From 55d761b8787afa60c361c30799f3bff681e42b1b Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Wed, 4 Sep 2013 23:51:28 -0400
Subject: [PATCH 01/39] ath10k: add missing braces to
 ath10k_pci_tx_pipe_cleanup

The indentation here implies this was meant to be
a multi-statement if, but it lacks the braces.

kvalo: add "ath10k: " prefix

Signed-off-by: Dave Jones <davej@fedoraproject.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 622901d5237f..880fcd0d604f 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -1316,7 +1316,7 @@ 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)
+		if (netbuf != CE_SENDLIST_ITEM_CTXT) {
 			/*
 			 * Indicate the completion to higer layer to free
 			 * the buffer
@@ -1325,6 +1325,7 @@ static void ath10k_pci_tx_pipe_cleanup(struct ath10k_pci_pipe *pipe_info)
 			ar_pci->msg_callbacks_current.tx_completion(ar,
 								    netbuf,
 								    id);
+		}
 	}
 }
 

From a9bf05062d13b255d4231ca2637892d9b381a9d4 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Tue, 3 Sep 2013 11:43:55 +0300
Subject: [PATCH 02/39] ath10k: add trace event ath10k_htt_stats

For analysing various data path statistics in user space.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c |  5 ++++-
 drivers/net/wireless/ath/ath10k/trace.h  | 21 +++++++++++++++++++++
 2 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 9bb0ae89fdba..af31d2f6c9a1 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>
 
@@ -1198,8 +1199,10 @@ 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/trace.h b/drivers/net/wireless/ath/ath10k/trace.h
index 85e806bf7257..bf1ceb88438b 100644
--- a/drivers/net/wireless/ath/ath10k/trace.h
+++ b/drivers/net/wireless/ath/ath10k/trace.h
@@ -158,6 +158,27 @@ 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 */

From db66ea0442daaa4ee5fb8b3083bde1728ca3a9ba Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Tue, 3 Sep 2013 11:44:03 +0300
Subject: [PATCH 03/39] ath10k: implement ath10k_debug_start/stop()

Needed for the HTT stats implementation.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.c  |  6 ++++++
 drivers/net/wireless/ath/ath10k/debug.c | 10 ++++++++++
 drivers/net/wireless/ath/ath10k/debug.h | 11 +++++++++++
 3 files changed, 27 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 2dd39a82ae99..4f2b0e7f2065 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -630,6 +630,10 @@ 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 +651,7 @@ 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);
@@ -777,6 +782,7 @@ 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/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 09f535a1c767..219d4692e548 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -517,6 +517,15 @@ static const struct file_operations fops_chip_id = {
 	.llseek = default_llseek,
 };
 
+int ath10k_debug_start(struct ath10k *ar)
+{
+	return 0;
+}
+
+void ath10k_debug_stop(struct ath10k *ar)
+{
+}
+
 int ath10k_debug_create(struct ath10k *ar)
 {
 	ar->debug.debugfs_phy = debugfs_create_dir("ath10k",
@@ -541,6 +550,7 @@ int ath10k_debug_create(struct ath10k *ar)
 
 	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..9c442a82c493 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -42,6 +42,8 @@ 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 +52,15 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 				    struct wmi_stats_event *ev);
 
 #else
+int ath10k_debug_start(struct ath10k *ar)
+{
+	return 0;
+}
+
+void ath10k_debug_stop(struct ath10k *ar)
+{
+}
+
 static inline int ath10k_debug_create(struct ath10k *ar)
 {
 	return 0;

From a3d135e566e269c7cfed51355c422e7e1c0b2b39 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Tue, 3 Sep 2013 11:44:10 +0300
Subject: [PATCH 04/39] ath10k: add htt_stats_enable debugfs file

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h   |   3 +
 drivers/net/wireless/ath/ath10k/debug.c  | 111 +++++++++++++++++++++++
 drivers/net/wireless/ath/ath10k/htt.h    |   1 +
 drivers/net/wireless/ath/ath10k/htt_tx.c |  42 +++++++++
 4 files changed, 157 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 174c4b404223..22b17d6ed90e 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -246,6 +246,9 @@ 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 219d4692e548..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;
@@ -517,13 +520,115 @@ 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)
@@ -534,6 +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,
@@ -548,6 +656,9 @@ 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;
 }
 
diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index 26c78a907f97..e090902cfd84 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -1327,6 +1327,7 @@ 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_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index d4fb3875c918..b6b142046d83 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -203,6 +203,48 @@ int ath10k_htt_h2t_ver_req_msg(struct ath10k_htt *htt)
 	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);
+
+	ATH10K_SKB_CB(skb)->htt.is_conf = true;
+
+	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;
+	}
+
+	return 0;
+}
+
 int ath10k_htt_send_rx_ring_cfg_ll(struct ath10k_htt *htt)
 {
 	struct sk_buff *skb;

From 03fc137b509fa698d9337ee54f765aa13fefd319 Mon Sep 17 00:00:00 2001
From: Bartosz Markowski <bartosz.markowski@tieto.com>
Date: Tue, 3 Sep 2013 14:24:02 +0200
Subject: [PATCH 05/39] ath10k: set the UART baud rate to 19200

When configuring the host_interests over BMI, set the UART
baud rate to 19200. This is valid for QCA988X_2.0 devices.

kvalo: found during code review, there should not be any functionality
changes

Signed-off-by: Bartosz Markowski <bartosz.markowski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 4f2b0e7f2065..8a688393b380 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -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;
 }

From b8a1e00f1a9b861512b17038afda9e2f8b9f9bea Mon Sep 17 00:00:00 2001
From: Bartosz Markowski <bartosz.markowski@tieto.com>
Date: Tue, 3 Sep 2013 14:24:03 +0200
Subject: [PATCH 06/39] ath10k: remove obsolete INIT STATUS definitions

There's no functional changes. Just a small cleanup.

Signed-off-by: Bartosz Markowski <bartosz.markowski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/wmi.h | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 08860c475721..40fcf05ecc57 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -763,14 +763,6 @@ 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)
 

From 5440ce253779fd42ae3b0df64d77430274079ab7 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Tue, 3 Sep 2013 15:09:58 +0200
Subject: [PATCH 07/39] ath10k: prevent CE from looping indefinitely

The double while() could end up running forever.
Inner while() would complete very fast. However
the completion processing could take enough time
for more completions to flow in. In that case the
outer while() would not terminate and run again,
and again. This could happen especially on a slow
host system.

This could lead to a system freeze during heavy
traffic. Note: this doesn't solve all known
starvation issues yet.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/ce.c  | 57 ++++--------------
 drivers/net/wireless/ath/ath10k/ce.h  | 26 ++-------
 drivers/net/wireless/ath/ath10k/pci.c | 84 ++++++++++++---------------
 3 files changed, 52 insertions(+), 115 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index afe5df404ce6..924b29ec7a74 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -742,11 +742,6 @@ 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 +754,15 @@ 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 +853,7 @@ 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 +866,7 @@ 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);
diff --git a/drivers/net/wireless/ath/ath10k/ce.h b/drivers/net/wireless/ath/ath10k/ce.h
index cfef7d0fc5c4..fb22361bde21 100644
--- a/drivers/net/wireless/ath/ath10k/ce.h
+++ b/drivers/net/wireless/ath/ath10k/ce.h
@@ -116,17 +116,8 @@ 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;
@@ -181,11 +172,7 @@ 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. */
@@ -228,12 +215,7 @@ 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 */
diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 880fcd0d604f..0ad65a0f94ca 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -612,19 +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;
+	void *transfer_context;
+	u32 ce_data;
+	unsigned int nbytes;
+	unsigned int transfer_id;
 
-	do {
+	while (ath10k_ce_completed_send_next(ce_state, &transfer_context,
+					     &ce_data, &nbytes,
+					     &transfer_id) == 0) {
 		/*
 		 * For the send completion of an item in sendlist, just
 		 * increment num_sends_allowed. The upper layer callback will
@@ -655,38 +656,28 @@ 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 +700,7 @@ 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);
 }
@@ -1491,13 +1477,16 @@ 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;
@@ -1505,14 +1494,17 @@ 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");

From 4ed998dcbfa971e4c33edca7b2b39ce693014ba8 Mon Sep 17 00:00:00 2001
From: Bartosz Markowski <bartosz.markowski@tieto.com>
Date: Mon, 9 Sep 2013 17:50:45 +0200
Subject: [PATCH 08/39] ath10k: define ath10k_debug_start/_stop as static
 inline

Otherwise if CONFIG_ATH10K_DEBUGFS won't be set we will
end up with multiple definitions and compilation failure
in each place the header is included.

