Bug 459053 - Screen corrupts and freezes on Wayland after clicking on taskbar
Summary: Screen corrupts and freezes on Wayland after clicking on taskbar
Status: RESOLVED UPSTREAM
Alias: None
Product: kwin
Classification: Plasma
Component: wayland-generic (show other bugs)
Version: 5.25.5
Platform: Other Linux
: NOR crash
Target Milestone: ---
Assignee: KWin default assignee
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2022-09-13 11:07 UTC by hbr
Modified: 2022-10-18 14:48 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments
dmesg output (50.77 KB, text/plain)
2022-09-13 11:07 UTC, hbr
Details
journal output (3.21 MB, text/plain)
2022-09-13 11:08 UTC, hbr
Details

Note You need to log in before you can comment on or make changes to this bug.
Description hbr 2022-09-13 11:07:40 UTC
Created attachment 152022 [details]
dmesg output

After updating to "KDE Frameworks Version: 5.98.0" the GPU driver crashed twice after clicking on the taskbar (using Wayland).

This did not happen with older versions of "KDE Frameworks" and this is the only KDE related update I can remember doing before the crashes occured.

So far it happenend randomly after hours of using the PC, but both times it was triggered when I clicked on the taskbar. It does not happen every time I click on the taskbar tho.

It happened the first time on the same day I updated "KDE Frameworks" (I did reboot after the update) and again 2 days later (today).

The screen corrupts and the the desktop environment locks up, audio keeps playing (I can still switch to TTY to reboot).
The first time the GPU driver managed to recover and so did the desktop environment, but not the second time.

dmesg says this:

[drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting for fences timed out!
[drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=3132014, emitted seq=3132016
[drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process kwin_wayland pid 1020 thread kwin_wayla:cs0 pid 1053
amdgpu 0000:03:00.0: amdgpu: GPU reset begin!
...

Attached is the dmesg / journal output when the crash happened until I rebooted from TTY. Since the desktop environment was locked up and all applications got terminated I can't provide any other traces for now.

Systeminfo:
Operating System: EndeavourOS
KDE Plasma Version: 5.25.5
KDE Frameworks Version: 5.98.0
Qt Version: 5.15.6
Kernel Version: 5.19.7-arch1-1 (64-bit)
Graphics Platform: Wayland
Processors: 20 × 12th Gen Intel® Core™ i7-12700K
Memory: 31,1 GiB of RAM
Graphics Processor: AMD Radeon RX 6800 XT
Comment 1 hbr 2022-09-13 11:08:56 UTC
Created attachment 152024 [details]
journal output
Comment 2 hbr 2022-10-18 14:29:27 UTC
After posting this report it happened a couple of times more, even on X11.

But after a slew of updates (Linux Kernel 6, Plasma 5.26, Mesa 22.2, KDE Frameworks 5.99, ...) it hasn't happened again.

And after looking around more this appears to be an issue with Mesa and amdgpu that just keeps popping up randomly in some kernel versions. So it just looked like it was related to KWin and the update to frameworks 5.98 for me by sheer coincidence.

So this can be closed.