https://bugs.kde.org/show_bug.cgi?id=506782

--- Comment #5 from [email protected] ---
Okay, bear with me here - I forgot what the original issue was until I just
re-read it after experiencing a hang after unlock, which I got into my head as
the issue. However, I think it may have some insight as to the source of this
issue.

I disabled my auto-hiding panel, and today unlocked to find plasmashell hanging
again. I already had GDB attached in anticipation of this issue, which I
promptly used to break into plasmashell and see its stacktrace. I found that
the global app menu I have on one of my panels appears to be taking a long time
in the stack trace shown below:

#0  0x00007f4f04b3860d in syscall () at /usr/lib/libc.so.6
#1  0x00007f4f0555a9b1 in ??? () at /usr/lib/libQt6Core.so.6
#2  0x00007f4f05390f93 in QMetaObject::invokeMethodImpl(QObject*,
QtPrivate::QSlotObjectBase*, Qt::ConnectionType, long long, void const* const*,
char const* const*, QtPrivate::QMetaTypeInterface const* const*) () at
/usr/lib/libQt6Core.so.6
#3  0x00007f4f063f5c34 in ??? () at /usr/lib/libQt6DBus.so.6
#4  0x00007f4f063f6817 in ??? () at /usr/lib/libQt6DBus.so.6
#5  0x00007f4f053e78bc in QObjectPrivate::connectImpl(QObject const*, int,
QObject const*, void**, QtPrivate::QSlotObjectBase*, int, int const*,
QMetaObject const*) () at /usr/lib/libQt6Core.so.6
#6  0x00007f4f053e7c44 in QObject::connectImpl(QObject const*, void**, QObject
const*, void**, QtPrivate::QSlotObjectBase*, Qt::ConnectionType, int const*,
QMetaObject const*) () at /usr/lib/libQt6Core.so.6
#7  0x00007f4e640cc1d1 in ??? () at
/usr/lib/qt6/qml/org/kde/plasma/private/appmenu/libappmenuplugin.so
#8  0x00007f4f053e99d4 in ??? () at /usr/lib/libQt6Core.so.6
#9  0x00007f4ede05798b in ??? () at /usr/lib/libtaskmanager.so.6
#10 0x00007f4f053e99d4 in ??? () at /usr/lib/libQt6Core.so.6
#11 0x00007f4f056463b2 in
QIdentityProxyModelPrivate::sourceDataChanged(QModelIndex const&, QModelIndex
const&, QList<int> const&) () at /usr/lib/libQt6Core.so.6
#12 0x00007f4f053e99d4 in ??? () at /usr/lib/libQt6Core.so.6
#13 0x00007f4f0560fb84 in QAbstractItemModel::dataChanged(QModelIndex const&,
QModelIndex const&, QList<int> const&) () at /usr/lib/libQt6Core.so.6
#14 0x00007f4ede075fe7 in ??? () at /usr/lib/libtaskmanager.so.6
#15 0x00007f4f053e99d4 in ??? () at /usr/lib/libQt6Core.so.6
#16 0x00007f4ede070951 in ??? () at /usr/lib/libtaskmanager.so.6
#17 0x00007f4f063a62ee in ??? () at /usr/lib/libffi.so.8
#18 0x00007f4f063a29f8 in ??? () at /usr/lib/libffi.so.8
#19 0x00007f4f063a56ce in ffi_call () at /usr/lib/libffi.so.8
#20 0x00007f4f0807547a in ??? () at /usr/lib/libwayland-client.so.0
#21 0x00007f4f080762e9 in ??? () at /usr/lib/libwayland-client.so.0
#22 0x00007f4f080766bb in wl_display_dispatch_queue_pending () at
/usr/lib/libwayland-client.so.0
#23 0x00007f4f064d6f36 in ??? () at /usr/lib/libQt6WaylandClient.so.6
#24 0x00007f4f053ea001 in ??? () at /usr/lib/libQt6Core.so.6
#25 0x00007f4f05683109 in
QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) ()
    at /usr/lib/libQt6Core.so.6
#26 0x00007f4f05387d46 in
QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) ()
    at /usr/lib/libQt6Core.so.6
#27 0x00007f4f0537db7c in QCoreApplication::exec() () at
/usr/lib/libQt6Core.so.6
#28 0x000055f8fdb5ff78 in ??? ()
#29 0x00007f4f04a27b8b in __libc_start_call_main
    (main=main@entry=0x55f8fdb5cf30, argc=argc@entry=2,
argv=argv@entry=0x7ffec59a0258)
    at ../sysdeps/nptl/libc_start_call_main.h:58
#30 0x00007f4f04a27c4b in __libc_start_main_impl
    (main=0x55f8fdb5cf30, argc=2, argv=0x7ffec59a0258, init=<optimized out>,
fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7ffec59a0248) at
../csu/libc-start.c:360
#31 0x000055f8fdb60495 in ??? ()

This leads me to believe that the KWin hang may be tangentially related to the
app menu constantly updating, and those updates getting stuck in a queue while
KWin does... something. Now that I remember that I don't have to wait 24 hours
for the plasmashell hang, I'll follow the reproduction steps I gave myself
(:P), and see what's going on in a separate user. I'll also upgrade to Plasma
6.5 to see if the issue got fixed in the process.

By the way, the monitor turning back on issue has gone away since Plasma 6.4.4
or 6.4.5.

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to