Signed-off-by: Bartosz Markowski <bartosz.markowski@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/debug.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h
index 9c442a82c493..376a56b093a7 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -52,12 +52,12 @@ void ath10k_debug_read_target_stats(struct ath10k *ar,
 				    struct wmi_stats_event *ev);
 
 #else
-int ath10k_debug_start(struct ath10k *ar)
+static inline int ath10k_debug_start(struct ath10k *ar)
 {
 	return 0;
 }
 
-void ath10k_debug_stop(struct ath10k *ar)
+static inline void ath10k_debug_stop(struct ath10k *ar)
 {
 }
 

From f0bbea9981780e767c90955f47615e6f7382746f Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:55:32 +0300
Subject: [PATCH 09/39] ath10k: add BMI log level

Also clean BMI log messages and add few more.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/bmi.c   | 42 ++++++++++++++++---------
 drivers/net/wireless/ath/ath10k/debug.h |  1 +
 2 files changed, 28 insertions(+), 15 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/debug.h b/drivers/net/wireless/ath/ath10k/debug.h
index 376a56b093a7..e40e155abd87 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -32,6 +32,7 @@ enum ath10k_debug_mask {
 	ATH10K_DBG_HTT_DUMP	= 0x00000080,
 	ATH10K_DBG_MGMT		= 0x00000100,
 	ATH10K_DBG_DATA		= 0x00000200,
+	ATH10K_DBG_BMI		= 0x00000400,
 	ATH10K_DBG_ANY		= 0xffffffff,
 };
 

From b52b7688ac85819c9c99bfe0250913fd0053e665 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:55:38 +0300
Subject: [PATCH 10/39] ath10k: rename ATH10K_DBG_CORE to BOOT

core.c mostly deals with driver and firmware starting related
actions. And we can use the boot level also in other components,
like PCI and HTT.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.c  | 6 +++---
 drivers/net/wireless/ath/ath10k/debug.h | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 8a688393b380..0ed60fd34b7f 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, "%s\n", __func__);
 
 	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, "core wmi ready\n");
 	return 0;
 
 timeout:
@@ -203,7 +203,7 @@ static int ath10k_push_board_ext_data(struct ath10k *ar,
 		return ret;
 	}
 
-	ath10k_dbg(ATH10K_DBG_CORE,
+	ath10k_dbg(ATH10K_DBG_BOOT,
 		   "ath10k: Board extended Data download addr: 0x%x\n",
 		   board_ext_data_addr);
 
diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h
index e40e155abd87..fa581486626f 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -27,7 +27,7 @@ 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,

From effea9688dae6ac1073b9f086903ddda8dda0b5a Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:55:44 +0300
Subject: [PATCH 11/39] ath10k: cleanup debug messages in core.c

Fix them to follow the general logging style in ath10k.

While at it, add print id chip_id to the debug log.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 0ed60fd34b7f..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_BOOT, "%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_BOOT, "core wmi ready\n");
+	ath10k_dbg(ATH10K_DBG_BOOT, "boot wmi ready\n");
 	return 0;
 
 timeout:
@@ -204,7 +204,7 @@ static int ath10k_push_board_ext_data(struct ath10k *ar,
 	}
 
 	ath10k_dbg(ATH10K_DBG_BOOT,
-		   "ath10k: Board extended Data download addr: 0x%x\n",
+		   "boot push board extended data addr 0x%x\n",
 		   board_ext_data_addr);
 
 	if (board_ext_data_addr == 0)
@@ -722,6 +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. */

From 24cfade11e20255f21ae1ef50591208ab3fe6a4c Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:55:50 +0300
Subject: [PATCH 12/39] ath10k: add boot debug messages to pci.c and ce.c

To unify all boot related debug messages into one debug
level.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/ce.c  | 8 ++++++++
 drivers/net/wireless/ath/ath10k/pci.c | 6 ++++--
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index 924b29ec7a74..4f8373c5a86c 100644
--- a/drivers/net/wireless/ath/ath10k/ce.c
+++ b/drivers/net/wireless/ath/ath10k/ce.c
@@ -973,6 +973,10 @@ 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;
 }
 
@@ -1054,6 +1058,10 @@ 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/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 0ad65a0f94ca..899a8c0c189c 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -2367,10 +2367,10 @@ 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;
 		}
 	}
@@ -2496,6 +2496,8 @@ 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);

From 42a2efbc578339f2e314703a818efaefb4b4af81 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:55:56 +0300
Subject: [PATCH 13/39] ath10k: add boot debug messages to htc.c

To unify the boot debug level.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htc.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 7d445d3b849d..f03fd8d282dd 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -752,8 +752,8 @@ 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 +873,19 @@ 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);
 	}
 

From aad0b65f5cb83ce0473367838a960f46577da8ac Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:56:02 +0300
Subject: [PATCH 14/39] ath10k: add boot messages to htt.c

To unify the boot debug level.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c | 2 +-
 drivers/net/wireless/ath/ath10k/htt_tx.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index af31d2f6c9a1..a39fbf4536c0 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -502,7 +502,7 @@ 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",
+	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;
 
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index b6b142046d83..454812879604 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) *

From 60c3daa88873df9c90465c09fba0b7985f9db939 Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:56:07 +0300
Subject: [PATCH 15/39] ath10k: clean mac.c debug messages

Just to unify with the rest of debug messages. Minimal functional changes,
only major ones are removal of the awkward "else" style in debug
messages.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/mac.c | 180 +++++++++++++-------------
 1 file changed, 92 insertions(+), 88 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 248248b5c196..b8c81bbb8572 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -604,7 +604,7 @@ 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 +636,7 @@ 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 +665,7 @@ 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 +749,14 @@ 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 +946,8 @@ 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 +967,7 @@ 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 +1046,8 @@ 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,
@@ -1102,6 +1104,9 @@ static void ath10k_peer_assoc_h_phymode(struct ath10k *ar,
 		break;
 	}
 
+	ath10k_dbg(ATH10K_DBG_MAC, "mac peer %pM phymode %d\n",
+		   sta->addr, phymode);
+
 	arg->peer_phymode = phymode;
 	WARN_ON(phymode == MODE_UNKNOWN);
 }
@@ -1159,15 +1164,15 @@ 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 +1193,11 @@ 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
@@ -1200,10 +1206,10 @@ static void ath10k_bss_disassoc(struct ieee80211_hw *hw,
 	 * interfaces as it expects there is no rx when no interface is
 	 * running.
 	 */
+	ath10k_dbg(ATH10K_DBG_MAC, "mac vdev %d down\n", arvif->vdev_id);
+
+	/* FIXME: why don't we print error if wmi call fails? */
 	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_wmi_flush_tx(ar);
 
@@ -1330,8 +1336,8 @@ 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 +1437,8 @@ 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 +1541,7 @@ 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 +1553,7 @@ 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);
 
@@ -1886,7 +1894,7 @@ 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;
@@ -1975,7 +1983,7 @@ 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,
@@ -2054,8 +2062,6 @@ 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);
-
 	ar->free_vdev_map |= 1 << (arvif->vdev_id);
 
