SUMMARY *** Turning off external monitor with a laptop in clamshell mode using hybrid nvidia gpu, and letting computer go to sleep will cause kwin to crash. NOTE: If you are reporting a crash, please try to attach a backtrace with debug symbols. See https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports *** Unable to backtrace as `sudo coredumpctl gdb` will cause my root partition to run out of memory. Here is a reddit thread on the topic: https://old.reddit.com/r/kde/comments/1b9t888/coredump_kwin_wayland_crashes_plasma_6/ STEPS TO REPRODUCE 1. Make sure there is no sleep blocking going on 2. Turn off external monitor connected via usbc 3. Let computer sleep 4. kwin_wayland will crash and present a dialog: https://imgur.com/a/dpoDFSM OBSERVED RESULT Dr Konqi presents a crash report EXPECTED RESULT Normally login and continue doing what i was doing without crashes SOFTWARE/OS VERSIONS Operating System: Arch Linux KDE Plasma Version: 6.0.1 KDE Frameworks Version: 6.0.0 Qt Version: 6.6.2 Kernel Version: 6.7.9-arch1-1 (64-bit) Graphics Platform: Wayland Processors: 12 × Intel® Core™ i7-10750H CPU @ 2.60GHz Memory: 23.3 GiB of RAM Graphics Processor: Mesa Intel® UHD Graphics Manufacturer: LENOVO Product Name: 81Y6 System Version: Lenovo Legion 5 15IMH05H Nvidia 2060 mobile graphics
Unfortunately we would need a backtrace with debug symbols :( See https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports for instructions. Maybe adding some swap would fix the OOM problem?
(In reply to fanzhuyifan from comment #1) > Unfortunately we would need a backtrace with debug symbols :( > > See > https://community.kde.org/Guidelines_and_HOWTOs/Debugging/ > How_to_create_useful_crash_reports for instructions. > > Maybe adding some swap would fix the OOM problem? Its not a ram issue - I have a 10GB swap file. I only have 3 Gbs on root partition remaining and the backtrace generated fails with "SEGFAULT INTERNAL ERROR" that seems to be related to runninng out of memory. Is it possible to have the backtrace saved to my /home/ partition?
FYI this is for `sudo coredumpctl gdb`, I don't know how to do anything else as the instructions don't make sense to me.
(In reply to agapp11@hotmail.com from comment #2) > (In reply to fanzhuyifan from comment #1) > > Unfortunately we would need a backtrace with debug symbols :( > > > > See > > https://community.kde.org/Guidelines_and_HOWTOs/Debugging/ > > How_to_create_useful_crash_reports for instructions. > > > > Maybe adding some swap would fix the OOM problem? > > Its not a ram issue - I have a 10GB swap file. I only have 3 Gbs on root > partition remaining and the backtrace generated fails with "SEGFAULT > INTERNAL ERROR" that seems to be related to runninng out of memory. > > Is it possible to have the backtrace saved to my /home/ partition? I have finally figured out how to create a backtrace. Or at least think I have. It is attached, let me know if it worked
Created attachment 166954 [details] gdb.log file
Created attachment 166956 [details] gdb.txt with backtrace I have managed to get a backtrace to show up in the .txt. Let me know if anyhting else is needed.
(In reply to agapp11@hotmail.com from comment #6) > Created attachment 166956 [details] > gdb.txt with backtrace > > I have managed to get a backtrace to show up in the .txt. Let me know if > anyhting else is needed. This is nice and even more detailed that what I would expect normally ;) But I am not sure what this trace is for. If you use gdb, it should mention a segfault or a crash somewhere but this does not seem to be the case. With gdb, you need to launch or attach to kwin process, and get it to crash, then we will have a useful trace in gdb output. Or with coredumpctl after kwin has quit after a crash, you can run `coredumpctl info -1 kwin_wayland` to get essentially the same thing.
(In reply to Méven Car from comment #7) > (In reply to agapp11@hotmail.com from comment #6) > > Created attachment 166956 [details] > > gdb.txt with backtrace > > > > I have managed to get a backtrace to show up in the .txt. Let me know if > > anyhting else is needed. > > This is nice and even more detailed that what I would expect normally ;) > > But I am not sure what this trace is for. If you use gdb, it should mention > a segfault or a crash somewhere but this does not seem to be the case. > > With gdb, you need to launch or attach to kwin process, and get it to crash, > then we will have a useful trace in gdb output. > > Or with coredumpctl after kwin has quit after a crash, you can run > `coredumpctl info -1 kwin_wayland` to get essentially the same thing. I can't consistently reproduce the crash so I can't use gdb. It finally crashed again so using the coredump command I have attached the output
Created attachment 167081 [details] Crash of kwin immediately after the crash after waking computer from sleep and turning on monitor One thing I noticed is the crash of kwin only happens when Optimus-manager is running in the tray. I have not observed it happen without it.
