Bug 511874 - PlasmaShell crashed on log in, with inconsistent QPalette focus states in DrKonqi, after ANR'ing before (and during) the previous logout
Summary: PlasmaShell crashed on log in, with inconsistent QPalette focus states in DrK...
Status: RESOLVED NOT A BUG
Alias: None
Product: plasmashell
Classification: Plasma
Component: generic-crash (other bugs)
Version First Reported In: master
Platform: Fedora RPMs Linux
: NOR crash
Target Milestone: 1.0
Assignee: Plasma Bugs List
URL:
Keywords: drkonqi
Depends on:
Blocks:
 
Reported: 2025-11-09 17:33 UTC by Roke Julian Lockhart Beedell
Modified: 2025-11-13 16:44 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed/Implemented In:
Sentry Crash Report: https://crash-reports.kde.org/organizations/kde/issues/290964/events/e3f7b4bc9a3041c6a0b388b850eb46fa/


Attachments
New crash information added by DrKonqi (249.32 KB, text/plain)
2025-11-09 17:33 UTC, Roke Julian Lockhart Beedell
Details
A Screenshot Of The Inconsistent Window (29.89 KB, image/png)
2025-11-09 17:55 UTC, Roke Julian Lockhart Beedell
Details
A Screencast That Demonstrates The Inconsistency Of The Window's Colour Scheme Upon Selection (805.49 KB, video/webm)
2025-11-09 17:56 UTC, Roke Julian Lockhart Beedell
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Roke Julian Lockhart Beedell 2025-11-09 17:33:51 UTC
Application: plasmashell (6.5.2)

ApplicationNotResponding [ANR]: false
Qt Version: 6.10.0
Frameworks Version: 6.19.0
Operating System: Linux 6.17.7-300.fc43.x86_64 x86_64
Windowing System: Wayland
Distribution: "Fedora Linux 43 (KDE Plasma Desktop Edition)"
DrKonqi: 6.5.2 [CoredumpBackend]

-- Information about the crash:
Without any noticeable interaction from me, the sole visible panel hung. (I neglected to confirm whether the desktop was impacted, but I presume so.) To remediate this, I issued the "Log out" command via KRunner. This appeared to do nothing, so I repeated it. The sound for the first invocation played 15 s late, but I solely heard it once. 3 s after that, I invoked the command again, and approximately 5 s after that, I was dumped to SDDM. When I logged back in, I solely saw a black screen for 3 s. Then, DrKonqi appeared. However, as a screencast, that I shall upload after this, demonstrates, its QPalette is `breezedark` when focussed. However, it utilises Breeze Light when not focussed, as all other applications are, whether focused or not.

The reporter is unsure if this crash is reproducible.

-- Backtrace (Reduced):
#5  syscall () at ../sysdeps/unix/sysv/linux/x86_64/syscall.S:38
#6  0x00007f151d2f940a in sys_futex (addr1=0x557071df2794, op=9, val1=2, timeout=0x0, addr2=0x0, val3=-1) at ../src/util/futex.c:43
#7  futex_wait (addr=addr@entry=0x557071df2794, value=value@entry=2, timeout=timeout@entry=0x0) at ../src/util/futex.c:55
#8  0x00007f151d306830 in do_futex_fence_wait (fence=fence@entry=0x557071df2794, timeout=timeout@entry=false, abs_timeout=abs_timeout@entry=0) at ../src/util/u_queue.c:131
#9  0x00007f151d307a0c in _util_queue_fence_wait (fence=0x557071df2794) at ../src/util/u_queue.c:146
#10 util_queue_fence_wait (fence=0x557071df2794) at ../src/util/u_queue.h:181
#11 util_queue_fence_wait (fence=0x557071df2794) at ../src/util/u_queue.h:178
#12 util_queue_finish (queue=queue@entry=0x55706ae24a30) at ../src/util/u_queue.c:722
#13 0x00007f151d760aaa in si_set_debug_callback (ctx=0x557071fbba90, cb=0x0) at ../src/gallium/drivers/radeonsi/si_pipe.c:449
#14 0x00007f151d761318 in si_destroy_context (context=0x557071fbba90) at ../src/gallium/drivers/radeonsi/si_pipe.c:202
#15 0x00007f151d546a31 in tc_destroy (_pipe=0x5570724d4f80) at ../src/gallium/auxiliary/util/u_threaded_context.c:5400
#16 0x00007f151cec7260 in st_destroy_context_priv (st=st@entry=0x5570726d77d0, destroy_pipe=destroy_pipe@entry=true) at ../src/mesa/state_tracker/st_context.c:356
#17 0x00007f151cec8d4d in st_destroy_context (st=0x5570726d77d0) at ../src/mesa/state_tracker/st_context.c:967
#18 0x00007f151ce2993b in dri_destroy_context (ctx=0x55706b8b1480) at ../src/gallium/frontends/dri/dri_context.c:267
#19 0x00007f151ce2d6be in driDestroyContext (ctx=<optimized out>) at ../src/gallium/frontends/dri/dri_util.c:631
#20 0x00007f152df65669 in dri2_destroy_context (disp=<optimized out>, ctx=0x557070b43a10) at ../src/egl/drivers/dri2/egl_dri2.c:1271


