SUMMARY kwin_x11 was pegged at 100% and the GUI was somewhat unresponsive (mouse cursor still moved). "$ kwin_x11 --replace" resulted in a coredump. STEPS TO REPRODUCE 1. Normal every day use. OBSERVED RESULT Coredump EXPECTED RESULT SOFTWARE/OS VERSIONS Windows: macOS: Linux/KDE Plasma: Arch Linux (available in About System) KDE Plasma Version: 5.16.5 KDE Frameworks Version: 5.62.0 Qt Version: 5.13.1 ADDITIONAL INFORMATION While I was browsing in Chrome, action responses froze. I could move my mouse around, but there would be no action from any clicking or even ALT-Tab. I switched terminals (luckily?), and saw kwin_x11 was pegged at 100% CPU usage, and bouncing around among the cores. I tried to "$ kwin_x11 --replace", but after a short wait, there was a coredump: ***** Oct 17 21:51:31 computername systemd-coredump[1598113]: Process 1598070 (kwin_x11) of user 1000 dumped core. Stack trace of thread 1598078: #0 0x00007b753249af25 raise (libc.so.6) #1 0x00007b75320f26b0 _ZN6KCrash19defaultCrashHandlerEi (libKF5Crash.so.5) #2 0x00007b753249afb0 __restore_rt (libc.so.6) #3 0x00007b753249af25 raise (libc.so.6) #4 0x00007b7532484897 abort (libc.so.6) #5 0x00007b7530e2c8f8 _ZNK14QMessageLogger5fatalEPKcz (libQt5Core.so.5) #6 0x00007b7529374cec n/a (KWinX11Platform.so) #7 0x00007b7531061b70 _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5) #8 0x00007b753106eb78 _ZN6QTimer7timeoutENS_14QPrivateSignalE (libQt5Core.so.5) #9 0x00007b75310623e5 _ZN7QObject5eventEP6QEvent (libQt5Core.so.5) #10 0x00007b75319f14f5 _ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent (libQt5Widgets.so.5) #11 0x00007b75319fae11 _ZN12QApplication6notifyEP7QObjectP6QEvent (libQt5Widgets.so.5) #12 0x00007b75310359c2 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5) #13 0x00007b753108cd9d _ZN14QTimerInfoList14activateTimersEv (libQt5Core.so.5) #14 0x00007b753108b169 _ZN20QEventDispatcherUNIX13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5) #15 0x00007b75310344ec _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5) #16 0x00007b7530e66385 _ZN7QThread4execEv (libQt5Core.so.5) #17 0x00007b7530e675b0 n/a (libQt5Core.so.5) #18 0x00007b752ffd54cf start_thread (libpthread.so.0) #19 0x00007b753255e2d3 __clone (libc.so.6) Stack trace of thread 1598074: #0 0x00007b752ffdbf7a pthread_cond_timedwait@@GLIBC_2.3.2 (libpthread.so.0) #1 0x00007b7530e6d5cc _ZN14QWaitCondition4waitEP6QMutex14QDeadlineTimer (libQt5Core.so.5) #2 0x00007b7530e6d719 _ZN14QWaitCondition4waitEP6QMutexm (libQt5Core.so.5) #3 0x00007b7530e6ab9b n/a (libQt5Core.so.5) #4 0x00007b7530e675b0 n/a (libQt5Core.so.5) #5 0x00007b752ffd54cf start_thread (libpthread.so.0) #6 0x00007b753255e2d3 __clone (libc.so.6) Stack trace of thread 1598071: #0 0x00007b75325539ef __poll (libc.so.6) #1 0x00007b7530d87630 n/a (libxcb.so.1) #2 0x00007b7530d892db xcb_wait_for_event (libxcb.so.1) #3 0x00007b752a0d6101 n/a (libQt5XcbQpa.so.5) #4 0x00007b7530e675b0 n/a (libQt5Core.so.5) #5 0x00007b752ffd54cf start_thread (libpthread.so.0) #6 0x00007b753255e2d3 __clone (libc.so.6) Stack trace of thread 1598070: #0 0x00007b752ffdbc45 pthread_cond_wait@@GLIBC_2.3.2 (libpthread.so.0) #1 0x00007b7530d8783b n/a (libxcb.so.1) #2 0x00007b7530d88fef n/a (libxcb.so.1) #3 0x00007b7530d89162 xcb_wait_for_reply64 (libxcb.so.1) #4 0x00007b752ef20269 _XReply (libX11.so.6) #5 0x00007b752d551aa4 n/a (libGLX.so.0) #6 0x00007b752d551bfc n/a (libGLX.so.0) #7 0x00007b752d550708 n/a (libGLX.so.0) #8 0x00007b752d550c28 n/a (libGLX.so.0) #9 0x00007b752d54cab3 