no_ack %d\n",
+		   tx_done->msdu_id, !!tx_done->discard, !!tx_done->no_ack);
 
-	if (ATH10K_SKB_CB(txdesc)->htt.refcount > 0)
+	if (tx_done->msdu_id >= htt->max_num_pending_tx) {
+		ath10k_warn("warning: msdu_id %d too big, ignoring\n",
+			    tx_done->msdu_id);
 		return;
-
-	if (txfrag) {
-		ret = ath10k_skb_unmap(dev, txfrag);
-		if (ret)
-			ath10k_warn("txfrag unmap failed (%d)\n", ret);
-
-		dev_kfree_skb_any(txfrag);
 	}
 
+	msdu = htt->pending_tx[tx_done->msdu_id];
+	skb_cb = ATH10K_SKB_CB(msdu);
+
 	ret = ath10k_skb_unmap(dev, msdu);
 	if (ret)
 		ath10k_warn("data skb unmap failed (%d)\n", ret);
 
+	if (skb_cb->htt.frag_len)
+		skb_pull(msdu, skb_cb->htt.frag_len + skb_cb->htt.pad_len);
+
 	ath10k_report_offchan_tx(htt->ar, msdu);
 
 	info = IEEE80211_SKB_CB(msdu);
-	memset(&info->status, 0, sizeof(info->status));
 
-	if (ATH10K_SKB_CB(txdesc)->htt.discard) {
+	if (tx_done->discard) {
 		ieee80211_free_txskb(htt->ar->hw, msdu);
 		goto exit;
 	}
@@ -85,7 +84,7 @@ void ath10k_txrx_tx_unref(struct ath10k_htt *htt, struct sk_buff *txdesc)
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK))
 		info->flags |= IEEE80211_TX_STAT_ACK;
 
-	if (ATH10K_SKB_CB(txdesc)->htt.no_ack)
+	if (tx_done->no_ack)
 		info->flags &= ~IEEE80211_TX_STAT_ACK;
 
 	ieee80211_tx_status(htt->ar->hw, msdu);
@@ -93,36 +92,12 @@ void ath10k_txrx_tx_unref(struct ath10k_htt *htt, struct sk_buff *txdesc)
 
 exit:
 	spin_lock_bh(&htt->tx_lock);
-	htt->pending_tx[ATH10K_SKB_CB(txdesc)->htt.msdu_id] = NULL;
-	ath10k_htt_tx_free_msdu_id(htt, ATH10K_SKB_CB(txdesc)->htt.msdu_id);
+	htt->pending_tx[tx_done->msdu_id] = NULL;
+	ath10k_htt_tx_free_msdu_id(htt, tx_done->msdu_id);
 	__ath10k_htt_tx_dec_pending(htt);
-	if (bitmap_empty(htt->used_msdu_ids, htt->max_num_pending_tx))
+	if (htt->num_pending_tx == 0)
 		wake_up(&htt->empty_tx_wq);
 	spin_unlock_bh(&htt->tx_lock);
-
-	dev_kfree_skb_any(txdesc);
-}
-
-void ath10k_txrx_tx_completed(struct ath10k_htt *htt,
-			      const struct htt_tx_done *tx_done)
-{
-	struct sk_buff *txdesc;
-
-	ath10k_dbg(ATH10K_DBG_HTT, "htt tx completion msdu_id %u discard %d no_ack %d\n",
-		   tx_done->msdu_id, !!tx_done->discard, !!tx_done->no_ack);
-
-	if (tx_done->msdu_id >= htt->max_num_pending_tx) {
-		ath10k_warn("warning: msdu_id %d too big, ignoring\n",
-			    tx_done->msdu_id);
-		return;
-	}
-
-	txdesc = htt->pending_tx[tx_done->msdu_id];
-
-	ATH10K_SKB_CB(txdesc)->htt.discard = tx_done->discard;
-	ATH10K_SKB_CB(txdesc)->htt.no_ack = tx_done->no_ack;
-
-	ath10k_txrx_tx_unref(htt, txdesc);
 }
 
 static const u8 rx_legacy_rate_idx[] = {
@@ -293,6 +268,8 @@ void ath10k_process_rx(struct ath10k *ar, struct htt_rx_info *info)
 		   status->vht_nss,
 		   status->freq,
 		   status->band);
+	ath10k_dbg_dump(ATH10K_DBG_HTT_DUMP, NULL, "rx skb: ",
+			info->skb->data, info->skb->len);
 
 	ieee80211_rx(ar->hw, info->skb);
 }
diff --git a/drivers/net/wireless/ath/ath10k/txrx.h b/drivers/net/wireless/ath/ath10k/txrx.h
index e78632a76df7..356dc9c04c9e 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.h
+++ b/drivers/net/wireless/ath/ath10k/txrx.h
@@ -19,9 +19,8 @@
 
