| Summary: | Plasma shell freezes when cycling through task manager when window previews are being shown (hover or on grouped) | ||
|---|---|---|---|
| Product: | [Plasma] kwin | Reporter: | Marcelo Bossoni <mmbossoni> |
| Component: | wayland-generic | Assignee: | KWin default assignee <kwin-bugs-null> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | normal | CC: | desirev00, kde, nate, qydwhotmail |
| Priority: | NOR | Keywords: | regression |
| Version First Reported In: | 5.26.90 | ||
| Target Milestone: | --- | ||
| Platform: | Arch Linux | ||
| OS: | Linux | ||
| Latest Commit: | https://invent.kde.org/plasma/plasma-workspace/commit/b0bc2b9d52614326ad1a2be02e68114d208b759e | Version Fixed In: | 5.27 |
| Sentry Crash Report: | |||
|
Description
Marcelo Bossoni
2023-01-20 00:04:25 UTC
It seems to happen with other plasma popups as well, like notifications When this happens, can you open Konsole (perhaps using KRunner) and run the following? $ gdb attach $(pidof plasmashell) $ bt Paste the output here. [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/libthread_db.so.1". 0x00007f9d6809c4b6 in ?? () from /usr/lib/libc.so.6 (gdb) bt #0 0x00007f9d6809c4b6 in () at /usr/lib/libc.so.6 #1 0x00007f9d6809ecd0 in pthread_cond_wait () at /usr/lib/libc.so.6 #2 0x00007f9d5e8b8f08 in () at /usr/lib/libEGL_nvidia.so.0 #3 0x00007f9d5e8b9085 in () at /usr/lib/libEGL_nvidia.so.0 #4 0x00007f9d5e88ffca in () at /usr/lib/libEGL_nvidia.so.0 #5 0x00007f9d5e892a7d in () at /usr/lib/libEGL_nvidia.so.0 #6 0x00007f9d5e89bdee in () at /usr/lib/libEGL_nvidia.so.0 #7 0x00007f9d5e89c924 in () at /usr/lib/libEGL_nvidia.so.0 #8 0x00007f9d5e89ca2e in () at /usr/lib/libEGL_nvidia.so.0 #9 0x00007f9d5e847c23 in () at /usr/lib/libEGL_nvidia.so.0 #10 0x00007f9d609f8dba in () at /usr/lib/libnvidia-egl-wayland.so.1 #11 0x00007f9d5e8a878e in () at /usr/lib/libEGL_nvidia.so.0 #12 0x00007f9d5e847bc0 in () at /usr/lib/libEGL_nvidia.so.0 #13 0x00007f9d6390a6bd in () at /usr/lib/qt/plugins/wayland-graphics-integration-client/libqt-plugin-wayland-egl.so #14 0x00007f9d6a3d6b9c in () at /usr/lib/libQt5Quick.so.5 #15 0x00007f9d6a452739 in QQuickWindow::event(QEvent*) () at /usr/lib/libQt5Quick.so.5 #16 0x00007f9d4d459c9a in () at /usr/lib/qt/qml/org/kde/plasma/core/libcorebindingsplugin.so #17 0x00007f9d69578b5c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib/libQt5Widgets.so.5 #18 0x00007f9d6888de08 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /usr/lib/libQt5Core.so.5 #19 0x00007f9d68d333b0 in QPlatformWindow::deliverUpdateRequest() () at /usr/lib/libQt5Gui.so.5 #20 0x00007f9d688b1a70 in QObject::event(QEvent*) () at /usr/lib/libQt5Core.so.5 #21 0x00007f9d69578b5c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib/libQt5Widgets.so.5 #22 0x00007f9d6888de08 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /usr/lib/libQt5Core.so.5 #23 0x00007f9d6888e913 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () at /usr/lib/libQt5Core.so.5 #24 0x00007f9d688d4d48 in () at /usr/lib/libQt5Core.so.5 #25 0x00007f9d675f687b in g_main_context_dispatch () at /usr/lib/libglib-2.0.so.0 #26 0x00007f9d6764dc89 in () at /usr/lib/libglib-2.0.so.0 #27 0x00007f9d675f5132 in g_main_context_iteration () at /usr/lib/libglib-2.0.so.0 #28 0x00007f9d688d8b2c in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt5Core.so.5 #29 0x00007f9d688865ac in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt5Core.so.5 #30 0x00007f9d688910d9 in QCoreApplication::exec() () at /usr/lib/libQt5Core.so.5 #31 0x0000561d371ed158 in () #32 0x00007f9d6803c290 in () at /usr/lib/libc.so.6 #33 0x00007f9d6803c34a in __libc_start_main () at /usr/lib/libc.so.6 #34 0x0000561d371ed505 in () (gdb) Just as a test, I've downgraded to 5.26.5 to rule out library updates and it worked fine. So this really seems a breakage from 5.27. The backtrace is all in the NVIDIA driver, so if downgrading to 5.26.5 fixes it, I guess some change made for 5.27 interacts poorly with an existing quirk or bug in the NVIDIA driver. Should be fixed with the recent changes. Please test again in 5.27.0 Git commit 0878690518366410eede56d731abe6142f9186d9 by Fushan Wen. Committed on 25/01/2023 at 15:58. Pushed by fusionfuture into branch 'master'. libtaskmanager: don't recreate Screencasting This makes the preview transition between tasks less rigid. FIXED-IN: 5.27 M +15 -12 libtaskmanager/declarative/screencastingrequest.cpp https://invent.kde.org/plasma/plasma-workspace/commit/0878690518366410eede56d731abe6142f9186d9 Git commit b0bc2b9d52614326ad1a2be02e68114d208b759e by Fushan Wen. Committed on 25/01/2023 at 15:59. Pushed by fusionfuture into branch 'cherry-pick-08786905'. libtaskmanager: don't recreate Screencasting This makes the preview transition between tasks less rigid. FIXED-IN: 5.27 (cherry picked from commit 0878690518366410eede56d731abe6142f9186d9) M +15 -12 libtaskmanager/declarative/screencastingrequest.cpp https://invent.kde.org/plasma/plasma-workspace/commit/b0bc2b9d52614326ad1a2be02e68114d208b759e This exact bug is still happening to me on Fedora 37 KDE with the newest proprietary Nvidia drivers, on RTX 2060. Killall plasmashell and running it again fixes it. Turning off window previews also stops the freezing. See alse: https://bugs.kde.org/show_bug.cgi?id=466580 and https://bugs.kde.org/show_bug.cgi?id=466580 *** Bug 466681 has been marked as a duplicate of this bug. *** *** Bug 466681 has been marked as a duplicate of this bug. *** Still happening on 5.27.4 *** This bug has been marked as a duplicate of bug 450248 *** Post Mortem comment: This is NOT a duplicate of bug 450248 This seems to be a NVIDIA thing. I end up finding the root cause. I have __GL_SYNC_TO_VBLANK=0 set in my variables and that end up causing the issue. Removing the flag fixed the bug for me. |