Random "soft freezes" on Framework 13 AMD with Debian Trixie

Framework Laptop 13 Linux
1

Hey, everyone.

Been getting some non-frequent, but odd issues after updating my stable Debian to Trixie. Seems to happen when using Firefox.

One change i made was I wanted to try enabling Game Mode for the AMD GPU in the UEFI settings to see if that was fixed. Yes, I also updated the UEFI to 3.0.9. Who knows if that is causing the problem. I may disable it and test more.

Here’s the error messages that get produced.

Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:3 pasid:32774)
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:  in process firefox-esr pid 6280 thread firefox-es:cs0 pid 6369)
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:   in page starting at address 0x0000000000000000 from client 10
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00301430
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:          Faulty UTCL2 client ID: SQC (data) (0xa)
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:          MORE_FAULTS: 0x0
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:          WALKER_ERROR: 0x0
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:          PERMISSION_FAULTS: 0x3
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:          MAPPING_ERROR: 0x0
Aug 15 22:34:38 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu:          RW: 0x0
Aug 15 22:34:48 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu: Dumping IP State
Aug 15 22:34:48 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu: Dumping IP State Completed
Aug 15 22:34:48 lilfoot kernel: amdgpu 0000:c1:00.0: amdgpu: ring gfx_0.0.0 timeout, but soft recovered

Let me know if anyone else has similar ones.

Oh, by the way, is a new thread an appropriate place to post Debian Trixie problems, or is there already a thread for it? Haven’t checked these forums in a while.

2
  1. What is your linux kernel version?
  2. What is your Mesa version?

Taking a quick glance around for the GCVM_L2_PROTECTION_FAULT_STATUS, I see a lot of posts on NixOS, Arch, and other forums. I also found this resolved gitlab issue.

3

Ah, sure, I can post that information.

$ uname -a
Linux lilfoot 6.12.41+deb13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.12.41-1 (2025-08-12) x86_64 GNU/Linux

And cmdline info if that’s relevant, since I did change a few things to hide some warnings:

$ cat /proc/cmdline
BOOT_IMAGE=/vmlinuz-6.12.41+deb13-amd64 root=UUID=<UUID> ro rootflags=subvol=@lilfoot-rootfs quiet i8042.nopnp i8042.unlock=1

$ glxinfo | grep "OpenGL version"
OpenGL version string: 4.6 (Compatibility Profile) Mesa 25.0.7-2

I’ll take a look at the Gitlab page.