 #include "htt.h"
 
-void ath10k_txrx_tx_unref(struct ath10k_htt *htt, struct sk_buff *txdesc);
-void ath10k_txrx_tx_completed(struct ath10k_htt *htt,
-			      const struct htt_tx_done *tx_done);
+void ath10k_txrx_tx_unref(struct ath10k_htt *htt,
+			  const struct htt_tx_done *tx_done);
 void ath10k_process_rx(struct ath10k *ar, struct htt_rx_info *info);
 
 struct ath10k_peer *ath10k_peer_find(struct ath10k *ar, int vdev_id,
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 32fd5e735beb..6803ead9b9cf 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -23,30 +23,6 @@
 #include "wmi.h"
 #include "mac.h"
 
-void ath10k_wmi_flush_tx(struct ath10k *ar)
-{
-	int ret;
-
-	lockdep_assert_held(&ar->conf_mutex);
-
-	if (ar->state == ATH10K_STATE_WEDGED) {
-		ath10k_warn("wmi flush skipped - device is wedged anyway\n");
-		return;
-	}
-
-	ret = wait_event_timeout(ar->wmi.wq,
-				 atomic_read(&ar->wmi.pending_tx_count) == 0,
-				 5*HZ);
-	if (atomic_read(&ar->wmi.pending_tx_count) == 0)
-		return;
-
-	if (ret == 0)
-		ret = -ETIMEDOUT;
-
-	if (ret < 0)
-		ath10k_warn("wmi flush failed (%d)\n", ret);
-}
-
 int ath10k_wmi_wait_for_service_ready(struct ath10k *ar)
 {
 	int ret;
@@ -85,18 +61,14 @@ static struct sk_buff *ath10k_wmi_alloc_skb(u32 len)
 static void ath10k_wmi_htc_tx_complete(struct ath10k *ar, struct sk_buff *skb)
 {
 	dev_kfree_skb(skb);
-
-	if (atomic_sub_return(1, &ar->wmi.pending_tx_count) == 0)
-		wake_up(&ar->wmi.wq);
 }
 
-/* WMI command API */
-static int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb,
-			       enum wmi_cmd_id cmd_id)
+static int ath10k_wmi_cmd_send_nowait(struct ath10k *ar, struct sk_buff *skb,
+				      enum wmi_cmd_id cmd_id)
 {
 	struct ath10k_skb_cb *skb_cb = ATH10K_SKB_CB(skb);
 	struct wmi_cmd_hdr *cmd_hdr;
-	int status;
+	int ret;
 	u32 cmd = 0;
 
 	if (skb_push(skb, sizeof(struct wmi_cmd_hdr)) == NULL)
@@ -107,26 +79,87 @@ static int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb,
 	cmd_hdr = (struct wmi_cmd_hdr *)skb->data;
 	cmd_hdr->cmd_id = __cpu_to_le32(cmd);
 
-	if (atomic_add_return(1, &ar->wmi.pending_tx_count) >
-	    WMI_MAX_PENDING_TX_COUNT) {
-		/* avoid using up memory when FW hangs */
-		dev_kfree_skb(skb);
-		atomic_dec(&ar->wmi.pending_tx_count);
-		return -EBUSY;
-	}
-
 	memset(skb_cb, 0, sizeof(*skb_cb));
+	ret = ath10k_htc_send(&ar->htc, ar->wmi.eid, skb);
+	trace_ath10k_wmi_cmd(cmd_id, skb->data, skb->len, ret);
 
-	trace_ath10k_wmi_cmd(cmd_id, skb->data, skb->len);
+	if (ret)
+		goto err_pull;
 
