Comment # 21 on bug 1196967 from Takashi Iwai

(In reply to Alexandre Vicenzi from comment #19)
> (In reply to Takashi Iwai from comment #17)
> > If you can still reproduce the problem with 5.19, at best, report this to
> > the upstream.  It should be fixed there at first.
> > 
> 
> I was not able to reproduce this issue with 5.19.

Aha, so it's been fixed between 5.18 and 5.19, supposedly?

> > Just to be sure: you've tested vanilla kernel, and 5.15.x is only to 5.15.13?
> > I'm asking it because 5.15.23 contains a known problematic stuff (the commit
> > f0a60c7c4edee3f108ec435d457149175d0b0afa).
> 
> Yes, the vanilla versions I tested were only 5.14.15, 5.15.13, and 5.16.16.

OK, any chance to give a quick try with the revert of
f0a60c7c4edee3f108ec435d457149175d0b0afa?