AMDGPU crashes with 6.18.13-100.fc42.x86_64

Framework Laptop 13 Linux
1

Anyone else with AMD 7040 FW13 and running Fedora 42 and KDE plasma seeing AMD GPU crashes with kernel 6.18.13-100.fc42.x86_64 that weren’t present before? I’ve gotten 2 over the last 3 days whereas before I was mostly stable.

Mar 01 11:36:06 fedora kwin_wayland[2876]: With the output of ‘sudo dmesg’ and ‘journalctl --user-unit plasma-kwin_wayland --boot 0’
Mar 01 11:36:06 fedora kwin_wayland[2876]: Please report this at
Mar 01 11:36:06 fedora kwin_wayland[2876]: The main thread was hanging temporarily!
Mar 01 11:36:06 fedora kwin_wayland[2876]: Pageflip timed out! This is a bug in the amdgpu kernel driver
Mar 01 11:36:06 fedora kernel: amdgpu 0000:c1:00.0: amdgpu: [drm] ERROR [CRTC:80:crtc-0] hw_done or flip_done timed out
Mar 01 11:36:06 fedora kernel: amdgpu 0000:c1:00.0: [drm] ERROR [CRTC:80:crtc-0] flip_done timed out
Mar 01 11:35:56 fedora kernel: amdgpu 0000:c1:00.0: [drm] ERROR dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
Mar 01 11:35:55 fedora kernel: amdgpu 0000:c1:00.0: [drm] ERROR dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
Mar 01 11:35:55 fedora kernel: amdgpu 0000:c1:00.0: [drm] ERROR dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data

I’m adding amdgpu.dcdebugmask=0x10 since searching suggests tons of problems with PSR and AMD GPU, just weird I never needed it before.

2

I saw this just once, with Arch Linux kernel 6.18.9 a few days ago. Haven’t seen it before or since (now on kernel 6.18.13).

I’m running Cinnamon desktop on Xorg. It seemed like everything froze, and then after a minute things were just unbearably laggy and unresponsive, but I was able to painfully collect logs and reboot.

Kernel logs were just being “spammed” with:

[287894.374215] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287894.634890] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287895.902596] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287896.139358] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287896.375798] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287896.634629] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287897.906065] amdgpu 0000:c1:00.0: [drm] *ERROR* Error queueing DMUB command: status=2
[287897.906069] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287898.142376] amdgpu 0000:c1:00.0: [drm] *ERROR* Error queueing DMUB command: status=2
[287898.142378] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[287898.378850] amdgpu 0000:c1:00.0: [drm] *ERROR* Error queueing DMUB command: status=2
[287898.378851] amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data

It seems to be rare enough for me that I’m not going to try to prevent it with amdgpu.dcdebugmask

3

Also noticed it on Fedora Silverblue 43 6.18.13-200.fc43.x86_64 on a FW13 AMD7040, I get similar logs:

Feb 23 19:05:31 ... kernel: amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
Feb 23 19:05:31 ... kernel: amdgpu 0000:c1:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
Feb 23 19:05:41 ... kernel: amdgpu 0000:c1:00.0: [drm] *ERROR* [CRTC:80:crtc-0] flip_done timed out

It started happening recently (not having seen this issue before), and even had it happen twice in one evening. There was another thread on the forum with this same error, naming GNOME Night Light feature as known potential cause for the issue. After turning off Night Light, I haven’t seen the crash anymore (anecdotal, but might help others).

5

I do use kde night color temp change, so that definitely could be a factor.

I haven’t yet seen it happen again since I killed PSR, just kind of weird it crops up now.