-	status = ath10k_htc_send(&ar->htc, ar->wmi.eid, skb);
-	if (status) {
+	return 0;
+
+err_pull:
+	skb_pull(skb, sizeof(struct wmi_cmd_hdr));
+	return ret;
+}
+
+static void ath10k_wmi_tx_beacon_nowait(struct ath10k_vif *arvif)
+{
+	struct wmi_bcn_tx_arg arg = {0};
+	int ret;
+
+	lockdep_assert_held(&arvif->ar->data_lock);
+
+	if (arvif->beacon == NULL)
+		return;
+
+	arg.vdev_id = arvif->vdev_id;
+	arg.tx_rate = 0;
+	arg.tx_power = 0;
+	arg.bcn = arvif->beacon->data;
+	arg.bcn_len = arvif->beacon->len;
+
+	ret = ath10k_wmi_beacon_send_nowait(arvif->ar, &arg);
+	if (ret)
+		return;
+
+	dev_kfree_skb_any(arvif->beacon);
+	arvif->beacon = NULL;
+}
+
+static void ath10k_wmi_tx_beacons_iter(void *data, u8 *mac,
+				       struct ieee80211_vif *vif)
+{
+	struct ath10k_vif *arvif = ath10k_vif_to_arvif(vif);
+
+	ath10k_wmi_tx_beacon_nowait(arvif);
+}
+
+static void ath10k_wmi_tx_beacons_nowait(struct ath10k *ar)
+{
+	spin_lock_bh(&ar->data_lock);
+	ieee80211_iterate_active_interfaces_atomic(ar->hw,
+						   IEEE80211_IFACE_ITER_NORMAL,
+						   ath10k_wmi_tx_beacons_iter,
+						   NULL);
+	spin_unlock_bh(&ar->data_lock);
+}
+
+static void ath10k_wmi_op_ep_tx_credits(struct ath10k *ar)
+{
+	/* try to send pending beacons first. they take priority */
+	ath10k_wmi_tx_beacons_nowait(ar);
+
+	wake_up(&ar->wmi.tx_credits_wq);
+}
+
+static int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb,
+			       enum wmi_cmd_id cmd_id)
+{
+	int ret = -EINVAL;
+
+	wait_event_timeout(ar->wmi.tx_credits_wq, ({
+		/* try to send pending beacons first. they take priority */
+		ath10k_wmi_tx_beacons_nowait(ar);
+
+		ret = ath10k_wmi_cmd_send_nowait(ar, skb, cmd_id);
+		(ret != -EAGAIN);
+	}), 3*HZ);
+
+	if (ret)
 		dev_kfree_skb_any(skb);
-		atomic_dec(&ar->wmi.pending_tx_count);
-		return status;
-	}
 
-	return 0;
+	return ret;
 }
 
 static int ath10k_wmi_event_scan(struct ath10k *ar, struct sk_buff *skb)
@@ -748,10 +781,8 @@ static void ath10k_wmi_event_host_swba(struct ath10k *ar, struct sk_buff *skb)
 	int i = -1;
 	struct wmi_bcn_info *bcn_info;
 	struct ath10k_vif *arvif;
-	struct wmi_bcn_tx_arg arg;
 	struct sk_buff *bcn;
 	int vdev_id = 0;
-	int ret;
 
 	ath10k_dbg(ATH10K_DBG_MGMT, "WMI_HOST_SWBA_EVENTID\n");
 
@@ -808,17 +839,17 @@ static void ath10k_wmi_event_host_swba(struct ath10k *ar, struct sk_buff *skb)
 		ath10k_wmi_update_tim(ar, arvif, bcn, bcn_info);
 		ath10k_wmi_update_noa(ar, arvif, bcn, bcn_info);
 
-		arg.vdev_id = arvif->vdev_id;
-		arg.tx_rate = 0;
-		arg.tx_power = 0;
-		arg.bcn = bcn->data;
-		arg.bcn_len = bcn->len;
+		spin_lock_bh(&ar->data_lock);
+		if (arvif->beacon) {
+			ath10k_warn("SWBA overrun on vdev %d\n",
+				    arvif->vdev_id);
+			dev_kfree_skb_any(arvif->beacon);
+		}
 
-		ret = ath10k_wmi_beacon_send(ar, &arg);
-		if (ret)
-			ath10k_warn("could not send beacon (%d)\n", ret);
+		arvif->beacon = bcn;
 
-		dev_kfree_skb_any(bcn);
+		ath10k_wmi_tx_beacon_nowait(arvif);
+		spin_unlock_bh(&ar->data_lock);
 	}
 }
 
@@ -1024,7 +1055,7 @@ static int ath10k_wmi_ready_event_rx(struct ath10k *ar, struct sk_buff *skb)
 	return 0;
 }
 
