Amdgpu crash while playing elden ring

[kyral]

that is the only game where i’ve been able to reproduce this issue.
after roughly 30mns of gameplay, the driver seems to crash and the gpu then resets, the whole system itself doesn’t crash, only the driver seems to do so.
this happens since a few days and despite rolling back to the oldest snapshot that i have (7 june 2024), the issue still persists.
it’ve tried migrating on slowroll as well but the issue is still there so i went back to TW.

i use KDE + Wayland

here’s the dmesg output after the gpu reset

[ 4111.108508] [T29488] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=4308193, emitted seq=4308195
[ 4111.108787] [T29488] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process eldenring.exe pid 32170 thread vkd3d_queue pid 32616
[ 4111.109034] [T29488] amdgpu 0000:0c:00.0: amdgpu: GPU reset begin!
[ 4111.602315] [T29488] amdgpu 0000:0c:00.0: amdgpu: MODE1 reset
[ 4111.602322] [T29488] amdgpu 0000:0c:00.0: amdgpu: GPU mode1 reset
[ 4111.602401] [T29488] amdgpu 0000:0c:00.0: amdgpu: GPU smu mode1 reset
[ 4112.111923] [T29488] amdgpu 0000:0c:00.0: amdgpu: GPU reset succeeded, trying to resume
[ 4112.112386] [T29488] [drm] PCIE GART of 512M enabled (table at 0x0000008000300000).
[ 4112.112460] [T29488] [drm] VRAM is lost due to GPU reset!
[ 4112.112463] [T29488] amdgpu 0000:0c:00.0: amdgpu: PSP is resuming...
[ 4112.188047] [T29488] amdgpu 0000:0c:00.0: amdgpu: reserve 0xa00000 from 0x82fd000000 for PSP TMR
[ 4112.287845] [T29488] amdgpu 0000:0c:00.0: amdgpu: RAS: optional ras ta ucode is not available
[ 4112.301408] [T29488] amdgpu 0000:0c:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[ 4112.301414] [T29488] amdgpu 0000:0c:00.0: amdgpu: SMU is resuming...
[ 4112.301419] [T29488] amdgpu 0000:0c:00.0: amdgpu: smu driver if version = 0x0000000e, smu fw if version = 0x00000012, smu fw program = 0, version = 0x00413e00 (65.62.0)
[ 4112.301423] [T29488] amdgpu 0000:0c:00.0: amdgpu: SMU driver if version not matched
[ 4112.301480] [T29488] amdgpu 0000:0c:00.0: amdgpu: use vbios provided pptable
[ 4112.359343] [T29488] amdgpu 0000:0c:00.0: amdgpu: SMU is resumed successfully!
[ 4112.360624] [T29488] [drm] DMUB hardware initialized: version=0x02020020
[ 4112.627351] [T29488] [drm] kiq ring mec 2 pipe 1 q 0
[ 4112.631854] [T29488] [drm] VCN decode and encode initialized successfully(under DPG Mode).
[ 4112.632410] [T29488] [drm] JPEG decode initialized successfully.
[ 4112.632455] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[ 4112.632458] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
[ 4112.632459] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
[ 4112.632461] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 5 on hub 0
[ 4112.632463] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 6 on hub 0
[ 4112.632464] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 7 on hub 0
[ 4112.632466] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 8 on hub 0
[ 4112.632467] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 9 on hub 0
[ 4112.632469] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 10 on hub 0
[ 4112.632471] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring kiq_0.2.1.0 uses VM inv eng 11 on hub 0
[ 4112.632472] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
[ 4112.632474] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
[ 4112.632476] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 8
[ 4112.632477] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 8
[ 4112.632479] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 8
[ 4112.632481] [T29488] amdgpu 0000:0c:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 8
[ 4112.636463] [T29488] amdgpu 0000:0c:00.0: amdgpu: recover vram bo from shadow start
[ 4112.659855] [T29488] amdgpu 0000:0c:00.0: amdgpu: recover vram bo from shadow done
[ 4112.659883] [T29488] amdgpu 0000:0c:00.0: amdgpu: GPU reset(2) succeeded!
[ 4112.659932] [T32616] [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!

[boopie]

Me too. I’m trying to play elden ring DLC these days and this crash happens about every 15-20 mins. I have a AMD 5700xt and so far I’ve made sure I’ve installed amdgpu, vulkan-radeon, and mesa and everything works fine for most other games. I also have tried undervolting my GPU and setting the clock speeds using CoreCtl.

I read somewhere that the RDNA on 5700xt is bad at dealing with the elden ring 60fps cap idk. What GPU are you using, it looks like it happens to 5000-7000 xt cards mostly, while using any graphics software.

The 2 things I haven’t tried yet are mesa-git because building it is a lil tough and then also trying adding some amdgpu kernel flags to grub, neither of which is a confirmed solution. I’ll lyk if i find anything stable.

[kyral]

i’ve been able to solve the issue by lowering the undervolt, for some reason no game or benchmark made my gpu crach like elden ring did.
so i’ve reduced the UV and evertyhing is fine now.
tho i won’t bother much longer with elden ring, it’s an awful game anyway

[boopie]

I fixed it by opening steam through Lutris and turning on feral gamemode in the runner settings idk if that does anything tho since it probably launches steam in gamemode not eldin ring. Maybe gamemoderun %command% in steam run options works.