Hi Ping-Ke,
On Thu, Apr 10, 2025 at 6:30 AM Ping-Ke Shih <pkshih@xxxxxxxxxxx> wrote:
[...]
> > @@ -718,10 +718,7 @@ static u8 rtw_sdio_get_tx_qsel(struct rtw_dev *rtwdev, struct sk_buff *skb,
> > case RTW_TX_QUEUE_H2C:
> > return TX_DESC_QSEL_H2C;
> > case RTW_TX_QUEUE_MGMT:
> > - if (rtw_chip_wcpu_11n(rtwdev))
> > - return TX_DESC_QSEL_HIGH;
> > - else
> > - return TX_DESC_QSEL_MGMT;
> > + return TX_DESC_QSEL_MGMT;
>
> Do you remember why you did the special deal with 11n chips?
> And this RFC looks good to me. (except to commit message, but this is RFC)
I don't remember - and Jernej said the same thing.
However, since we got the first 802.11n hardware for testing long
after this part was written my suggestion is: let's roll this into a
proper patch, Cc Fiona Klute <fiona.klute@xxxxxx> (author of RTL8723CS
support) on the resulting patch(es) and then apply the patches
(assuming nobody observes any problems).
To make this a non-RFC patch the following steps are needed (in my opinion):
- split the change into two patches (one which unconditionally calls
rtw_sdio_indicate_tx_status())
- another one for the TX_DESC_QSEL_MGMT mapping
- each of the patches should include their own description
- I checked the history and it seems that both problems were
introduced with the original commit, meaning both patches should get
the following line (above the Signed-off-by): Fixes: 65371a3f14e7
("wifi: rtw88: sdio: Add HCI implementation for SDIO based chipsets")
- (plus anything Ping-Ke has to add :-) )
Best regards,
Martin