-static void ath10k_wmi_event_process(struct ath10k *ar, struct sk_buff *skb)
+static void ath10k_wmi_process_rx(struct ath10k *ar, struct sk_buff *skb)
 {
 	struct wmi_cmd_hdr *cmd_hdr;
 	enum wmi_event_id id;
@@ -1143,64 +1174,18 @@ static void ath10k_wmi_event_process(struct ath10k *ar, struct sk_buff *skb)
 	dev_kfree_skb(skb);
 }
 
-static void ath10k_wmi_event_work(struct work_struct *work)
-{
-	struct ath10k *ar = container_of(work, struct ath10k,
-					 wmi.wmi_event_work);
-	struct sk_buff *skb;
-
-	for (;;) {
-		skb = skb_dequeue(&ar->wmi.wmi_event_list);
-		if (!skb)
-			break;
-
-		ath10k_wmi_event_process(ar, skb);
-	}
-}
-
-static void ath10k_wmi_process_rx(struct ath10k *ar, struct sk_buff *skb)
-{
-	struct wmi_cmd_hdr *cmd_hdr = (struct wmi_cmd_hdr *)skb->data;
-	enum wmi_event_id event_id;
-
-	event_id = MS(__le32_to_cpu(cmd_hdr->cmd_id), WMI_CMD_HDR_CMD_ID);
-
-	/* some events require to be handled ASAP
-	 * thus can't be defered to a worker thread */
-	switch (event_id) {
-	case WMI_HOST_SWBA_EVENTID:
-	case WMI_MGMT_RX_EVENTID:
-		ath10k_wmi_event_process(ar, skb);
-		return;
-	default:
-		break;
-	}
-
-	skb_queue_tail(&ar->wmi.wmi_event_list, skb);
-	queue_work(ar->workqueue, &ar->wmi.wmi_event_work);
-}
-
 /* WMI Initialization functions */
 int ath10k_wmi_attach(struct ath10k *ar)
 {
 	init_completion(&ar->wmi.service_ready);
 	init_completion(&ar->wmi.unified_ready);
-	init_waitqueue_head(&ar->wmi.wq);
-
-	skb_queue_head_init(&ar->wmi.wmi_event_list);
-	INIT_WORK(&ar->wmi.wmi_event_work, ath10k_wmi_event_work);
+	init_waitqueue_head(&ar->wmi.tx_credits_wq);
 
 	return 0;
 }
 
 void ath10k_wmi_detach(struct ath10k *ar)
 {
-	/* HTC should've drained the packets already */
-	if (WARN_ON(atomic_read(&ar->wmi.pending_tx_count) > 0))
-		ath10k_warn("there are still pending packets\n");
-
-	cancel_work_sync(&ar->wmi.wmi_event_work);
-	skb_queue_purge(&ar->wmi.wmi_event_list);
 }
 
 int ath10k_wmi_connect_htc_service(struct ath10k *ar)
@@ -1215,6 +1200,7 @@ int ath10k_wmi_connect_htc_service(struct ath10k *ar)
 	/* these fields are the same for all service endpoints */
 	conn_req.ep_ops.ep_tx_complete = ath10k_wmi_htc_tx_complete;
 	conn_req.ep_ops.ep_rx_complete = ath10k_wmi_process_rx;
+	conn_req.ep_ops.ep_tx_credits = ath10k_wmi_op_ep_tx_credits;
 
 	/* connect to control service */
 	conn_req.service_id = ATH10K_HTC_SVC_ID_WMI_CONTROL;
@@ -2125,7 +2111,8 @@ int ath10k_wmi_peer_assoc(struct ath10k *ar,
 	return ath10k_wmi_cmd_send(ar, skb, WMI_PEER_ASSOC_CMDID);
 }
 
-int ath10k_wmi_beacon_send(struct ath10k *ar, const struct wmi_bcn_tx_arg *arg)
+int ath10k_wmi_beacon_send_nowait(struct ath10k *ar,
+				  const struct wmi_bcn_tx_arg *arg)
 {
 	struct wmi_bcn_tx_cmd *cmd;
 	struct sk_buff *skb;
@@ -2141,7 +2128,7 @@ int ath10k_wmi_beacon_send(struct ath10k *ar, const struct wmi_bcn_tx_arg *arg)
 	cmd->hdr.bcn_len  = __cpu_to_le32(arg->bcn_len);
 	memcpy(cmd->bcn, arg->bcn, arg->bcn_len);
 