 	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
@@ -2066,6 +2072,9 @@ 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 +2116,20 @@ 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 +2154,41 @@ 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 +2201,15 @@ 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 +2219,12 @@ 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 +2248,15 @@ 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 +2267,15 @@ 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 +2285,16 @@ 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 +2485,26 @@ 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 +2515,13 @@ 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 +2529,13 @@ 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 +2755,13 @@ 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 +2796,13 @@ 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)

From 38a1d47ed5cf4b55f494be44940a1fda3281feab Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 17:56:14 +0300
Subject: [PATCH 16/39] ath10k: print phymode as a string

Makes it easier to read debug logs.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/mac.c |  9 ++++--
 drivers/net/wireless/ath/ath10k/wmi.h | 42 +++++++++++++++++++++++++++
 2 files changed, 49 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index b8c81bbb8572..2b118569030f 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);
@@ -1104,8 +1109,8 @@ static void ath10k_peer_assoc_h_phymode(struct ath10k *ar,
 		break;
 	}
 
-	ath10k_dbg(ATH10K_DBG_MAC, "mac peer %pM phymode %d\n",
-		   sta->addr, phymode);
+	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);
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 40fcf05ecc57..ab46582d4eaa 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

From 7cc45e98e539ad98d8af57e4126aaac3e6b8f341 Mon Sep 17 00:00:00 2001
From: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 18:19:55 +0300
Subject: [PATCH 17/39] ath10k: Calculate correct peer PHY mode for VHT

The peer PHY mode for 11ac operation needs to be determined
properly based on the channel bandwidth being used. Fix
this so that the proper mode is given to the firmware.

kvalo: earlier we used 11na-ht20 in STA mode for 11ac AP peer, this
patch changes that to 11ac-vht80. I didn't notice any change in
throughput in my tests, but nevertheless it's the right thing
to do.

Signed-off-by: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/mac.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 2b118569030f..8b9fb660519b 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -1080,8 +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) {
@@ -1095,7 +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

From e9bb0aa39bce2f0a159c562c322c2d2746e6560d Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@qca.qualcomm.com>
Date: Sun, 8 Sep 2013 18:36:11 +0300
Subject: [PATCH 18/39] ath10k: delete struct ce_sendlist

struct ce_sendlist is useless as we always add just one buffer onto it.
And most importantly, it's ugly as it doesn't use skb properly.

Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/ce.c  | 48 +++++----------------------
 drivers/net/wireless/ath/ath10k/ce.h  | 44 +++---------------------
 drivers/net/wireless/ath/ath10k/pci.c | 40 ++++++++--------------
 3 files changed, 25 insertions(+), 107 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/ce.c b/drivers/net/wireless/ath/ath10k/ce.c
index 4f8373c5a86c..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);
diff --git a/drivers/net/wireless/ath/ath10k/ce.h b/drivers/net/wireless/ath/ath10k/ce.h
index fb22361bde21..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
 
 /*
@@ -124,24 +123,6 @@ struct ath10k_ce_pipe {
 	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;
 
@@ -175,13 +156,6 @@ void ath10k_ce_send_cb_register(struct ath10k_ce_pipe *ce_state,
 				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
@@ -193,10 +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=======================*/
 
@@ -307,16 +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/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 899a8c0c189c..ecaf21315244 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -626,17 +626,9 @@ static void ath10k_pci_ce_send_done(struct ath10k_ce_pipe *ce_state)
 	while (ath10k_ce_completed_send_next(ce_state, &transfer_context,
 					     &ce_data, &nbytes,
 					     &transfer_id) == 0) {
-		/*
-		 * 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;
-		}
+		spin_lock_bh(&pipe_info->pipe_lock);
+		pipe_info->num_sends_allowed++;
+		spin_unlock_bh(&pipe_info->pipe_lock);
 
 		compl = get_free_compl(pipe_info);
 		if (!compl)
@@ -714,13 +706,10 @@ 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;
 
@@ -735,8 +724,6 @@ 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) {
@@ -747,7 +734,8 @@ 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);
 
@@ -1302,16 +1290,14 @@ 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);
 	}
 }
 

From 3699ddc55fc1d1689a411ada9d74e584f7987f46 Mon Sep 17 00:00:00 2001
From: Jingoo Han <jg1.han@samsung.com>
Date: Tue, 10 Sep 2013 20:25:11 +0900
Subject: [PATCH 19/39] wireless: ath10k: remove unnecessary pci_set_drvdata()

The driver core clears the driver data to NULL after device_release
or on probe failure. Thus, it is not needed to manually clear the
device driver data to NULL.

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/pci.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index ecaf21315244..f1faf4691f91 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -2501,7 +2501,6 @@ err_region:
 err_device:
 	pci_disable_device(pdev);
 err_ar:
-	pci_set_drvdata(pdev, NULL);
 	ath10k_core_destroy(ar);
 err_ar_pci:
 	/* call HIF PCI free here */
@@ -2529,7 +2528,6 @@ static void ath10k_pci_remove(struct pci_dev *pdev)
 
 	ath10k_core_unregister(ar);
 
-	pci_set_drvdata(pdev, NULL);
 	pci_iounmap(pdev, ar_pci->mem);
 	pci_release_region(pdev, BAR_NUM);
 	pci_clear_master(pdev);

From eeed3765f327755b499c81eaf453e9d52a27c01c Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:52 +0200
Subject: [PATCH 20/39] ath10k: simplify HTC credits calculation

Credit calculation was overly complex
unnecessarily. Now skb dequeing is more unified.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htc.c | 58 +++++----------------------
 1 file changed, 10 insertions(+), 48 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index f03fd8d282dd..1da0f828f458 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -167,49 +167,6 @@ err:
 	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,
@@ -224,11 +181,16 @@ static void ath10k_htc_send_work(struct work_struct *work)
 			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);
+		skb = __skb_dequeue(&ep->tx_queue);
+
+		if (ep->tx_credit_flow_enabled) {
+			credits = DIV_ROUND_UP(skb->len,
+					       htc->target_credit_size);
+			if (ep->tx_credits < credits) {
+				__skb_queue_head(&ep->tx_queue, skb);
+				skb = NULL;
+			}
+		}
 		spin_unlock_bh(&htc->tx_lock);
 
 		if (!skb)

From 88e65fc33752f54dfbabc766de15f3be6989f1d1 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:53 +0200
Subject: [PATCH 21/39] ath10k: add HTC TX credits replenishing notification

This will allow higher layers to anticipate and
act upon TX credits renewal. This will be
important for some future rework of WMI command
submission.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htc.c | 9 +++++++++
 drivers/net/wireless/ath/ath10k/htc.h | 1 +
 2 files changed, 10 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 1da0f828f458..49da4e509084 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -150,6 +150,9 @@ err:
 	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);
+
 	/* 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 */
@@ -302,6 +305,12 @@ ath10k_htc_process_credit_report(struct ath10k_htc *htc,
 		ep = &htc->endpoint[report->eid];
 		ep->tx_credits += report->credits;
 
+		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);
+		}
+
 		if (ep->tx_credits && !skb_queue_empty(&ep->tx_queue))
 			queue_work(htc->ar->workqueue, &ep->send_work);
 	}
diff --git a/drivers/net/wireless/ath/ath10k/htc.h b/drivers/net/wireless/ath/ath10k/htc.h
index e1dd8c761853..92ca29bf9620 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 */

From be8b39439009579a4569598539ea97e227f99d98 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:54 +0200
Subject: [PATCH 22/39] ath10k: make WMI commands block by design

This will be necessary for further changes in
command submission scheme.

Once HTC is cleaned up WMI commands will finally
block.