glXGetClientString (libGLX.so.0) #10 0x00007b75280d7803 n/a (libqxcb-glx-integration.so) #11 0x00007b75280d8199 n/a (libqxcb-glx-integration.so) #12 0x00007b7529373310 _ZN4KWin21X11StandalonePlatform19createOpenGLBackendEv (KWinX11Platform.so) #13 0x00007b751ada3422 _ZN4KWin11SceneOpenGL11createSceneEP7QObject (KWinSceneOpenGL.so) #14 0x00007b751ada6bdb _ZNK4KWin13OpenGLFactory6createEP7QObject (KWinSceneOpenGL.so) #15 0x00007b75322c3015 _ZN4KWin10Compositor33slotCompositingOptionsInitializedEv (libkwin.so.5) #16 0x00007b75322c3fdc _ZN4KWin10Compositor5setupEv (libkwin.so.5) #17 0x00007b753106244a _ZN7QObject5eventEP6QEvent (libQt5Core.so.5) #18 0x00007b75319f14f5 _ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent (libQt5Widgets.so.5) #19 0x00007b75319fae11 _ZN12QApplication6notifyEP7QObjectP6QEvent (libQt5Widgets.so.5) #20 0x00007b75310359c2 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5) #21 0x00007b7531038739 _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt5Core.so.5) #22 0x00007b753108ada5 _ZN20QEventDispatcherUNIX13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5) #23 0x00007b752a0d74df n/a (libQt5XcbQpa.so.5) #24 0x00007b75310344ec _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5) #25 0x00007b753103c326 _ZN16QCoreApplication4execEv (libQt5Core.so.5) #26 0x00007b753262f588 kdemain (libkdeinit5_kwin_x11.so) #27 0x00007b7532486153 __libc_start_main (libc.so.6) #28 0x000058f2a7e2605e _start (kwin_x11) Stack trace of thread 1598077: #0 0x00007b7532553ae6 ppoll (libc.so.6) #1 0x00007b7531089973 _Z12qt_safe_pollP6pollfdmPK8timespec (libQt5Core.so.5) #2 0x00007b753108b003 _ZN20QEventDispatcherUNIX13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5) #3 0x00007b75310344ec _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5) #4 0x00007b7530e66385 _ZN7QThread4execEv (libQt5Core.so.5) #5 0x00007b752f75e849 n/a (libQt5Qml.so.5) #6 0x00007b7530e675b0 n/a (libQt5Core.so.5) #7 0x00007b752ffd54cf start_thread (libpthread.so.0) #8 0x00007b753255e2d3 __clone (libc.so.6) Stack trace of thread 1598072: #0 0x00007b7532553ae6 ppoll (libc.so.6) #1 0x00007b7531089973 _Z12qt_safe_pollP6pollfdmPK8timespec (libQt5Core.so.5) #2 0x00007b753108b003 _ZN20QEventDispatcherUNIX13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5) #3 0x00007b75310344ec _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5) #4 0x00007b7530e66385 _ZN7QThread4execEv (libQt5Core.so.5) #5 0x00007b752f18cb37 n/a (libQt5DBus.so.5) #6 0x00007b7530e675b0 n/a (libQt5Core.so.5) #7 0x00007b752ffd54cf start_thread (libpthread.so.0) #8 0x00007b753255e2d3 __clone (libc.so.6) ***** SIGTERM didn't kill the process, but SIGKILL did. Yes, I'll upgrade to 5.17, but thought this may be something that could very well have carried over to later versions, and continue to do so if not addressed (even if it is a fluke). I am a bit doubtful the coredump would show the cause of the hangup, but rather the failed attempt at replacing the instance.
Does this issue happen in 5.17?
It was a one-off event that never happened before 5.16.5 and only that one time in that version and not in the time I've used 5.17.1. I WILL say though that so far in this time since installing 5.17.1, plasmashell has disappeared 3 times. By that, I mean the process is still up and running (no CPU pegging), but the entire display disappears (no desktop or status bar). I had to "--replace" it each time. And the icing on top of that nasty cake is that NOTHING shows in the log about any type of crash.
