[Bug 201957] amdgpu: ring gfx timeout

[bugzilla-daemon at kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=201957

--- Comment #72 from Martin von Wittich (martin.von.wittich at iserv.eu) ---
I can confirm that adding "amdgpu.dpm=0" to the kernel command line seems to
resolve this issue - I enabled that option on 2022-06-12 13:24, and my system
didn't crash at all on 2022-06-12 - 2022-06-14 (I was on vacation from
2022-06-15 on and didn't use my computer from then on).

I don't use Linux for gaming and therefore can't comment how badly this affects
gaming performance, but I did notice mpv could no longer play 1080p x264 video
without stuttering when it defaults to --vo=gpu. Using another --vo like sdl
seems to be a viable workaround.

> Did you try with the latest Linux Kernel? I had a lot of gpu lockups like this. Also try these kernel parameters : "amdgpu.ppfeaturemask=0xffffbffb  amdgpu.noretry=0 amdgpu.lockup_timeout=0 amdgpu.gpu_recovery=1 amdgpu.audio=0 amdgpu.deep_color=1 amd_iommu=on iommu=pt"" ( you might also try with amdgpu.ppfeaturemask=0xfffd7fff or amdgpu.ppfeaturemask=0xffffffff )

I'll try these next.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.