(In reply to agapp11@hotmail.com from comment #9) > Created attachment 167081 [details] > Crash of kwin immediately after the crash after waking computer from sleep > and turning on monitor > > One thing I noticed is the crash of kwin only happens when Optimus-manager > is running in the tray. I have not observed it happen without it. This lacks debuginfo that would make it 100 times more useful, luckily that can be added easily. For arch/Endeavour and reading your precedent output it should be as simple as running `DEBUGINFOD_URLS="https://debuginfod.archlinux.org/" coredumpctl gdb 1973961`, this will take a little while and download a bunch of data. Once the last line is a prompt `(gdb) `, type in `bt` then hit enter and you will have a good output. You can quit gdb with Ctrl + D or close your terminal afterwards.
(In reply to Méven Car from comment #10) > (In reply to agapp11@hotmail.com from comment #9) > > Created attachment 167081 [details] > > Crash of kwin immediately after the crash after waking computer from sleep > > and turning on monitor > > > > One thing I noticed is the crash of kwin only happens when Optimus-manager > > is running in the tray. I have not observed it happen without it. > > This lacks debuginfo that would make it 100 times more useful, luckily that > can be added easily. > For arch/Endeavour and reading your precedent output it should be as simple > as running `DEBUGINFOD_URLS="https://debuginfod.archlinux.org/" coredumpctl > gdb 1973961`, this will take a little while and download a bunch of data. > Once the last line is a prompt `(gdb) `, type in `bt` then hit enter and you > will have a good output. > You can quit gdb with Ctrl + D or close your terminal afterwards. I finally got another crash. I tried using the command you specified but it did not work as it gave an output of "no match found". Using `sudo DEBUGINFOD_URLS="https://debuginfod.archlinux.org/" coredumpctl gdb kwin_wayland` I got the output attached
Created attachment 167259 [details] Coredump using sudo DEBUGINFOD_URLS="https://debuginfod.archlinux.org/" coredumpctl gdb kwin_wayland I hope it is correct and useful this time. Please let me know
(In reply to agapp11@hotmail.com from comment #12) > Created attachment 167259 [details] > Coredump using sudo DEBUGINFOD_URLS="https://debuginfod.archlinux.org/" > coredumpctl gdb kwin_wayland > > I hope it is correct and useful this time. Please let me know Yes fantastic ! (gdb) bt #0 doActivate<false> (sender=0x723878002840, signal_index=3, argv=0x7fffb699e490) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/tools/qscopedpointer.h:110 #1 0x00007238c73e8c07 in KWin::DrmGpu::activeChanged (_t1=<optimized out>, this=0x723878002840) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_drm_gpu.cpp:333 #2 KWin::DrmGpu::setActive (this=0x723878002840, active=true) at /usr/src/debug/kwin/kwin-6.0.2/src/backends/drm/drm_gpu.cpp:734 #3 0x00007238c4b90ca9 in QtPrivate::QSlotObjectBase::call (a=0x7fffb699e5d0, r=0x7238b8009810, this=0x60e9eb54f5a0, this=<optimized out>, r=<optimized out>, a=<optimized out>) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qobjectdefs_impl.h:433 #4 doActivate<false> (sender=0x60e9eb54a960, signal_index=5, argv=0x7fffb699e5d0) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qobject.cpp:4039 #5 0x00007238c717edd7 in KWin::Session::deviceResumed (this=<optimized out>, _t1=<optimized out>) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session.cpp:236 #6 0x00007238c718b9bb in KWin::LogindSession::handleResumeDevice (fileDescriptor=..., minor=<optimized out>, major=<optimized out>, this=0x60e9eb54a960) at /usr/src/debug/kwin/kwin-6.0.2/src/core/session_logind.cpp:330 #7 KWin::LogindSession::qt_static_metacall (_o=0x60e9eb54a960, _id=<optimized out>, _a=0x7fffb699e848, _c=<optimized