[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#1027483: linux-image-5.10.0-20-amd64: sound with firmware-sof-intel on Intel Lake stops working with newer kernel



Control: tags -1 + moreinfo

On Sun, Jan 01, 2023 at 04:34:28PM +0100, Salvatore Bonaccorso wrote:
> Control: reassign -1 src:linux 5.10.158-2
> 
> Hi Malvin,
> 
> On Sun, Jan 01, 2023 at 02:36:28PM +0100, Malvin Gattinger wrote:
> > Package: linux-image-5.10.0-20-amd64
> > Version: 5.10.158-2
> > Severity: normal
> > X-Debbugs-Cc: malvin@w4eg.eu
> > 
> > Dear Maintainer,
> > 
> > I am unsure if this problem belongs to linux-image-amd64 as a
> > regression or is actually a but in firmware-sof-intel.
> > 
> > On an HP ProBook with this sound card (from lspci):
> > 
> > 00:1f.3 Multimedia audio controller: Intel Corporation Tiger Lake-LP
> > Smart Sound Technology Audio Controller (rev 20)
> > 
> > Updating from linux-image-5.10.0-19-amd64 to
> > linux-image-5.10.0-20-amd64 breaks sound.
> > 
> > Sound output from VLC works fine immediately after boot, but using
> > the microphone or other audio use cases later disables all output.
> > Unfortunately the only reproducible example on this laptop was to
> > open the proprietary zoom video-conferencing software which
> > immediately stops all output, but other symptoms are discussed here:
> > 
> > https://forums.debian.net/viewtopic.php?t=153585
> > 
> > https://forums.debian.net/viewtopic.php?p=764594
> > 
> > PulseAudio claims that the output is still running, i.e. pavucontrol
> > and KDE audio settings still show a moving bar for the output.
> > 
> > The only workaround for now is to use the older kernel from
> > linux-image-5.10.0-19-amd64.
> 
> To pin point the possibly breaking change in the kernel, would you be
> able to build kernels, first both upstream kernels mathcing the
> 5.10.149 (for the 19 ABI one), and 5.10.158 and then bisect betweeen
> the two?
> 
> https://wiki.debian.org/DebianKernel/GitBisect

Following that, I wonder if #1027483, #1026834 (and maybe #1010733)
will have all the same underlying cause.

Regards,
Salvatore


Reply to: