Display Freeze due to amdgpu_job_timedout on Linux 6.9.7
Brief summary of the problem:
During normal office activity in Gnome (web browsing in Chrome, meetings in Teams), the display occasionally freezes for a couple of seconds. The kernel log reports:
[ 5630.310783] amdgpu 0000:c1:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:2 pasid:32779)
[ 5630.310790] amdgpu 0000:c1:00.0: amdgpu: in process chrome pid 4330 thread chrome:cs0 pid 4376)
[ 5630.310792] amdgpu 0000:c1:00.0: amdgpu: in page starting at address 0x0000984d3d082000 from client 10
[ 5630.310793] amdgpu 0000:c1:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00201430
[ 5630.310794] amdgpu 0000:c1:00.0: amdgpu: Faulty UTCL2 client ID: SQC (data) (0xa)
[ 5630.310795] amdgpu 0000:c1:00.0: amdgpu: MORE_FAULTS: 0x0
[ 5630.310796] amdgpu 0000:c1:00.0: amdgpu: WALKER_ERROR: 0x0
[ 5630.310796] amdgpu 0000:c1:00.0: amdgpu: PERMISSION_FAULTS: 0x3
[ 5630.310797] amdgpu 0000:c1:00.0: amdgpu: MAPPING_ERROR: 0x0
[ 5630.310797] amdgpu 0000:c1:00.0: amdgpu: RW: 0x0
[ 5640.451988] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, but soft recoveredAt the moment this happens maybe once a day. There doesn't seem to be anything specific to reproduce it.
This seems to happen since I updated to Linux 6.9.7. Earlier 6.9 versions where fine. Might have also been a Chrome update.
Hardware description:
- CPU: AMD Ryzen 7 7840U w/ Radeon 780M Graphics
- GPU: c1:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Phoenix1 [1002:15bf] (rev c4)
- System Memory: 32 GiB
- Display(s): Builtin laptop display + LG 27UL850-W 4K display via USB-C
- Type of Display Connection: USB-C
System information:
- Distro name and Version: NixOS 24.05
- Kernel version: 6.9.7 NixOS SMP PREEMPT_DYNAMIC Thu Jun 27 11:52:32 UTC 2024 x86_64 GNU/Linux
- Custom kernel: N/A
- AMD official driver version: N/A
How to reproduce the issue:
Not clear yet.
Attached files:
Log files (for system lockups / game freezes / crashes)
Edited by Julian Stecklina