This requires for SWBA to be processed in a
non-atomic context for now. Once other necessary
changes are in this will be reverted.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h  |  1 +
 drivers/net/wireless/ath/ath10k/trace.h | 11 +++--
 drivers/net/wireless/ath/ath10k/wmi.c   | 56 +++++++++++++++----------
 3 files changed, 43 insertions(+), 25 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 22b17d6ed90e..c953a3317d7b 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -114,6 +114,7 @@ struct ath10k_wmi {
 	struct completion unified_ready;
 	atomic_t pending_tx_count;
 	wait_queue_head_t wq;
+	wait_queue_head_t tx_credits_wq;
 
 	struct sk_buff_head wmi_event_list;
 	struct work_struct wmi_event_work;
diff --git a/drivers/net/wireless/ath/ath10k/trace.h b/drivers/net/wireless/ath/ath10k/trace.h
index bf1ceb88438b..fd53130376c8 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
 	)
 );
 
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 32fd5e735beb..66cd8921df27 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -90,13 +90,12 @@ static void ath10k_wmi_htc_tx_complete(struct ath10k *ar, struct sk_buff *skb)
 		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 +106,40 @@ 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);
-
-	status = ath10k_htc_send(&ar->htc, ar->wmi.eid, skb);
-	if (status) {
-		dev_kfree_skb_any(skb);
-		atomic_dec(&ar->wmi.pending_tx_count);
-		return status;
-	}
+	if (ret)
+		goto err_pull;
 
 	return 0;
+
+err_pull:
+	skb_pull(skb, sizeof(struct wmi_cmd_hdr));
+	return ret;
+}
+
+static void ath10k_wmi_op_ep_tx_credits(struct ath10k *ar,
+					enum ath10k_htc_ep_id eid)
+{
+	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, ({
+		ret = ath10k_wmi_cmd_send_nowait(ar, skb, cmd_id);
+		(ret != -EAGAIN);
+	}), 3*HZ);
+
+	if (ret)
+		dev_kfree_skb_any(skb);
+
+	return ret;
 }
 
 static int ath10k_wmi_event_scan(struct ath10k *ar, struct sk_buff *skb)
@@ -1168,7 +1181,6 @@ static void ath10k_wmi_process_rx(struct ath10k *ar, struct sk_buff *skb)
 	/* 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;
@@ -1186,6 +1198,7 @@ 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);
+	init_waitqueue_head(&ar->wmi.tx_credits_wq);
 
 	skb_queue_head_init(&ar->wmi.wmi_event_list);
 	INIT_WORK(&ar->wmi.wmi_event_work, ath10k_wmi_event_work);
@@ -1215,6 +1228,7 @@ 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;

From 12acbc43c1c302022984bf0af89ac5f0a24b133a Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:55 +0200
Subject: [PATCH 23/39] ath10k: simplify HTC command submitting

The patch removes HTC endpoint tx workers in
favour of direct command submission. This makes a
lot more sense for data path.

mac80211 queues are effectively stopped/woken up
in a more timely fashion preventing build up of
frames. It's possible to push more traffic than
the device/system is able to handle and have no
hiccups or performance degradation with UDP
traffic.

WMI commands will now report errors properly and
possibly block as they actively can wait for tx
credits to become available.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htc.c | 178 +++++++-------------------
 drivers/net/wireless/ath/ath10k/htc.h |   4 -
 drivers/net/wireless/ath/ath10k/wmi.c |   3 +-
 3 files changed, 44 insertions(+), 141 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 49da4e509084..d0d72125df81 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -117,99 +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);
-
-	if (ep->ep_ops.ep_tx_credits)
-		ep->ep_ops.ep_tx_credits(htc->ar);
-
-	/* 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 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);
-		skb = __skb_dequeue(&ep->tx_queue);
-
-		if (ep->tx_credit_flow_enabled) {
-			credits = DIV_ROUND_UP(skb->len,
-					       htc->target_credit_size);
-			if (ep->tx_credits < credits) {
-				__skb_queue_head(&ep->tx_queue, skb);
-				skb = NULL;
-			}
-		}
-		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;
@@ -219,18 +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;
 	}
-
-	__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);
+	skb_push(skb, sizeof(struct ath10k_htc_hdr));
+
+	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,
@@ -243,39 +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 */
 /***********/
@@ -310,9 +231,6 @@ ath10k_htc_process_credit_report(struct ath10k_htc *htc,
 			ep->ep_ops.ep_tx_credits(htc->ar);
 			spin_lock_bh(&htc->tx_lock);
 		}
-
-		if (ep->tx_credits && !skb_queue_empty(&ep->tx_queue))
-			queue_work(htc->ar->workqueue, &ep->send_work);
 	}
 	spin_unlock_bh(&htc->tx_lock);
 }
@@ -570,10 +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);
 	}
 }
 
@@ -916,18 +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 92ca29bf9620..4716d331e6b6 100644
--- a/drivers/net/wireless/ath/ath10k/htc.h
+++ b/drivers/net/wireless/ath/ath10k/htc.h
@@ -316,15 +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/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 66cd8921df27..ff407c2f1d2b 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -120,8 +120,7 @@ err_pull:
 	return ret;
 }
 
-static void ath10k_wmi_op_ep_tx_credits(struct ath10k *ar,
-					enum ath10k_htc_ep_id eid)
+static void ath10k_wmi_op_ep_tx_credits(struct ath10k *ar)
 {
 	wake_up(&ar->wmi.tx_credits_wq);
 }

From ed54388a38d817dce7fe22e7dc80fc13b1a6838e Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:56 +0200
Subject: [PATCH 24/39] ath10k: improve beacon submission latency

The patch prevents beacon misses in some case of
heavy load on a system.

If a beacon can't be transmitted directly from an
SWBA event it will be left in arvif->beacon and
transmission will be retried once TX credits
become available.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h |  1 +
 drivers/net/wireless/ath/ath10k/mac.c  |  7 +++
 drivers/net/wireless/ath/ath10k/wmi.c  | 74 +++++++++++++++++++++-----
 drivers/net/wireless/ath/ath10k/wmi.h  |  3 +-
 4 files changed, 71 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index c953a3317d7b..14b7d3de6883 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -204,6 +204,7 @@ 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;
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 8b9fb660519b..6c3e9d1f80d9 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -2075,6 +2075,13 @@ static void ath10k_remove_interface(struct ieee80211_hw *hw,
 
 	mutex_lock(&ar->conf_mutex);
 
+	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);
 
 	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index ff407c2f1d2b..9152daea9d96 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -120,8 +120,53 @@ err_pull:
 	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);
 }
 
@@ -131,6 +176,9 @@ static int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb,
 	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);
@@ -760,10 +808,8 @@ 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");
 
@@ -820,17 +866,17 @@ 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);
 	}
 }
 
@@ -1181,6 +1227,7 @@ static void ath10k_wmi_process_rx(struct ath10k *ar, struct sk_buff *skb)
 	 * thus can't be defered to a worker thread */
 	switch (event_id) {
 	case WMI_MGMT_RX_EVENTID:
+	case WMI_HOST_SWBA_EVENTID:
 		ath10k_wmi_event_process(ar, skb);
 		return;
 	default:
@@ -2138,7 +2185,8 @@ 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;
@@ -2154,7 +2202,7 @@ 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 ab46582d4eaa..b100431f2241 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -3110,7 +3110,8 @@ 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);

From 7cc23016366e183dcaf23afa4a0dca61ff7f787a Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:57 +0200
Subject: [PATCH 25/39] ath10k: remove wmi pending count limit

It is no longer used nor necessary since WMI
commands can block.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h |  2 --
 drivers/net/wireless/ath/ath10k/mac.c  |  9 --------
 drivers/net/wireless/ath/ath10k/wmi.c  | 32 --------------------------
 drivers/net/wireless/ath/ath10k/wmi.h  |  2 --
 4 files changed, 45 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 14b7d3de6883..c2b6a766dabd 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -112,8 +112,6 @@ 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;
 	wait_queue_head_t tx_credits_wq;
 
 	struct sk_buff_head wmi_event_list;
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 6c3e9d1f80d9..11aa13e7587f 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -1224,8 +1224,6 @@ static void ath10k_bss_disassoc(struct ieee80211_hw *hw,
 	/* FIXME: why don't we print error if wmi call fails? */
 	ret = ath10k_wmi_vdev_down(ar, arvif->vdev_id);
 