out>) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session_logind.cpp:176 #8 0x00007238c718bbf1 in KWin::LogindSession::qt_static_metacall (_a=0x7fffb699e848, _id=0, _c=QMetaObject::InvokeMetaMethod, _o=0x60e9eb54a960) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session_logind.cpp:170 #9 KWin::LogindSession::qt_metacall (this=0x60e9eb54a960, _c=QMetaObject::InvokeMetaMethod, _id=0, _a=0x7fffb699e848) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session_logind.cpp:216 #10 0x00007238c61130ce in QDBusConnectionPrivate::deliverCall (this=0x7238b8009370, object=0x60e9eb54a960, msg=..., metaTypes=<optimized out>, slotIdx=9) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/dbus/qdbusintegrator.cpp:977 #11 0x00007238c4b7c2c7 in QObject::event (this=0x60e9eb54a960, e=0x7238b8010dc0) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qobject.cpp:1437 #12 0x00007238c5af438b in QApplicationPrivate::notify_helper (this=<optimized out>, receiver=0x60e9eb54a960, e=0x7238b8010dc0) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/widgets/kernel/qapplication.cpp:3296 #13 0x00007238c4b39818 in QCoreApplication::notifyInternal2 (receiver=0x60e9eb54a960, event=event@entry=0x7238b8010dc0) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qcoreapplication.cpp:1121 #14 0x00007238c4b39b9b in QCoreApplication::sendEvent (event=0x7238b8010dc0, receiver=<optimized out>) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qcoreapplication.cpp:1539 #15 QCoreApplicationPrivate::sendPostedEvents (receiver=0x0, event_type=0, data=0x60e9eb412430) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qcoreapplication.cpp:1901 #16 0x00007238c4c9f18f in QEventDispatcherUNIX::processEvents (this=0x60e9eb47f460, flags=flags@entry=...) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qeventdispatcher_unix.cpp:431 #17 0x00007238c55b26e2 in QUnixEventDispatcherQPA::processEvents (this=<optimized out>, flags=...) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/gui/platform/unix/qunixeventdispatcher.cpp:27 #18 0x00007238c4b43d6e in QEventLoop::processEvents (flags=..., this=0x7fffb699edb0) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qeventloop.cpp:100 #19 QEventLoop::exec (this=0x7fffb699edb0, flags=...) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qeventloop.cpp:182 #20 0x00007238c4b3c2b8 in QCoreApplication::exec () at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/global/qflags.h:74 #21 0x00007238c5af0f0a in QApplication::exec () at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/widgets/kernel/qapplication.cpp:2574 #22 0x000060e9e99d3dee in main (argc=<optimized out>, argv=<optimized out>) at /usr/src/debug/kwin/kwin-6.0.2/src/main_wayland.cpp:609
A possibly relevant merge request was started @ https://invent.kde.org/plasma/kwin/-/merge_requests/5433
*** Bug 483125 has been marked as a duplicate of this bug. ***
Git commit aa465efa5178c240059f6d974440cfb85bea99d2 by Méven Car. Committed on 15/03/2024 at 16:31. Pushed by meven into branch 'master'. wayland: DrmLeaseDevice, use Q_OBJECT macro M +1 -0 src/wayland/drmlease_v1_p.h https://invent.kde.org/plasma/kwin/-/commit/aa465efa5178c240059f6d974440cfb85bea99d2
Git commit 84412b22df01c5b00074e486cad3a432b04c953d by Méven Car. Committed on 15/03/2024 at 16:51. Pushed by meven into branch 'Plasma/6.0'. wayland: DrmLeaseDevice, use Q_OBJECT macro (cherry picked from commit aa465efa5178c240059f6d974440cfb85bea99d2) M +1 -0 src/wayland/drmlease_v1_p.h https://invent.kde.org/plasma/kwin/-/commit/84412b22df01c5b00074e486cad3a432b04c953d
Does this look like the same cause affecting https://bugs.kde.org/show_bug.cgi?id=483243 (possible duplicate)
(In reply to andy from comment #18) > Does this look like the same cause affecting > https://bugs.kde.org/show_bug.cgi?id=483243 (possible duplicate) This is not not a duplicate but looks a bit similar.
