CS2 Crashing amdgpu GPU or driver

[bjornpijnacker]

Your system information

• System information from steam (Steam -> Help -> System Information) in a gist System Information: https://gist.github.com/bjornpijnacker/6343826eda9c4af623fba39d1ea1d952 Steam Runtime Diagnostics

• Have you checked for system updates?: Yes

Please describe your issue in as much detail as possible:

Describe what you expected should happen and what did happen. Please link any large pastes as a Github Gist.

Steps for reproducing this issue:

1. Launch CS2 on Fedora Linux 41
2. Play for an hour or so in a competitive match
3. System freezes and I am booted to GDM login. Game remains running and a full system restart is required.

Log snippet, indicating an issue originating from CS2 Vulkan:

[ 4614.327849] amdgpu 0000:03:00.0: amdgpu: Dumping IP State
[ 4614.330348] amdgpu 0000:03:00.0: amdgpu: Dumping IP State Completed
[ 4614.340384] amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 timeout, signaled seq=17735888, emitted seq=17735890
[ 4614.340386] amdgpu 0000:03:00.0: amdgpu: Process information: process cs2 pid 29348 thread VKRenderThread pid 29384
[ 4616.340522] amdgpu 0000:03:00.0: amdgpu: MES failed to respond to msg=RESET
[ 4616.340526] [drm:amdgpu_mes_reset_legacy_queue [amdgpu]] *ERROR* failed to reset legacy queue
[ 4616.340693] amdgpu 0000:03:00.0: amdgpu: GPU reset begin!
[ 4618.580628] amdgpu 0000:03:00.0: amdgpu: MES failed to respond to msg=REMOVE_QUEUE
[ 4618.580631] [drm:amdgpu_mes_unmap_legacy_queue [amdgpu]] *ERROR* failed to unmap legacy queue
[ 4618.814047] [drm:gfx_v11_0_hw_fini [amdgpu]] *ERROR* failed to halt cp gfx
[ 4618.853764] amdgpu 0000:03:00.0: amdgpu: MODE1 reset
[ 4618.853766] amdgpu 0000:03:00.0: amdgpu: GPU mode1 reset
[ 4618.853815] amdgpu 0000:03:00.0: amdgpu: GPU smu mode1 reset
[ 4619.359816] amdgpu 0000:03:00.0: amdgpu: GPU reset succeeded, trying to resume
[ 4619.359887] [drm] PCIE GART of 512M enabled (table at 0x0000008001300000).
[ 4619.359915] [drm] VRAM is lost due to GPU reset!
[ 4619.359925] amdgpu 0000:03:00.0: amdgpu: PSP is resuming...
[ 4619.429546] amdgpu 0000:03:00.0: amdgpu: reserve 0x1300000 from 0x85fc000000 for PSP TMR
[ 4619.573925] amdgpu 0000:03:00.0: amdgpu: RAP: optional rap ta ucode is not available
[ 4619.573927] amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[ 4619.573929] amdgpu 0000:03:00.0: amdgpu: SMU is resuming...
[ 4619.573931] amdgpu 0000:03:00.0: amdgpu: smu driver if version = 0x0000003d, smu fw if version = 0x00000040, smu fw program = 0, smu fw version = 0x004e8000 (78.128.0)
[ 4619.573934] amdgpu 0000:03:00.0: amdgpu: SMU driver if version not matched
[ 4619.753693] amdgpu 0000:03:00.0: amdgpu: SMU is resumed successfully!
[ 4619.764286] [drm] DMUB hardware initialized: version=0x07002A00
[ 4619.893407] amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[ 4619.893409] amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[ 4619.893410] amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[ 4619.893411] amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[ 4619.893411] amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[ 4619.893412] amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[ 4619.893412] amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[ 4619.893413] amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[ 4619.893413] amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[ 4619.893414] amdgpu 0000:03:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[ 4619.893414] amdgpu 0000:03:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
[ 4619.893415] amdgpu 0000:03:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
[ 4619.893415] amdgpu 0000:03:00.0: amdgpu: ring vcn_unified_1 uses VM inv eng 1 on hub 8
[ 4619.893416] amdgpu 0000:03:00.0: amdgpu: ring jpeg_dec uses VM inv eng 4 on hub 8
[ 4619.893417] amdgpu 0000:03:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 14 on hub 0
[ 4619.896692] amdgpu 0000:03:00.0: amdgpu: GPU reset(2) succeeded!
[ 4619.896978] [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!

[ 4614.340386] amdgpu 0000:03:00.0: amdgpu: Process information: process cs2 pid 29348 thread VKRenderThread pid 29384 is the part that pointed me to CS2. I haven't had it happen anywhere else.

I've looked at https://gitlab.freedesktop.org/drm/amd/-/issues/3131, where people report similar issues. However, CS2 does not run my GPU above regular boost frequency at all. It's at about 80% utilization. I have also looked at #3386, which is a different issue. I am missing the [130882.405656] [drm:gfx_v10_0_priv_reg_irq [amdgpu]] ERROR Illegal register access in command stream logline that @tadzik reports.