Reported using DrKonqi
Comment 1 Roke Julian Lockhart Beedell 2025-11-09 17:33:52 UTC
Created attachment 186651 [details]
New crash information added by DrKonqi

DrKonqi auto-attaching complete backtrace.
Comment 3 Roke Julian Lockhart Beedell 2025-11-09 17:55:08 UTC
Created attachment 186652 [details]
A Screenshot Of The Inconsistent Window

Apologies about the blur. See https://bugs.kde.org/show_bug.cgi?id=510982#c6.
Comment 4 Roke Julian Lockhart Beedell 2025-11-09 17:56:31 UTC
Created attachment 186654 [details]
A Screencast That Demonstrates The Inconsistency Of The Window's Colour Scheme Upon Selection

I have attached the screencast, as I aforestated.
Comment 5 Roke Julian Lockhart Beedell 2025-11-09 18:07:56 UTC
Comment on attachment 186654 [details]
A Screencast That Demonstrates The Inconsistency Of The Window's Colour Scheme Upon Selection

Perhaps, this inconsistent appearance correlates with https://discuss.kde.org/t/settings-day-night-cycle-isnt-working/41524?u=rokejulianlockhart, if this bug caused it.
Comment 6 TraceyC 2025-11-13 01:44:13 UTC
Thanks for the bug report! Unfortunately it reports multiple distinct issues, which will make it not actionable. See https://community.kde.org/Get_Involved/Issue_Reporting#Multiple_issues_in_a_single_Bugzilla_ticket for more explanation.

The crash has the same backtrace as bug 507936, and doesn't need to be reported again.

Can I ask you to submit a new bug report for each specific issue that isn't the crash? Thanks again!
Comment 7 Roke Julian Lockhart Beedell 2025-11-13 03:08:18 UTC
(In reply to TraceyC from comment #6)

> Thanks for the bug report! Unfortunately it reports multiple distinct
> issues, which will make it not actionable. See
> https://community.kde.org/Get_Involved/
> Issue_Reporting#Multiple_issues_in_a_single_Bugzilla_ticket for more
> explanation.
> 
> Can I ask you to submit a new bug report for each specific issue that isn't
> the crash? Thanks again!

Thanks, too. However, I was only really reporting the crash, unless you think that the other information is actionable, considering that I don't see a way to reproduce it. I'm a little wary to file bugs that'll likely be closed as WAITINGFORINFO, since I presume it's not a great use of your time.

> The crash has the same backtrace as bug 507936, and doesn't need to be
> reported again.

Got any ideas on how to tell what it is before I report? Abrt tells me when a trace matches something I've already reported, but Konqi doesn't. Unless I just don't report ANRs for PlasmaShell, I'm not sure how to tell whether one's a Mesa bug or not. I can't really read traces.
Comment 8 TraceyC 2025-11-13 16:44:37 UTC
(In reply to Roke Julian Lockhart Beedell from comment #7)
> Thanks, too. However, I was only really reporting the crash, unless you
> think that the other information is actionable, considering that I don't see
> a way to reproduce it. I'm a little wary to file bugs that'll likely be
> closed as WAITINGFORINFO, since I presume it's not a great use of your time.

Thank you for being considerate. If you can find a way to reliably reproduce the theming issue with Dr Konqi, it would be worth reporting. Without a way to reproduce, I'm afraid you're right, we wouldn't be able to do much.

> > The crash has the same backtrace as bug 507936, and doesn't need to be
> > reported again.
> 
> Got any ideas on how to tell what it is before I report? Abrt tells me when
> a trace matches something I've already reported, but Konqi doesn't. Unless I
> just don't report ANRs for PlasmaShell, I'm not sure how to tell whether
> one's a Mesa bug or not. I can't really read traces.

Unfortunately, the only way to tell if a crash is the same as another is to read the backtrace. In the case of this and the other report, the backtraces are identical, if you were interested in seeing how we compare. They are an excellent example to learn from.  We also have a howto on reading the backtraces, if you're interested 
https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging/Identifying_duplicates#Identifying_duplicates_(crashes)_:_C++_Backtraces