The tid of 11a station with WMM disable reported by FW is 0x10 in
tx completion. The tid 16 is mapped to a NULL txq since buffer
MMPDU capbility is not supported. Then 11a station's airtime will
not be registered due to NULL txq check. As a results, airtime of
11a station keeps unchanged in debugfs system.
Mask the tid along with IEEE80211_QOS_CTL_TID_MASK to make it in
the valid range.
Hardwares tested : QCA9984
Firmwares tested : 10.4-3.10-00047
Signed-off-by: Yibo Zhao <[email protected]>
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 38a5814..f883f2a 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2744,7 +2744,8 @@ static void ath10k_htt_rx_tx_compl_ind(struct ath10k *ar,
continue;
}
- tid = FIELD_GET(HTT_TX_PPDU_DUR_INFO0_TID_MASK, info0);
+ tid = FIELD_GET(HTT_TX_PPDU_DUR_INFO0_TID_MASK, info0) &
+ IEEE80211_QOS_CTL_TID_MASK;
tx_duration = __le32_to_cpu(ppdu_dur->tx_duration);
ieee80211_sta_register_airtime(peer->sta, tid, tx_duration, 0);
--
1.9.1
ieee80211_sta_register_airtime uses ieee80211_ac_from_tid(tid) which
already applies a mask, so should be returning IEEE80211_AC_BE = 2
already, there has been recent changes to some of this so maybe I'm
looking at the wrong decision but I don't think this change makes a
difference
On Wed, Feb 19, 2020 at 8:02 PM Yibo Zhao <[email protected]> wrote:
>
> The tid of 11a station with WMM disable reported by FW is 0x10 in
> tx completion. The tid 16 is mapped to a NULL txq since buffer
> MMPDU capbility is not supported. Then 11a station's airtime will
> not be registered due to NULL txq check. As a results, airtime of
> 11a station keeps unchanged in debugfs system.
>
> Mask the tid along with IEEE80211_QOS_CTL_TID_MASK to make it in
> the valid range.
>
> Hardwares tested : QCA9984
> Firmwares tested : 10.4-3.10-00047
>
> Signed-off-by: Yibo Zhao <[email protected]>
> ---
> drivers/net/wireless/ath/ath10k/htt_rx.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
> index 38a5814..f883f2a 100644
> --- a/drivers/net/wireless/ath/ath10k/htt_rx.c
> +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
> @@ -2744,7 +2744,8 @@ static void ath10k_htt_rx_tx_compl_ind(struct ath10k *ar,
> continue;
> }
>
> - tid = FIELD_GET(HTT_TX_PPDU_DUR_INFO0_TID_MASK, info0);
> + tid = FIELD_GET(HTT_TX_PPDU_DUR_INFO0_TID_MASK, info0) &
> + IEEE80211_QOS_CTL_TID_MASK;
> tx_duration = __le32_to_cpu(ppdu_dur->tx_duration);
>
> ieee80211_sta_register_airtime(peer->sta, tid, tx_duration, 0);
> --
> 1.9.1
Yibo Zhao <[email protected]> wrote:
> The tid of 11a station with WMM disable reported by FW is 0x10 in
> tx completion. The tid 16 is mapped to a NULL txq since buffer
> MMPDU capbility is not supported. Then 11a station's airtime will
> not be registered due to NULL txq check. As a results, airtime of
> 11a station keeps unchanged in debugfs system.
>
> Mask the tid along with IEEE80211_QOS_CTL_TID_MASK to make it in
> the valid range.
>
> Hardwares tested : QCA9984
> Firmwares tested : 10.4-3.10-00047
>
> Signed-off-by: Yibo Zhao <[email protected]>
> Signed-off-by: Kalle Valo <[email protected]>
Patch applied to ath-next branch of ath.git, thanks.
f9680c75d187 ath10k: fix not registering airtime of 11a station with WMM disable
--
https://patchwork.kernel.org/patch/11393223/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches