diff mbox series

[V6,1/2] ath11k: switch to using ieee80211_tx_status_ext()

Message ID 20200803145444.473810-1-john@phrozen.org (mailing list archive)
State Changes Requested
Delegated to: Kalle Valo
Headers show
Series [V6,1/2] ath11k: switch to using ieee80211_tx_status_ext() | expand

Commit Message

John Crispin Aug. 3, 2020, 2:54 p.m. UTC
This allows us to pass HE rates down into the stack.

Signed-off-by: John Crispin <john@phrozen.org>
---
Changes in V5
* fix sparse warnings

 drivers/net/wireless/ath/ath11k/dp_tx.c | 24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)

Comments

Kalle Valo Aug. 19, 2020, 5:34 p.m. UTC | #1
John Crispin <john@phrozen.org> writes:

> This allows us to pass HE rates down into the stack.
>
> Signed-off-by: John Crispin <john@phrozen.org>

On what hardware and firmware was this tested?
John Crispin Aug. 20, 2020, 5:16 a.m. UTC | #2
On 19.08.20 19:34, Kalle Valo wrote:
> John Crispin <john@phrozen.org> writes:
>
>> This allows us to pass HE rates down into the stack.
>>
>> Signed-off-by: John Crispin <john@phrozen.org>
> On what hardware and firmware was this tested?
>
Miles can give more insight, the patch was written by him, I just 
rebased it, removed the checkpatch warnings and did some basic testing.

     John
Kalle Valo Aug. 26, 2020, 2:58 p.m. UTC | #3
John Crispin <john@phrozen.org> wrote:

> This allows us to pass HE rates down into the stack.
> 
> Signed-off-by: John Crispin <john@phrozen.org>
> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

Dropping the per and expecting Miles will address the issues. Miles, as I did
some changes to the patches in pending branch please incorporate my changes in
v7:

https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=pending&id=52a632ec5dd0eab7e82051164d42bd35fe06af00

https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=pending&id=0ed5b5fc10d08b565183008a14267779d6da4bd4

2 patches set to Changes Requested.

11698305 [V6,1/2] ath11k: switch to using ieee80211_tx_status_ext()
11698303 [V6,2/2] ath11k: add more HE fields into radiotap header
diff mbox series

Patch

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 1af76775b1a8..a1f16f589b98 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -393,9 +393,12 @@  static void ath11k_dp_tx_complete_msdu(struct ath11k *ar,
 				       struct sk_buff *msdu,
 				       struct hal_tx_status *ts)
 {
+	struct ieee80211_tx_status status = { };
 	struct ath11k_base *ab = ar->ab;
 	struct ieee80211_tx_info *info;
 	struct ath11k_skb_cb *skb_cb;
+	struct ath11k_peer *peer;
+	struct ath11k_sta *arsta;
 
 	if (WARN_ON_ONCE(ts->buf_rel_source != HAL_WBM_REL_SRC_MODULE_TQM)) {
 		/* Must not happen */
@@ -459,12 +462,23 @@  static void ath11k_dp_tx_complete_msdu(struct ath11k *ar,
 		ath11k_dp_tx_cache_peer_stats(ar, msdu, ts);
 	}
 
-	/* NOTE: Tx rate status reporting. Tx completion status does not have
-	 * necessary information (for example nss) to build the tx rate.
-	 * Might end up reporting it out-of-band from HTT stats.
-	 */
 
-	ieee80211_tx_status(ar->hw, msdu);
+	spin_lock_bh(&ab->base_lock);
+	peer = ath11k_peer_find_by_id(ab, ts->peer_id);
+	if (peer) {
+		arsta = (struct ath11k_sta *)peer->sta->drv_priv;
+		status.sta = peer->sta;
+		status.skb = msdu;
+		status.info = info;
+		status.rate = &arsta->last_txrate;
+	}
+	rcu_read_unlock();
+	if (peer)
+		ieee80211_tx_status_ext(ar->hw, &status);
+	else
+		dev_kfree_skb_any(msdu);
+	spin_unlock_bh(&ab->base_lock);
+	return;
 
 exit:
 	rcu_read_unlock();