> is that NOTHING shows in the log about any type of crash. Could you please attach kwin's log to this bug report? KWin prints a warning message before crashing. Also, it appears that the backtrace in the first comment lacks debug symbols. Can you provide a new one with debug symbols? Given that you use Arch Linux, you need to repackage kwin. [1] [1] https://wiki.archlinux.org/index.php/Debug_-_Getting_Traces#CMAKE_(KDE)_applications
Uh, is there some special location for the kwin log that would still exist after the past few weeks? The nearest entry in the journal was: ***** Oct 17 21:47:50 computerName kwin_x11[1163]: qt.qpa.xcb: QXcbConnection: XCB error: 3 (BadWindow), sequence: 45756, resource id: 41328738, major code: 20 (GetProperty), minor code: 0 ***** but that isn't uncommon. Like I had mentioned before, it was a one-off event, and it looks like what you're asking from me is possibly going to cause me a lot of grief... I'd like to think I won't have any problem with doing a makepkg for kwin from https://download.kde.org/stable/plasma/5.16.5/, but am I going to end up screwing myself over with all the other packages I have being newer? Besides, wouldn't I still need to have the coredump file? I've had SO many crashes with KDE related stuff over the past month, mainly with kdeinit5, but I think that was related to thumbnails. That all settled down after I upgraded and now I'm contending with crashes from Kodi. So I've got like 4 crash files for Kodi, and everything else shows as "missing" in coredumpctl. I have a feeling, I'm going to need a lot more guidance than that one link you provided, and even then, I already mentioned it was a one-off and have no clue how to force the crash.
Dear Bug Submitter, This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information as soon as possible and set the bug status as REPORTED. Due to regular bug tracker maintenance, if the bug is still in NEEDSINFO status with no change in 30 days the bug will be closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging If you have already provided the requested information, please mark the bug as REPORTED so that the KDE team knows that the bug is ready to be confirmed. Thank you for helping us make KDE software even better for everyone!
Well gee "Bug Janitor Service" (aka bot), it would seem that I had made an appropriate response, but received no additional guidance in order to provide the "NEEDSINFO BACKTRACE". Is there a bot for that (I say sarcastically)? I figured the "Status" was meant for devs to change according to their needs, but sure, I can change it to "REPORTED". However, I would be unable to confidently set it to "RESOLVED" or "WORKSFORME" since it is unknown if the bug was ever found and corrected. I'm doing whatever part I'm able to. At least the issue still hasn't come up again (yet)... Though a different issue comes up every now and again where plasmashell rendering seems to go kaput, and I have to "plasmashell --replace" to bring it up again. It's been that way for over a month, but I don't believe I've posted about it. The thing is, there doesn't seem to be any crash involved. The last time it happened, I was able to ALT-TAB through windows, but there was no application panel or anything.
Oh.. On a second look, it appears like OpenGL freeze protection kicked in. That's why kwin crashed. It appears like kwin froze when calling glXGetClientString() #0 0x00007b752ffdbc45 pthread_cond_wait@@GLIBC_2.3.2 (libpthread.so.0) #1 0x00007b7530d8783b n/a (libxcb.so.1) #2 0x00007b7530d88fef n/a (libxcb.so.1) #3 0x00007b7530d89162 xcb_wait_for_reply64 (libxcb.so.1) #4 0x00007b752ef20269 _XReply (libX11.so.6) #5 0x00007b752d551aa4 n/a (libGLX.so.0) #6 0x00007b752d551bfc n/a (libGLX.so.0) #7 0x00007b752d550708 n/a (libGLX.so.0) #8 0x00007b752d550c28 n/a (libGLX.so.0) #9 0x00007b752d54cab3 glXGetClientString (libGLX.so.0) #10 0x00007b75280d7803 n/a (libqxcb-glx-integration.so) #11 0x00007b75280d8199 n/a (libqxcb-glx-integration.so) #12 0x00007b7529373310 _ZN4KWin21X11StandalonePlatform19createOpenGLBackendEv (KWinX11Platform.so) In either case, it points to a faulty driver. It's also a quite old bug report now. Either way, if the issue is still present, please report it to your graphics driver developers.