Dear Maintainer,

On a system with Intel Raptor Lake and NVIDIA RTX 4050 (hybrid graphics),
the kernel module `sof-audio-pci-intel-tgl` fails repeatedly after resume
from suspend.
This results in temporary freezes and instability in the GNOME desktop
environment.

The issue is reproducible and occurs consistently after resuming from
suspend. Kernel logs show the following errors:

    sof-audio-pci-intel-tgl 0000:00:1f.3: ipc tx error for 0x60010000
(msg/reply size: 108/20): -22
    sof-audio-pci-intel-tgl 0000:00:1f.3: HW params ipc failed for stream 1
    sof-audio-pci-intel-tgl 0000:00:1f.3: error: set pcm hw_params after
resume

These errors coincide with a brief freeze of the GNOME interface, which
recovers after a few seconds. The system remains usable, but the experience
is degraded.

Workaround: adding `snd_intel_dspcfg.dsp_driver=1` to the kernel command
line probably could resolve the issue by forcing the legacy HDA driver.
I would like however to avoid this action, I therefore reported this bug.

Expected outcome: GNOME should resume normally without freezing, and the
audio subsystem should recover cleanly.
let me know if further logs or testing are needed.

Additionally, I am currently running the older Debian 12 kernel
(6.1.0-37-amd64) after upgrading to Debian 13, because the newer 6.12
kernel introduced severe instability on my system.
I had to revert to a more stable environment while attempting to resolve
the issues related to sof-audio and GNOME freezes and Nvidia Driver
problems.

Hello alessandro,

Can you please provide the following things for us to help you?:

 - Output of lspci -vvnn (as root)
 - full kernel log (`sudo journalctl -k` or `sudo dmesg`)

Does it help to unload the module (i.e.
`sudo rmmod snd-sof-pci-intel-tgl`) when the problem occurs? If yes, do
the problems reappear if you load the module again
(`sudo modprobe snd-sof-pci-intel-tgl`)? Does the problem also occur if
you rmmod the module before suspend and reload it after resume?

Did you use earlier kernels where this problem didn't occur? If yes,
which was the last good version?

Your kernel reports to be tainted due to a proprietary kernel module, so
I assume that you're using the nvidia driver? Does the problem also
occur without that one, e.g. if you switch to nouveau?

Best regards
Uwe

Hello Uwe,

After a few days of troubleshooting, I’ve managed to stabilize my
system with the Trixie kernel set. I’ve now completed the transition to
Debian 13, running kernel 6.12.41+deb13-amd64, which is currently
stable enough for daily use.

I mention this because I accidentally ran `apt autoremove`, which
unfortunately removed the old kernel 6.1.x sources, headers, and
related packages. As a result, I’m unable to provide further logs or
documentation from that environment. Apologies for the inconvenience.

Interestingly, I discovered that some of the freezes after suspend in
kernel 6.12 were caused by an SMB mount defined in `/etc/fstab`, which
was timing out after 180 seconds. I’m not entirely sure if this was
also contributing to the freezes in kernel 6.1, but it’s a possibility
worth considering.

Kind regards,
Alex