diff mbox

ath10k: Fix mgmt tx status for 10.4.3 firmware.

Message ID 1456782654-24723-1-git-send-email-greearb@candelatech.com (mailing list archive)
State Not Applicable
Delegated to: Kalle Valo
Headers show

Commit Message

Ben Greear Feb. 29, 2016, 9:50 p.m. UTC
From: Ben Greear <greearb@candelatech.com>

When testing a 10.4.3 firmware in station mode, I notice that
when the AP is powered down, the ath10k does not notice AP is gone
because mgt frames get tx status of 3, which is not handled.

It appears that status 3 means something similar to failed-retry.

Treating it thus lets the station disconnect properly.

Tested against a non-stock 10.4.3 firmware, but likely upstream
firmware acts similarly in this case.

Signed-off-by: Ben Greear <greearb@candelatech.com>
---

This patch is against a heavily patched tree based on 4.4, hopefully
it can be applied to upstream code w/out too much trouble.

Someone should verify this on stock 10.4-ish firmware before applying.

 drivers/net/wireless/ath/ath10k/htt.h    |  4 +++-
 drivers/net/wireless/ath/ath10k/htt_rx.c | 11 +++++++++--
 2 files changed, 12 insertions(+), 3 deletions(-)

Comments

Felix Fietkau March 1, 2016, 9:26 a.m. UTC | #1
On 2016-02-29 22:50, greearb@candelatech.com wrote:
> From: Ben Greear <greearb@candelatech.com>
> 
> When testing a 10.4.3 firmware in station mode, I notice that
> when the AP is powered down, the ath10k does not notice AP is gone
> because mgt frames get tx status of 3, which is not handled.
> 
> It appears that status 3 means something similar to failed-retry.
> 
> Treating it thus lets the station disconnect properly.
> 
> Tested against a non-stock 10.4.3 firmware, but likely upstream
> firmware acts similarly in this case.
> 
> Signed-off-by: Ben Greear <greearb@candelatech.com>
> ---
> 
> This patch is against a heavily patched tree based on 4.4, hopefully
> it can be applied to upstream code w/out too much trouble.
> 
> Someone should verify this on stock 10.4-ish firmware before applying.
> 
>  drivers/net/wireless/ath/ath10k/htt.h    |  4 +++-
>  drivers/net/wireless/ath/ath10k/htt_rx.c | 11 +++++++++--
>  2 files changed, 12 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
> index c31a31f..de663a6 100644
> --- a/drivers/net/wireless/ath/ath10k/htt.h
> +++ b/drivers/net/wireless/ath/ath10k/htt.h
> @@ -309,7 +309,9 @@ struct htt_mgmt_tx_desc {
>  enum htt_mgmt_tx_status {
>  	HTT_MGMT_TX_STATUS_OK    = 0,
>  	HTT_MGMT_TX_STATUS_RETRY = 1,
> -	HTT_MGMT_TX_STATUS_DROP  = 2
> +	HTT_MGMT_TX_STATUS_DROP  = 2,
> +	HTT_MGMT_TX_STATUS_TXFILT = 3 /* Seems to be logically similar to
> +					 RETRY failure. */
In that case the driver should probably set IEEE80211_TX_STAT_TX_FILTERED

- Felix
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Kalle Valo June 16, 2016, 1:57 p.m. UTC | #2
greearb@candelatech.com writes:

> From: Ben Greear <greearb@candelatech.com>
>
> When testing a 10.4.3 firmware in station mode, I notice that
> when the AP is powered down, the ath10k does not notice AP is gone
> because mgt frames get tx status of 3, which is not handled.
>
> It appears that status 3 means something similar to failed-retry.
>
> Treating it thus lets the station disconnect properly.
>
> Tested against a non-stock 10.4.3 firmware, but likely upstream
> firmware acts similarly in this case.
>
> Signed-off-by: Ben Greear <greearb@candelatech.com>
> ---
>
> This patch is against a heavily patched tree based on 4.4, hopefully
> it can be applied to upstream code w/out too much trouble.
>
> Someone should verify this on stock 10.4-ish firmware before applying.

Has anyone tested this with stock driver and stock firmware?

Full patch here:

https://patchwork.kernel.org/patch/8460831/
diff mbox

Patch

diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index c31a31f..de663a6 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -309,7 +309,9 @@  struct htt_mgmt_tx_desc {
 enum htt_mgmt_tx_status {
 	HTT_MGMT_TX_STATUS_OK    = 0,
 	HTT_MGMT_TX_STATUS_RETRY = 1,
-	HTT_MGMT_TX_STATUS_DROP  = 2
+	HTT_MGMT_TX_STATUS_DROP  = 2,
+	HTT_MGMT_TX_STATUS_TXFILT = 3 /* Seems to be logically similar to
+					 RETRY failure. */
 };
 
 /*=== target -> host messages ===============================================*/
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 2953aa7..88918ab 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -1721,8 +1721,10 @@  static void ath10k_htt_rx_frm_tx_compl(struct ath10k *ar,
 		break;
 	}
 
-	ath10k_dbg(ar, ATH10K_DBG_HTT, "htt tx completion num_msdus %d\n",
-		   resp->data_tx_completion.num_msdus);
+	ath10k_dbg(ar, ATH10K_DBG_HTT,
+		   "htt tx completion num_msdus %d status: %d  discard: %d  no-ack: %d\n",
+		   resp->data_tx_completion.num_msdus, status,
+		   (int)tx_done.discard, (int)tx_done.no_ack);
 
 	if (test_bit(ATH10K_FW_FEATURE_TXRATE_CT, ar->fw_features) &&
 	    ar->wmi.op_version != ATH10K_FW_WMI_OP_VERSION_10_4) {
@@ -2406,6 +2408,7 @@  void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 			tx_done.success = true;
 			break;
 		case HTT_MGMT_TX_STATUS_RETRY:
+		case HTT_MGMT_TX_STATUS_TXFILT:
 			tx_done.no_ack = true;
 			break;
 		case HTT_MGMT_TX_STATUS_DROP:
@@ -2413,6 +2416,10 @@  void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 			break;
 		}
 
+		ath10k_dbg(ar, ATH10K_DBG_HTT,
+			   "mgmt-tx-completion, status: %d  discard: %d  no-ack: %d\n",
+			   status, (int)tx_done.discard, (int)tx_done.no_ack);
+
 		ath10k_txrx_tx_unref(htt, &tx_done);
 		ath10k_mac_tx_push_pending(ar);
 		break;