-	ath10k_wmi_flush_tx(ar);
-
 	arvif->def_wep_key_index = 0;
 }
 
@@ -1664,8 +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");
@@ -1699,10 +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);
@@ -1924,7 +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;
 }
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 9152daea9d96..b29d2b954c0e 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;
@@ -85,9 +61,6 @@ 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);
 }
 
 static int ath10k_wmi_cmd_send_nowait(struct ath10k *ar, struct sk_buff *skb,
@@ -1243,7 +1216,6 @@ 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);
 	init_waitqueue_head(&ar->wmi.tx_credits_wq);
 
 	skb_queue_head_init(&ar->wmi.wmi_event_list);
@@ -1254,10 +1226,6 @@ int ath10k_wmi_attach(struct ath10k *ar)
 
 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);
 }
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index b100431f2241..2c52c23107dd 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -3044,7 +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 */
@@ -3057,7 +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,

From 5c6c82df6b574eeb9a8db97833899dc39b1f6a6d Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Fri, 13 Sep 2013 14:16:58 +0200
Subject: [PATCH 26/39] ath10k: remove wmi event worker thread

It's not really necessary to have this processed
in a worker. There are no sleepable calls (and
actually shouldn't be).

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h |  3 --
 drivers/net/wireless/ath/ath10k/wmi.c  | 44 +-------------------------
 2 files changed, 1 insertion(+), 46 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index c2b6a766dabd..fcf94ee8fb62 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -113,9 +113,6 @@ struct ath10k_wmi {
 	struct completion service_ready;
 	struct completion unified_ready;
 	wait_queue_head_t tx_credits_wq;
-
-	struct sk_buff_head wmi_event_list;
-	struct work_struct wmi_event_work;
 };
 
 struct ath10k_peer_stat {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index b29d2b954c0e..6803ead9b9cf 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -1055,7 +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;
@@ -1174,43 +1174,6 @@ 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_MGMT_RX_EVENTID:
-	case WMI_HOST_SWBA_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)
 {
@@ -1218,16 +1181,11 @@ int ath10k_wmi_attach(struct ath10k *ar)
 	init_completion(&ar->wmi.unified_ready);
 	init_waitqueue_head(&ar->wmi.tx_credits_wq);
 
-	skb_queue_head_init(&ar->wmi.wmi_event_list);
-	INIT_WORK(&ar->wmi.wmi_event_work, ath10k_wmi_event_work);
-
 	return 0;
 }
 
 void ath10k_wmi_detach(struct ath10k *ar)
 {
-	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)

From 9e84f653ac16b718dd2305abe64a29bb3428621e Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 19 Sep 2013 09:44:09 +0200
Subject: [PATCH 27/39] ath10k: fix tracing build for ath10k_wmi_cmd

Commit be8b394390 ("ath10k: make WMI commands block by design") broke
the build if CONFIG_ATH10K_TRACING was enabled.

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/trace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath10k/trace.h b/drivers/net/wireless/ath/ath10k/trace.h
index fd53130376c8..90817ddc92ba 100644
--- a/drivers/net/wireless/ath/ath10k/trace.h
+++ b/drivers/net/wireless/ath/ath10k/trace.h
@@ -119,7 +119,7 @@ TRACE_EVENT(ath10k_wmi_cmd,
 		__field(unsigned int, id)
 		__field(size_t, buf_len)
 		__dynamic_array(u8, buf, buf_len)
-		__field(int ret)
+		__field(int, ret)
 	),
 
 	TP_fast_assign(

From 1073ab2e9b0d45f318052a3f3690405fc04eda75 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 18 Sep 2013 15:22:17 +0200
Subject: [PATCH 28/39] ath10k: fix num_sends_allowed replenishing

Commit e9bb0aa39 ("ath10k: delete struct ce_sendlist") broke
num_sends_allowed incrementing. num_sends_allowed
exceeded initial values and could overflow.

This code was supposed to replenish
num_sends_allowed for partial sendlist items (i.e.
before final sendlist item from a sendlist was
completed and could be processed by completion
handlers).

Fortunately it seems it did not cause any major breakage,
yet.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/pci.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index f1faf4691f91..dff23d97bed0 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -626,10 +626,6 @@ static void ath10k_pci_ce_send_done(struct ath10k_ce_pipe *ce_state)
 	while (ath10k_ce_completed_send_next(ce_state, &transfer_context,
 					     &ce_data, &nbytes,
 					     &transfer_id) == 0) {
-		spin_lock_bh(&pipe_info->pipe_lock);
-		pipe_info->num_sends_allowed++;
-		spin_unlock_bh(&pipe_info->pipe_lock);
-
 		compl = get_free_compl(pipe_info);
 		if (!compl)
 			break;

From 0945baf7d4245b79e01c2a790186c41fb4b27dd2 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 18 Sep 2013 14:43:18 +0200
Subject: [PATCH 29/39] ath10k: use num_pending_tx instead of msdu id bitmap

It's more efficient to simply check num_pending_tx
value instead of traversing whole bitmap of
msdu ids.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/mac.c  | 3 +--
 drivers/net/wireless/ath/ath10k/txrx.c | 2 +-
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 11aa13e7587f..5a56833c4907 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -2853,8 +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);
diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index 68b6faefd1d8..37b819675d20 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -96,7 +96,7 @@ exit:
 	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);
 	__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);
 

From 27bb178dae7b8165ffe7028fbd6f616fb7157c6d Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 18 Sep 2013 14:43:19 +0200
Subject: [PATCH 30/39] ath10k: avoid needless memset on TX path

This reduces number of memory accesses and
hopefully contributes to better performance in the
future.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htc.c    | 2 +-
 drivers/net/wireless/ath/ath10k/htt_tx.c | 4 +++-
 drivers/net/wireless/ath/ath10k/mac.c    | 4 +++-
 drivers/net/wireless/ath/ath10k/txrx.c   | 1 -
 4 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index d0d72125df81..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++;
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index 454812879604..c4bbf74815f4 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -384,9 +384,11 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	/* refcount is decremented by HTC and HTT completions until it reaches
 	 * zero and is freed */
 	skb_cb = ATH10K_SKB_CB(txdesc);
+	skb_cb->htt.is_conf = false;
 	skb_cb->htt.msdu_id = msdu_id;
 	skb_cb->htt.refcount = 2;
 	skb_cb->htt.msdu = msdu;
+	skb_cb->htt.txfrag = NULL;
 
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
@@ -505,7 +507,6 @@ 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;
 
@@ -555,6 +556,7 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	/* refcount is decremented by HTC and HTT completions until it reaches
 	 * zero and is freed */
 	skb_cb = ATH10K_SKB_CB(txdesc);
+	skb_cb->htt.is_conf = false;
 	skb_cb->htt.msdu_id = msdu_id;
 	skb_cb->htt.refcount = 2;
 	skb_cb->htt.txfrag = txfrag;
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 5a56833c4907..0d367e46ba9a 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -1757,7 +1757,9 @@ 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)->is_mapped = false;
+	ATH10K_SKB_CB(skb)->is_aborted = false;
+	ATH10K_SKB_CB(skb)->htt.is_offchan = false;
 	ATH10K_SKB_CB(skb)->htt.vdev_id = vdev_id;
 	ATH10K_SKB_CB(skb)->htt.tid = tid;
 
diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index 37b819675d20..f6fed31ac545 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -75,7 +75,6 @@ void ath10k_txrx_tx_unref(struct ath10k_htt *htt, struct sk_buff *txdesc)
 	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) {
 		ieee80211_free_txskb(htt->ar->hw, msdu);

From 0a89f8a01d0b28ca646f3aae9ffe1a7fc9d5b266 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 18 Sep 2013 14:43:20 +0200
Subject: [PATCH 31/39] ath10k: decouple HTT TX completions

Until now the all MSDU transfer related structures
were freed when all resources were unreferenced.