It seems that I am still getting kwin_wayland crashes even after updating to plasma 6.03 on arch. Backtrace is attached
Created attachment 167957 [details] Kwin_wayland crash on Plasma 6.03
(In reply to agapp11@hotmail.com from comment #21) > Created attachment 167957 [details] > Kwin_wayland crash on Plasma 6.03 [Current thread is 1 (Thread 0x7dac5a90ba00 (LWP 2243))] (gdb) bt #0 doActivate<false> (sender=0x5c1be80bd0e0, signal_index=3, argv=0x7ffe55023cd0) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/tools/qscopedpointer.h:110 #1 0x00007dac625e9447 in KWin::DrmGpu::activeChanged (_t1=<optimized out>, this=0x5c1be80bd0e0) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_drm_gpu.cpp:333 #2 KWin::DrmGpu::setActive (this=0x5c1be80bd0e0, active=true) at /usr/src/debug/kwin/kwin-6.0.3.1/src/backends/drm/drm_gpu.cpp:736 #3 0x00007dac5fd90ca9 in QtPrivate::QSlotObjectBase::call (a=0x7ffe55023e10, r=0x7dac54009800, this=0x5c1be7fc9f60, this=<optimized out>, r=<optimized out>, a=<optimized out>) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qobjectdefs_impl.h:433 #4 doActivate<false> (sender=0x5c1be7fee6e0, signal_index=5, argv=0x7ffe55023e10) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qobject.cpp:4039 #5 0x00007dac6237ee37 in KWin::Session::deviceResumed (this=<optimized out>, _t1=<optimized out>) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session.cpp:236 #6 0x00007dac6238ba1b in KWin::LogindSession::handleResumeDevice (fileDescriptor=..., minor=<optimized out>, major=<optimized out>, this=0x5c1be7fee6e0) at /usr/src/debug/kwin/kwin-6.0.3.1/src/core/session_logind.cpp:330 #7 KWin::LogindSession::qt_static_metacall (_o=0x5c1be7fee6e0, _id=<optimized out>, _a=0x7ffe55024088, _c=<optimized out>) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session_logind.cpp:176 #8 0x00007dac6238bc51 in KWin::LogindSession::qt_static_metacall (_a=0x7ffe55024088, _id=0, _c=QMetaObject::InvokeMetaMethod, _o=0x5c1be7fee6e0) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session_logind.cpp:170 #9 KWin::LogindSession::qt_metacall (this=0x5c1be7fee6e0, _c=QMetaObject::InvokeMetaMethod, _id=0, _a=0x7ffe55024088) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_session_logind.cpp:216 #10 0x00007dac613130ce in QDBusConnectionPrivate::deliverCall (this=0x7dac54009360, object=0x5c1be7fee6e0, msg=..., metaTypes=<optimized out>, slotIdx=9) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/dbus/qdbusintegrator.cpp:977 #11 0x00007dac5fd7c2c7 in QObject::event (this=0x5c1be7fee6e0, e=0x7dac54025100) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qobject.cpp:1437 #12 0x00007dac60cf438b in QApplicationPrivate::notify_helper (this=<optimized out>, receiver=0x5c1be7fee6e0, e=0x7dac54025100) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/widgets/kernel/qapplication.cpp:3296 #13 0x00007dac5fd39818 in QCoreApplication::notifyInternal2 (receiver=0x5c1be7fee6e0, event=event@entry=0x7dac54025100) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qcoreapplication.cpp:1121 #14 0x00007dac5fd39b9b in QCoreApplication::sendEvent (event=0x7dac54025100, receiver=<optimized out>) at /usr/src/debug/qt6-base/qtbase-everywhere-src-6.6.2/src/corelib/kernel/qcoreapplication.cpp:1539
A possibly relevant merge request was started @ https://invent.kde.org/plasma/kwin/-/merge_requests/5539
*** Bug 485182 has been marked as a duplicate of this bug. ***
Any update on this? It is quite bothersome to have your desktop and apps crash every time you wake from sleep? Is there anything else I need to include? Perhaps another backtrace?
I got an email you guys need a backtrace? I have attached one. Please let me know if there is anything else I can do.
Created attachment 169705 [details] Plasma 6.0.4 Backtrace for Wayland Kwin Crash
A possibly relevant merge request was started @ https://invent.kde.org/plasma/kwin/-/merge_requests/5760
Git commit b232e803190a408ec22bb94bfa0f4ef0cbba8eed by Xaver Hugl. Committed on 22/05/2024 at 20:33. Pushed by zamundaaa into branch 'master'. backends/drm: don't update outputs from within DrmGpu DrmBackend::updateOutputs may delete DrmGpu objects, so calling it from within a DrmGpu method can cause problems FIXED-IN: 6.1.0 M +4 -1 src/backends/drm/drm_backend.cpp M +0 -2 src/backends/drm/drm_gpu.cpp https://invent.kde.org/plasma/kwin/-/commit/b232e803190a408ec22bb94bfa0f4ef0cbba8eed