-	return ath10k_wmi_cmd_send(ar, skb, WMI_BCN_TX_CMDID);
+	return ath10k_wmi_cmd_send_nowait(ar, skb, WMI_BCN_TX_CMDID);
 }
 
 static void ath10k_wmi_pdev_set_wmm_param(struct wmi_wmm_params *params,
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 08860c475721..2c52c23107dd 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -508,6 +508,48 @@ enum wmi_phy_mode {
 	MODE_MAX        = 14
 };
 
+static inline const char *ath10k_wmi_phymode_str(enum wmi_phy_mode mode)
+{
+	switch (mode) {
+	case MODE_11A:
+		return "11a";
+	case MODE_11G:
+		return "11g";
+	case MODE_11B:
+		return "11b";
+	case MODE_11GONLY:
+		return "11gonly";
+	case MODE_11NA_HT20:
+		return "11na-ht20";
+	case MODE_11NG_HT20:
+		return "11ng-ht20";
+	case MODE_11NA_HT40:
+		return "11na-ht40";
+	case MODE_11NG_HT40:
+		return "11ng-ht40";
+	case MODE_11AC_VHT20:
+		return "11ac-vht20";
+	case MODE_11AC_VHT40:
+		return "11ac-vht40";
+	case MODE_11AC_VHT80:
+		return "11ac-vht80";
+	case MODE_11AC_VHT20_2G:
+		return "11ac-vht20-2g";
+	case MODE_11AC_VHT40_2G:
+		return "11ac-vht40-2g";
+	case MODE_11AC_VHT80_2G:
+		return "11ac-vht80-2g";
+	case MODE_UNKNOWN:
+		/* skip */
+		break;
+
+		/* no default handler to allow compiler to check that the
+		 * enum is fully handled */
+	};
+
+	return "<unknown>";
+}
+
 #define WMI_CHAN_LIST_TAG	0x1
 #define WMI_SSID_LIST_TAG	0x2
 #define WMI_BSSID_LIST_TAG	0x3
@@ -763,14 +805,6 @@ struct wmi_service_ready_event {
 	struct wlan_host_mem_req mem_reqs[1];
 } __packed;
 
-/*
- * status consists of  upper 16 bits fo int status and lower 16 bits of
- * module ID that retuned status
- */
-#define WLAN_INIT_STATUS_SUCCESS   0x0
-#define WLAN_GET_INIT_STATUS_REASON(status)    ((status) & 0xffff)
-#define WLAN_GET_INIT_STATUS_MODULE_ID(status) (((status) >> 16) & 0xffff)
-
 #define WMI_SERVICE_READY_TIMEOUT_HZ (5*HZ)
 #define WMI_UNIFIED_READY_TIMEOUT_HZ (5*HZ)
 
@@ -3010,7 +3044,6 @@ struct wmi_force_fw_hang_cmd {
 
 #define WMI_MAX_EVENT 0x1000
 /* Maximum number of pending TXed WMI packets */
-#define WMI_MAX_PENDING_TX_COUNT 128
 #define WMI_SKB_HEADROOM sizeof(struct wmi_cmd_hdr)
 
 /* By default disable power save for IBSS */
@@ -3023,7 +3056,6 @@ int ath10k_wmi_attach(struct ath10k *ar);
 void ath10k_wmi_detach(struct ath10k *ar);
 int ath10k_wmi_wait_for_service_ready(struct ath10k *ar);
 int ath10k_wmi_wait_for_unified_ready(struct ath10k *ar);
-void ath10k_wmi_flush_tx(struct ath10k *ar);
 
 int ath10k_wmi_connect_htc_service(struct ath10k *ar);
 int ath10k_wmi_pdev_set_channel(struct ath10k *ar,
@@ -3076,7 +3108,8 @@ int ath10k_wmi_set_ap_ps_param(struct ath10k *ar, u32 vdev_id, const u8 *mac,
 			       enum wmi_ap_ps_peer_param param_id, u32 value);
 int ath10k_wmi_scan_chan_list(struct ath10k *ar,
 			      const struct wmi_scan_chan_list_arg *arg);
-int ath10k_wmi_beacon_send(struct ath10k *ar, const struct wmi_bcn_tx_arg *arg);
+int ath10k_wmi_beacon_send_nowait(struct ath10k *ar,
+				  const struct wmi_bcn_tx_arg *arg);
 int ath10k_wmi_pdev_set_wmm_params(struct ath10k *ar,
 			const struct wmi_pdev_set_wmm_params_arg *arg);
 int ath10k_wmi_request_stats(struct ath10k *ar, enum wmi_stats_id stats_id);