Re: [PATCH rtw-next v2 1/2] wifi: rtw88: move the bmc high queue decision to tx.c

["Luka Gejak" <luka.gejak@linux.dev>]

Hi Ping-Ke,

October 22, 2026 at 7:11 AM, "Luka Gejak" <luka.gejak@linux.dev> wrote:


> 
> October 22, 2026 at 4:20 AM, "Ping-Ke Shih" <pkshih@realtek.com> wrote:
> 
> > 
> > Luka Gejak <luka.gejak@linux.dev> wrote:
> >  
> >  
> >  USB decides in usb.c which broadcast and multicast (bmc) frames may use
> >  the high queue. SDIO needs the same decision, so move the helpers to
> >  tx.c behind a single exported rtw_tx_bmc_use_hiq() and keep the budget
> >  state in struct rtw_dev.
> >  
> >  No functional change.
> >  
> >  By my review, I think it doesn't change logic at all, but I hope if
> >  someone can help to test USB devices to confirm it still works well.
> > 
> I ran the same test with rtl8723du and there is no regression.
> 
> Best regards,
> Luka Gejak
>

I was thinking, as staging got removed, for 10-ish weeks(until 7.5)
users of RTL8723BS won't have AP mode. So I was wondering if these two
patches could pass as fixes into rc2 release(or whatever is the first
rc into which rtw tree is merged), as this would be DHCP stall fix. To
elaborate: when this driver was merged it, was exposingAP mode to
userspace, due to RTW88 doing it globally. Therefore AP mode existed,
but wasn't working properly. So maybe it's a bit of a stretch, but I
think it's a valid proposal to send it as a fix into next rc release.

Best regards,
Luka Gejak