You are not logged in.
No matter the game, periodically I get these freezes in display on Wayland Hyprland. All I could find linking to it was this issue submitted to the amd code repo. Other than that I am at a loss. Wondering if any of you had any issues? I have tested with all kernels since 5.x.x all the way to latest stable 6.7.6 without success (without kernel flags that might change amdgpus behavior).
Variant 1:
Mar 01 14:09:33 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Variant 2:
Mar 01 14:13:10 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* flip_done timed out
Full example:
Feb 06 15:40:46 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 06 15:49:53 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* flip_done timed out
Feb 06 15:50:04 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* flip_done timed out
Feb 06 15:50:14 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* flip_done timed out
Feb 06 15:50:37 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 07 03:51:55 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 07 05:50:38 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 07 18:57:12 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 07 20:27:42 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 07 22:22:46 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 08 10:51:38 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 08 17:09:44 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 12 02:40:03 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 12 03:28:54 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
Feb 12 04:45:00 atelier kernel: amdgpu 0000:09:00.0: [drm] *ERROR* [CRTC:91:crtc-0] flip_done timed out
It can either go away on it's own and the display will fix itself or it can stay like that for hours even days. As for the backtrace, it's identical to the one sent in the issue I brought up before.
Last edited by demeter555 (2024-03-02 00:04:20)
Offline
No matter the game
Native wayland or xwayland?
All steam?
Is https://archlinux.org/packages/extra/x86_64/warsow/ affected?
eglgears_wayland? (from mesa-utils)
Same problem on weston or X11?
Online
Both native Waylandm and XWayland. Nothing is affected immediately, all applications can trigger it. Some Steam games on Proton might trigger it faster but eventually the display always ends up freezing whether it's on Chromium or the terminal alone. eglgears, glxgears and vkcube all unaffected; no artifacts to be found or relevant kernel logs.
Offline
Did you already exclude hardware failure?
Offline
Before that: did you rule out it's hyprland (eg. try openbox or weston)?
Or does
the terminal alone
mean the linux console w/o any display server?
Is any of this you also?
https://gitlab.freedesktop.org/drm/amd/-/issues/2950
https://gitlab.freedesktop.org/drm/amd/-/issues/2075
https://www.reddit.com/r/archlinux/comm … ly_update/
Online
The first link is the one I attached as well. By "the terminal alone" I was just exemplifying it happens regardless of the binary being run. I seriously doubt it's a hardware failure since this GPU barely has 6 months being used. The TTY itself does not produce the same error, though "https://www.reddit.com/r/archlinux/comm … ly_update/" does sound familiar.
Offline
did you rule out it's hyprland (eg. try openbox or weston)?
What I meant is whether you're the reporter of any of those bugs.
The first amdgpu one is runs on X11 w/ VRR enabled => please test the beavior on an X11 session and post the xorg log you get.
Online