Now HTC transfer is freed independently and HTT
transfer is so too.

This yields a way more simpler ath10k_skb_cb and
should possibly enable parallel pipe processing
(which is now serialized in
ath10k_pci_process_ce routine).

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h   |  8 ---
 drivers/net/wireless/ath/ath10k/htt_rx.c |  4 +-
 drivers/net/wireless/ath/ath10k/htt_tx.c | 64 +++---------------------
 drivers/net/wireless/ath/ath10k/mac.c    |  2 -
 drivers/net/wireless/ath/ath10k/txrx.c   | 52 ++++++-------------
 drivers/net/wireless/ath/ath10k/txrx.h   |  5 +-
 6 files changed, 27 insertions(+), 108 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index fcf94ee8fb62..4563f800a291 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -52,18 +52,10 @@ 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 */
 } __packed;
 
 static inline struct ath10k_skb_cb *ATH10K_SKB_CB(struct sk_buff *skb)
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index a39fbf4536c0..62ea9c8f80c4 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -1140,7 +1140,7 @@ 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: {
@@ -1174,7 +1174,7 @@ 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;
 	}
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index c4bbf74815f4..06946d2191aa 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -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)
@@ -192,8 +168,6 @@ 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);
@@ -233,8 +207,6 @@ int ath10k_htt_h2t_stats_req(struct ath10k_htt *htt, u8 mask, u64 cookie)
 	req->cookie_lsb = cpu_to_le32(cookie & 0xffffffff);
 	req->cookie_msb = cpu_to_le32((cookie & 0xffffffff00000000ULL) >> 32);
 
-	ATH10K_SKB_CB(skb)->htt.is_conf = true;
-
 	ret = ath10k_htc_send(&htt->ar->htc, htt->eid, skb);
 	if (ret) {
 		ath10k_warn("failed to send htt type stats request: %d", ret);
@@ -321,8 +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);
@@ -335,7 +305,6 @@ 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;
@@ -364,7 +333,7 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 		res = msdu_id;
 		goto err;
 	}
-	htt->pending_tx[msdu_id] = txdesc;
+	htt->pending_tx[msdu_id] = msdu;
 	spin_unlock_bh(&htt->tx_lock);
 
 	res = ath10k_skb_map(dev, msdu);
@@ -381,15 +350,6 @@ 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.is_conf = false;
-	skb_cb->htt.msdu_id = msdu_id;
-	skb_cb->htt.refcount = 2;
-	skb_cb->htt.msdu = msdu;
-	skb_cb->htt.txfrag = NULL;
-
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
 		goto err;
@@ -417,7 +377,6 @@ 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 sk_buff *txdesc = NULL;
 	struct sk_buff *txfrag = NULL;
 	u8 vdev_id = ATH10K_SKB_CB(msdu)->htt.vdev_id;
@@ -470,7 +429,7 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 		res = msdu_id;
 		goto err;
 	}
-	htt->pending_tx[msdu_id] = txdesc;
+	htt->pending_tx[msdu_id] = msdu;
 	spin_unlock_bh(&htt->tx_lock);
 
 	res = ath10k_skb_map(dev, msdu);
@@ -553,15 +512,6 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 	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.is_conf = false;
-	skb_cb->htt.msdu_id = msdu_id;
-	skb_cb->htt.refcount = 2;
-	skb_cb->htt.txfrag = txfrag;
-	skb_cb->htt.msdu = msdu;
-
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
 		goto err;
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 0d367e46ba9a..9112e6d6f75c 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -1757,8 +1757,6 @@ static void ath10k_tx(struct ieee80211_hw *hw,
 		ath10k_tx_h_seq_no(skb);
 	}
 
-	ATH10K_SKB_CB(skb)->is_mapped = false;
-	ATH10K_SKB_CB(skb)->is_aborted = false;
 	ATH10K_SKB_CB(skb)->htt.is_offchan = false;
 	ATH10K_SKB_CB(skb)->htt.vdev_id = vdev_id;
 	ATH10K_SKB_CB(skb)->htt.tid = tid;
diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index f6fed31ac545..15395afae957 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -44,21 +44,25 @@ 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 sk_buff *msdu, *txfrag;
 	int ret;
 
-	if (ATH10K_SKB_CB(txdesc)->htt.refcount == 0)
-		return;
+	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);
 
-	ATH10K_SKB_CB(txdesc)->htt.refcount--;
-
-	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;
+	}
+
+	msdu = htt->pending_tx[tx_done->msdu_id];
+	txfrag = ATH10K_SKB_CB(msdu)->htt.txfrag;
 
 	if (txfrag) {
 		ret = ath10k_skb_unmap(dev, txfrag);
@@ -76,7 +80,7 @@ void ath10k_txrx_tx_unref(struct ath10k_htt *htt, struct sk_buff *txdesc)
 
 	info = IEEE80211_SKB_CB(msdu);
 
-	if (ATH10K_SKB_CB(txdesc)->htt.discard) {
+	if (tx_done->discard) {
 		ieee80211_free_txskb(htt->ar->hw, msdu);
 		goto exit;
 	}
@@ -84,7 +88,7 @@ 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);
@@ -92,36 +96,12 @@ 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 (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[] = {
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,

From 2f3773bcaf9fbf3ddb9d4315e3a3ca5e4b376cef Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 18 Sep 2013 14:43:21 +0200
Subject: [PATCH 32/39] ath10k: cleanup HTT TX functions

Use a saner goto scheme for failure handling. Also
group operations more sensibly.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt_tx.c | 143 +++++++++++------------
 1 file changed, 69 insertions(+), 74 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index 06946d2191aa..5e738d80138f 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -315,30 +315,30 @@ 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;
 	}
+	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;
@@ -352,22 +352,22 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 	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;
 }
 
@@ -379,6 +379,7 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)msdu->data;
 	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;
@@ -390,7 +391,17 @@ 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);
@@ -401,46 +412,34 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	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)) {
+	use_frags = htt->target_version_major < 3 ||
+		    !ieee80211_is_mgmt(hdr->frame_control);
+
+	if (use_frags) {
 		txfrag = dev_alloc_skb(frag_len);
 		if (!txfrag) {
 			res = -ENOMEM;
-			goto err;
+			goto err_free_txdesc;
 		}
 	}
 
 	if (!IS_ALIGNED((unsigned long)txdesc->data, 4)) {
 		ath10k_warn("htt alignment check failed. dropping packet.\n");
 		res = -EIO;
-		goto err;
+		goto err_free_txfrag;
 	}
 
-	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;
-	}
-	htt->pending_tx[msdu_id] = msdu;
-	spin_unlock_bh(&htt->tx_lock);
-
 	res = ath10k_skb_map(dev, msdu);
 	if (res)
-		goto err;
+		goto err_free_txfrag;
 
-	/* 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)) {
+	if (use_frags) {
 		/* tx fragment list must be terminated with zero-entry */
 		skb_put(txfrag, frag_len);
 		tx_frags = (struct htt_data_tx_desc_frag *)txfrag->data;
@@ -451,7 +450,7 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 		res = ath10k_skb_map(dev, txfrag);
 		if (res)
-			goto err;
+			goto err_unmap_msdu;
 
 		ath10k_dbg(ATH10K_DBG_HTT, "txfrag 0x%llx\n",
 			   (unsigned long long) ATH10K_SKB_CB(txfrag)->paddr);
@@ -476,15 +475,11 @@ 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;
@@ -493,14 +488,10 @@ 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
+	if (use_frags)
 		frags_paddr = ATH10K_SKB_CB(txfrag)->paddr;
+	else
+		frags_paddr = ATH10K_SKB_CB(msdu)->paddr;
 
 	cmd->hdr.msg_type        = HTT_H2T_MSG_TYPE_TX_FRM;
 	cmd->data_tx.flags0      = flags0;
@@ -514,23 +505,27 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
-		goto err;
+		goto err_restore;
 
 	return 0;
-err:
-	if (txfrag)
+
+err_restore:
+	if (use_frags)
 		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_free_txfrag:
+	if (use_frags)
+		dev_kfree_skb_any(txfrag);
+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;
 }

From 1f8bb1518eee321d94477ca7bcbb153c47d43ba4 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 18 Sep 2013 14:43:22 +0200
Subject: [PATCH 33/39] ath10k: use msdu headroom to store txfrag

Instead of allocating sk_buff for a mere 16-byte
tx fragment list buffer use headroom of the
original msdu sk_buff.

