On Sat, 2025-04-26 at 22:26 +0200, Janusz Dziedzic wrote: > > In some way though this also means your kernel patch _did_ introduce a > > userspace regression, so I'm not entirely sure why we should do it at > > all? > > > I think > - only HE report wrong MCS/PPE len - EHT report it correctly Yeah that's unfortunate. > - only bug in iw, hostapd recalc it based on HE capab > - looks odd to skip kernel patch only due to some user mode app bug > That's ... what happens? That's how "don't break userspace" works :) So I think we should just leave it. We can take it as a historical quirk and try to do better in the future (as we already did for EHT), but I don't see value in breaking iw for such a thing. johannes
