RE: Error in rtw89 using Kernel 7.1.4

[Ping-Ke Shih <pkshih@realtek.com>]

Adrodoc <adrodoc55@googlemail.com> wrote:
> Ok, I'll try 6.6.9-arch1-1 for longer to see if I encounter the SER error.
> 
> I don't quite understand what you mean with:
> > Therefore, if the infrequent SER doesn't bother your daily work, they suggest to just keep it as was.
> 
> Do you mean I should keep using the new kernels and ignore the error?

Yes, because SER L1 ("SER catches error: 0x1000") means firmware detect
and fix an error already, and notify driver it does (driver no need to
do anything). Therefore, I'd suggest to ignore this message if it doesn't
bother your daily use. 

> The disconnects aren't the end of the world, but they are annoying and
> I was hoping I could help you improve the driver to fix the problem.
> So I would like to continue, not just for myself, but to help improve
> this for others too. I have an off the shelf prebuilt PC
> (https://psref.lenovo.com/Detail/Legion/Lenovo_Legion_T5_26AMR5?M=90RC00UUGE),
> I only added a second SSD, so I am sure I am not the only one with
> this issue.

But if the disconnection is annoying to you, we should continue to dig
the problem.

> 
> > we can bisect the commits in detail.
> > I need your help to build and test driver ko.
> > The second way is to debug on the latest kernel and apply debug patch to dump more detail needed by
> internal experts.
> 
> I have never built drivers before, and I don't have a lot of
> experience with C, but I am a developer, so we can try that. I am
> ready to try both options.
> 
> Something else I was wondering: In the error logs I see this message:
> > firmware failed to ack for leaving ps mode

This is another problem. I'm not sure if it is related to SER L1 mentioned
above. 

> 
> I am guessing ps mode is power saving mode, which is weird, because
> the error happens during zoom calls and remote desktop sessions, which
> use the connection constantly. So I would not expect the Wifi-Card to
> go into power save mode here.
> I read on https://wiki.archlinux.org/title/Network_configuration/Wireless
> that in quite a few cases they suggest disabling power save.
> I checked and power save seems to be enabled on my system:
> $ iw dev wlan0 get power_save
> Power save: on
> 
> Should I try to disable it? Would that give you any new information?

Yes, it is worth to try by setting
iw dev wlan0 set power_save off

> 
> The arch wiki also contains a chapter specific to rtw89:
> https://wiki.archlinux.org/title/Network_configuration/Wireless#RTW89
> They suggest disabling ASPM-related features using the config below.
> /etc/modprobe.d/70-rtw89.conf
>   options rtw89_pci disable_aspm_l1=y disable_aspm_l1ss=y
>   options rtw89_core disable_ps_mode=y

These are also worth to try, because there might be interoperability issue
on PCI ASPM. After setting this, do cold reboot to take effect. 

> 
> See also:
> https://github.com/lwfinger/rtw89#option-configuration
> https://github.com/lwfinger/rtw89/issues/275#issuecomment-1784155449
> 
> However this sounds like it affects notebooks, whereas I have a desktop PC.
> 
> Should I try this? Would that give you any new information?

Trying power saving parameters will help to the message
'firmware failed to ack for leaving ps mode' at least.

Ping-Ke