This decreases CPU cache pressure and improves
performance.

Measured improvement on AP135 is 560mbps ->
590mbps of UDP TX briding traffic.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/core.h   |  4 +-
 drivers/net/wireless/ath/ath10k/htt_tx.c | 82 ++++++++++++------------
 drivers/net/wireless/ath/ath10k/mac.c    |  4 ++
 drivers/net/wireless/ath/ath10k/txrx.c   | 16 ++---
 4 files changed, 53 insertions(+), 53 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 4563f800a291..292ad4577c98 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -54,7 +54,9 @@ struct ath10k_skb_cb {
 		u8 vdev_id;
 		u8 tid;
 		bool is_offchan;
-		struct sk_buff *txfrag;
+
+		u8 frag_len;
+		u8 pad_len;
 	} __packed htt;
 } __packed;
 
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index 5e738d80138f..3b93c6a01c6c 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -307,7 +307,8 @@ int ath10k_htt_mgmt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	struct device *dev = htt->ar->dev;
 	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;
@@ -350,6 +351,9 @@ 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));
 
+	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_unmap_msdu;
@@ -377,13 +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 = 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;
@@ -407,7 +410,6 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	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) {
@@ -421,41 +423,44 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	use_frags = htt->target_version_major < 3 ||
 		    !ieee80211_is_mgmt(hdr->frame_control);
 
-	if (use_frags) {
-		txfrag = dev_alloc_skb(frag_len);
-		if (!txfrag) {
-			res = -ENOMEM;
-			goto err_free_txdesc;
-		}
-	}
-
 	if (!IS_ALIGNED((unsigned long)txdesc->data, 4)) {
 		ath10k_warn("htt alignment check failed. dropping packet.\n");
 		res = -EIO;
-		goto err_free_txfrag;
+		goto err_free_txdesc;
+	}
+
+	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;
 	}
 
 	res = ath10k_skb_map(dev, msdu);
 	if (res)
-		goto err_free_txfrag;
+		goto err_pull_txfrag;
 
 	if (use_frags) {
+		dma_sync_single_for_cpu(dev, skb_cb->paddr, msdu->len,
+					DMA_TO_DEVICE);
+
 		/* 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_unmap_msdu;
-
-		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",
@@ -488,35 +493,28 @@ 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;
 
-	if (use_frags)
-		frags_paddr = ATH10K_SKB_CB(txfrag)->paddr;
-	else
-		frags_paddr = ATH10K_SKB_CB(msdu)->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);
+	memcpy(cmd->data_tx.prefetch, hdr, prefetch_len);
 
 	res = ath10k_htc_send(&htt->ar->htc, htt->eid, txdesc);
 	if (res)
-		goto err_restore;
+		goto err_unmap_msdu;
 
 	return 0;
 
-err_restore:
-	if (use_frags)
-		ath10k_skb_unmap(dev, txfrag);
 err_unmap_msdu:
 	ath10k_skb_unmap(dev, msdu);
-err_free_txfrag:
-	if (use_frags)
-		dev_kfree_skb_any(txfrag);
+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:
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 9112e6d6f75c..99a9bad3f398 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -3342,6 +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/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index 15395afae957..57931d0fae4b 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -49,7 +49,8 @@ void ath10k_txrx_tx_unref(struct ath10k_htt *htt,
 {
 	struct device *dev = htt->ar->dev;
 	struct ieee80211_tx_info *info;
-	struct sk_buff *msdu, *txfrag;
+	struct ath10k_skb_cb *skb_cb;
+	struct sk_buff *msdu;
 	int ret;
 
 	ath10k_dbg(ATH10K_DBG_HTT, "htt tx completion msdu_id %u discard %d no_ack %d\n",
@@ -62,20 +63,15 @@ void ath10k_txrx_tx_unref(struct ath10k_htt *htt,
 	}
 
 	msdu = htt->pending_tx[tx_done->msdu_id];
-	txfrag = ATH10K_SKB_CB(msdu)->htt.txfrag;
-
-	if (txfrag) {
-		ret = ath10k_skb_unmap(dev, txfrag);
-		if (ret)
-			ath10k_warn("txfrag unmap failed (%d)\n", ret);
-
-		dev_kfree_skb_any(txfrag);
-	}
+	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);

From f6dc2095a0da8906ec65be98cfc508aaadcf5d28 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 26 Sep 2013 10:12:22 +0300
Subject: [PATCH 34/39] ath10k: report A-MSDU subframes individually

HW reports each A-MSDU subframe as a separate
sk_buff. It is impossible to configure it to
behave differently.

Until now ath10k was reconstructing A-MSDUs from
subframes which involved a lot of memory
operations. This proved to be a significant
contributor to degraded RX performance.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c | 204 +++++++++--------------
 drivers/net/wireless/ath/ath10k/txrx.c   |   2 +
 2 files changed, 84 insertions(+), 122 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 62ea9c8f80c4..617df96e4ffd 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -41,6 +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;
@@ -591,136 +595,111 @@ 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];
 	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;
 
+		skb->ip_summed = ath10k_htt_rx_get_csum_state(skb);
+
+		/* First frame in an A-MSDU chain has more decapped data. */
 		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);
+			len = round_up(ieee80211_hdrlen(hdr->frame_control), 4);
+			len += round_up(ath10k_htt_rx_crypto_param_len(enctype),
+					4);
+			decap_hdr += len;
 		}
 
-		/* 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. */
+		switch (fmt) {
+		case RX_MSDU_DECAP_RAW:
+			skb_trim(skb, skb->len - FCS_LEN);
+			break;
+		case RX_MSDU_DECAP_NATIVE_WIFI:
+			break;
+		case RX_MSDU_DECAP_ETHERNET2_DIX:
+			len = 0;
+			len += sizeof(struct rfc1042_hdr);
+			len += sizeof(struct amsdu_subframe_hdr);
 
-		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);
-		}
-
-		if (fmt == RX_MSDU_DECAP_NATIVE_WIFI) {
-			/* Native Wifi decap inserts regular 802.11 header
-			 * in place of A-MSDU subframe header. */
-			hdr = (struct ieee80211_hdr *)skb->data;
-			skb_pull(skb, ieee80211_hdrlen(hdr->frame_control));
-
-			/* A-MSDU subframe header length */
-			decap_len += 6 + 6 + 2;
-
-			memcpy(skb_put(amsdu, decap_len), decap_hdr, decap_len);
-		}
-
-		if (fmt == RX_MSDU_DECAP_RAW)
-			skb_trim(skb, skb->len - 4); /* remove FCS */
-
-		memcpy(skb_put(amsdu, skb->len), skb->data, skb->len);
-
-		/* 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:
+			memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
+			break;
 		}
 
+		info->skb = skb;
+		info->encrypt_type = enctype;
 		skb = skb->next;
+		info->skb->next = NULL;
+
+		ath10k_process_rx(htt->ar, info);
 	}
 
-	info->skb = amsdu;
-	info->encrypt_type = enctype;
-
-	ath10k_htt_rx_free_msdu_chain(first);
-
-	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;
@@ -742,6 +721,8 @@ static int ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 			RX_MPDU_START_INFO0_ENCRYPT_TYPE);
 	hdr = (void *)skb->data - RX_HTT_HDR_STATUS_LEN;
 
