⨯ Check out our Monthly Survey Page to see what our users are running.
GPU crash with FSR
Sloprano Jan 11
My System Info

Linux Distribution: Manjaro
Desktop Environment: XCFE4
Graphics Card: AMD 6950XT
GPU Driver Version: video-linux 2018.05.04

Have you checked for system updates?: Yes

If this is not Steam, you can still give us a system readout using inxi with "inxi -SbCGxx" in terminal (please put that into Pastebin/Gist as linked above!):
https://pastebin.com/pTqGAYa3

You can note your issue below here

Wondered if anyone has experienced issues like GPU crash when using upscaling on an AMD GPU.

I have been gaming on linux for some months now and everything has been going smooth, but has always encountered issues using FSR in more demanding games. Baldur's Gate 3 and Darktide both, when FSR 1 og 2 was enable no matter the setting, the GPU would crash, monitors would go black and I would be presented with the login screen again. Attempt to launch the games again would most of the time fail with no errors until I restarted the computer. (And I would always have to restart twice, first restart just hangs on the three Manjaro dots. On the next boot, Manjaro starts fast as normal).

The problem was easily avoided by not using any upscaling in those games. But when I try to play Alan Wake 2 I find I cannot turn off FSR 2, and the moment the first level finishes loading the crash happens like described above.

Experimented with installing Windows on a seperate drive and running with the same setup, there it worked with no issues. Been reinstalling the drivers on manjaro and been trying to use mesa-git I've been reading about.



Note any particular steps to reproduce it below here
Turning on FSR 1 or 2 in any newer games and try to play.
BlackBloodRum Jan 11
After your GPU crashes, you probably (almost certainly) received an error message regarding the crash in your system logs.

Trigger the crash again, but this time grab the related lines from dmesg (using "dmesg | grep amdgpu" should suffice in most cases) from the crash. Without the error, we're all basically just taking wild guesses.
Sloprano Jan 12
Quoting: BlackBloodRumAfter your GPU crashes, you probably (almost certainly) received an error message regarding the crash in your system logs.

Trigger the crash again, but this time grab the related lines from dmesg (using "dmesg | grep amdgpu" should suffice in most cases) from the crash. Without the error, we're all basically just taking wild guesses.

Thanks for the tip, BlackBlood, Ill check that soon as I get home today.
Sloprano Jan 12
Here's the output of amdgpu.
https://pastebin.com/2WHZcAdX

The error I could discern was:
127.933782] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=52029, emitted seq=52031

Currently searching around about the message.
Sloprano Jan 13
I couldnt find anything about the error that made a difference for when running the game. Decided to do a backup and install Pop OS. Installed everything and made no further changes, Alan Wake 2 started with no issues and have been running smoothly for the last hour.

Biggest difference is that Alan Wake 2 didn't warn about "Your GPU is misisng Mesh Shader" after the reinstall, so wonder if it was something I had done wrong when using Manjaro.
LoudTechie Jan 13
Quoting: SlopranoHere's the output of amdgpu.
https://pastebin.com/2WHZcAdX

The error I could discern was:
127.933782] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=52029, emitted seq=52031

Currently searching around about the message.

I know it's too late, but your problem was too interesting, so here is my fix.
Quick skilled operator search reveals:
driver issue
should be fixed by upgrading to libdrm-amdgpu1 2.4.114-1 and kernel version 6.2+
This is manjaro type in:
sudo mhwd -a pci free 0300 or
sudo mhwd -a pci non-free 0300

Last edited by LoudTechie on 13 January 2024 at 9:50 am UTC
Sloprano Jan 14
Quoting: LoudTechie
Quoting: SlopranoHere's the output of amdgpu.
https://pastebin.com/2WHZcAdX

The error I could discern was:
127.933782] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=52029, emitted seq=52031

Currently searching around about the message.

I know it's too late, but your problem was too interesting, so here is my fix.
Quick skilled operator search reveals:
driver issue
should be fixed by upgrading to libdrm-amdgpu1 2.4.114-1 and kernel version 6.2+
This is manjaro type in:
sudo mhwd -a pci free 0300 or
sudo mhwd -a pci non-free 0300

Still thanks a lot for taking the time to look up, LoudTechie. Things are working on Pop and I will revisit your post if I try Manjaro again and run into the same issue.

Cheers!
While you're here, please consider supporting GamingOnLinux on:

Reward Tiers: Patreon. Plain Donations: PayPal.

This ensures all of our main content remains totally free for everyone! Patreon supporters can also remove all adverts and sponsors! Supporting us helps bring good, fresh content. Without your continued support, we simply could not continue!

You can find even more ways to support us on this dedicated page any time. If you already are, thank you!
Login / Register

Or login with...
Sign in with Steam Sign in with Google
Social logins require cookies to stay logged in.
Register Forgot Login?
★ Support Us
Popular this week
Search or view by category
Contact
Latest Comments
Latest Forum Posts
Misc