+	skb->ip_summed = ath10k_htt_rx_get_csum_state(skb);
+
 	switch (fmt) {
 	case RX_MSDU_DECAP_RAW:
 		/* remove trailing FCS */
@@ -782,7 +763,8 @@ static int ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 
 	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)
@@ -854,8 +836,6 @@ 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));
 
@@ -930,11 +910,6 @@ 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;
@@ -947,24 +922,9 @@ 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);
 		}
 	}
 
diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c
index 57931d0fae4b..5ae373a1e294 100644
--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c
@@ -268,6 +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);
 }

From 26d1e9c2610c36689d48f7167ddc007e970d5489 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 26 Sep 2013 10:12:23 +0300
Subject: [PATCH 35/39] ath10k: document decap modes

Clarify how each decap mode works in one place.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/rx_desc.h | 24 +++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

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
 };
 

From e3fbf8d22a7ba109d1542c50c063acc23ef818ff Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 26 Sep 2013 10:12:23 +0300
Subject: [PATCH 36/39] ath10k: cleanup RX decap handling

Simplify decapping code and make it easier to
understand.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c | 54 ++++++++++++------------
 1 file changed, 28 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 617df96e4ffd..cdb4e6da7cc2 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -669,11 +669,16 @@ static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
 
 		switch (fmt) {
 		case RX_MSDU_DECAP_RAW:
+			/* remove trailing FCS */
 			skb_trim(skb, skb->len - FCS_LEN);
 			break;
 		case RX_MSDU_DECAP_NATIVE_WIFI:
+			/* nothing to do */
 			break;
 		case RX_MSDU_DECAP_ETHERNET2_DIX:
+			/* strip ethernet header and insert decapped 802.11
+			 * header, amsdu subframe header and rfc1042 header */
+
 			len = 0;
 			len += sizeof(struct rfc1042_hdr);
 			len += sizeof(struct amsdu_subframe_hdr);
@@ -683,6 +688,8 @@ static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
 			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;
 		}
@@ -706,6 +713,8 @@ static void ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 	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) {
@@ -719,48 +728,41 @@ static void 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 */
 		break;
 	case RX_MSDU_DECAP_ETHERNET2_DIX:
-		/* macaddr[6] + macaddr[6] + ethertype[2] */
-		skb_pull(skb, 6 + 6 + 2);
+		/* strip ethernet header and insert decapped 802.11 header and
+		 * rfc1042 header */
+
+		rfc1042 = hdr;
+		rfc1042 += roundup(hdr_len, 4);
+		rfc1042 += roundup(ath10k_htt_rx_crypto_param_len(enctype), 4);
+
+		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:
-		/* macaddr[6] + macaddr[6] + len[2] */
-		/* we don't need this for non-A-MSDU */
-		skb_pull(skb, 6 + 6 + 2);
+		/* remove A-MSDU subframe header and insert
+		 * decapped 802.11 header. rfc1042 header is already there */
+
+		skb_pull(skb, sizeof(struct amsdu_subframe_hdr));
+		memcpy(skb_push(skb, hdr_len), hdr, hdr_len);
 		break;
 	}
 
-	if (fmt == RX_MSDU_DECAP_ETHERNET2_DIX) {
-		void *llc;
-		int llclen;
-
-		llclen = 8;
-		llc  = hdr;
-		llc += roundup(ieee80211_hdrlen(hdr->frame_control), 4);
-		llc += roundup(ath10k_htt_rx_crypto_param_len(enctype), 4);
-
-		skb_push(skb, llclen);
-		memcpy(skb->data, llc, llclen);
-	}
-
-	if (fmt >= RX_MSDU_DECAP_ETHERNET2_DIX) {
-		int len = ieee80211_hdrlen(hdr->frame_control);
-		skb_push(skb, len);
-		memcpy(skb->data, hdr, len);
-	}
-
 	info->skb = skb;
 	info->encrypt_type = enctype;
 

From 784f69d31bd57973ef1dab9f5110018e38d92cf3 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 26 Sep 2013 10:12:23 +0300
Subject: [PATCH 37/39] ath10k: fix Native Wifi decap mode RX

NWifi decap mode always reports 802.11 Data
Frames, even when QoS Data Frames are actually
received.

This made mac80211 not report frame priority
properly (since there was no QoS Control field).

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt_rx.c | 32 +++++++++++++++++++++---
 1 file changed, 29 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index cdb4e6da7cc2..c4637f18734f 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -618,7 +618,7 @@ static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
 	enum rx_msdu_decap_format fmt;
 	enum htt_rx_mpdu_encrypt_type enctype;
 	struct ieee80211_hdr *hdr;
-	u8 hdr_buf[64];
+	u8 hdr_buf[64], addr[ETH_ALEN], *qos;
 	unsigned int hdr_len;
 
 	rxd = (void *)skb->data - sizeof(*rxd);
@@ -673,7 +673,25 @@ static void ath10k_htt_rx_amsdu(struct ath10k_htt *htt,
 			skb_trim(skb, skb->len - FCS_LEN);
 			break;
 		case RX_MSDU_DECAP_NATIVE_WIFI:
-			/* nothing to do */
+			/* pull decapped header and copy DA */
+			hdr = (struct ieee80211_hdr *)skb->data;
+			hdr_len = ieee80211_hdrlen(hdr->frame_control);
+			memcpy(addr, ieee80211_get_DA(hdr), ETH_ALEN);
+			skb_pull(skb, hdr_len);
+
+			/* 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);
+
+			/* 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;
+
+			/* 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
@@ -739,7 +757,15 @@ static void ath10k_htt_rx_msdu(struct ath10k_htt *htt, struct htt_rx_info *info)
 		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:
 		/* strip ethernet header and insert decapped 802.11 header and

From 4d316c79a5dcc13dea8110f0fcf295a17b4b625b Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 26 Sep 2013 10:12:24 +0300
Subject: [PATCH 38/39] ath10k: align RX frames properly

Ethernet-like decapping mode leaves IP protocol
frame not aligned to 4-byte boundaries. This leads
to re-aligning in mac80211 which in turn leads to
poor CPU cache behaviour on some machines.

Since HW doesn't allow to change payload offset
properly the solution is to force HW to decap in
Native Wifi mode which always has 24-bytes long
802.11 header (even for QoS frames). This means IP
frame is properly aligned in this decap mode.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/hw.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

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

From 6e712d427cb0542afdd5220edb6e4f4f8a5b952d Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Tue, 24 Sep 2013 10:18:36 +0200
Subject: [PATCH 39/39] ath10k: replenish HTT RX buffers in a tasklet

This starves FW RX ring buffer in case of
excessive RX. This prevents from CPU being
overwhelmed by RX indications/completions by
naturally forbiddin FW to submit more RX.

This fixes RX starvation on slow machines when
under heavy RX traffic.

kvalo: remove extra newline

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
---
 drivers/net/wireless/ath/ath10k/htt.h    |  6 ++++
 drivers/net/wireless/ath/ath10k/htt_rx.c | 35 ++++++++++++++++++++++--
 2 files changed, 38 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index e090902cfd84..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
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index c4637f18734f..90d4f74c28d7 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -182,10 +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) {
 		/*
@@ -196,6 +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);
 }
@@ -217,6 +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 =
@@ -446,6 +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;
@@ -506,6 +532,9 @@ 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;
 
+	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;
@@ -956,7 +985,7 @@ static void ath10k_htt_rx_handler(struct ath10k_htt *htt,
 		}
 	}
 
-	ath10k_htt_rx_msdu_buff_replenish(htt);
+	tasklet_schedule(&htt->rx_replenish_task);
 }
 
 static void ath10k_htt_rx_frag_handler(struct ath10k_htt *htt,