Bug 217786

I had a crash in the same situation but with a slightly different backtrace:

Application: KWin (kwin), signal: Segmentation fault
[KCrash Handler]
#5  0x00007f1aff6d01e5 in KWin::DesktopGridEffect::setActive (this=0x91a0f0, active=<value optimized out>) at /usr/src/debug/kdebase-workspace-4.3.80/kwin/effects/desktopgrid/desktopgrid.cpp:996
#6  0x00007f1aff6d3903 in KWin::DesktopGridEffect::windowInputMouseEvent (this=0x91a0f0, e=<value optimized out>)
    at /usr/src/debug/kdebase-workspace-4.3.80/kwin/effects/desktopgrid/desktopgrid.cpp:575
#7  0x00007f1b0ff1a965 in KWin::EffectsHandlerImpl::checkInputWindowEvent (this=<value optimized out>, e=<value optimized out>) at /usr/src/debug/kdebase-workspace-4.3.80/kwin/effects.cpp:817
#8  0x00007f1b0fec46b2 in KWin::Workspace::workspaceEvent (this=0x838160, e=0x7fffdbd97400) at /usr/src/debug/kdebase-workspace-4.3.80/kwin/events.cpp:258
#9  0x00007f1b0fea3058 in KWin::Application::x11EventFilter (this=0x7fffdbd977f0, e=0x7fffdbd97400) at /usr/src/debug/kdebase-workspace-4.3.80/kwin/main.cpp:361
#10 0x00007f1b0c2ba511 in qt_x11EventFilter (ev=0x7fffdbd97400) at kernel/qapplication_x11.cpp:399
#11 0x00007f1b0c2c8df1 in QApplication::x11ProcessEvent (this=<value optimized out>, event=0x7fffdbd97400) at kernel/qapplication_x11.cpp:3231
#12 0x00007f1b0c2f3d62 in x11EventSourceDispatch (s=0x62e350, callback=<value optimized out>, user_data=<value optimized out>) at kernel/qguieventdispatcher_glib.cpp:146
#13 0x00007f1b056c7dde in g_main_context_dispatch () from /usr/lib64/libglib-2.0.so.0
#14 0x00007f1b056cb7a8 in ?? () from /usr/lib64/libglib-2.0.so.0
#15 0x00007f1b056cb8d0 in g_main_context_iteration () from /usr/lib64/libglib-2.0.so.0
#16 0x00007f1b0bb25b13 in QEventDispatcherGlib::processEvents (this=0x610ae0, flags=<value optimized out>) at kernel/qeventdispatcher_glib.cpp:407
#17 0x00007f1b0c2f394e in QGuiEventDispatcherGlib::processEvents (this=0x1242850, flags=<value optimized out>) at kernel/qguieventdispatcher_glib.cpp:202
#18 0x00007f1b0bafae52 in QEventLoop::processEvents (this=<value optimized out>, flags=) at kernel/qeventloop.cpp:149
#19 0x00007f1b0bafb22c in QEventLoop::exec (this=0x7fffdbd97730, flags=) at kernel/qeventloop.cpp:201
#20 0x00007f1b0bafef7b in QCoreApplication::exec () at kernel/qcoreapplication.cpp:981
#21 0x00007f1b0fea48be in kdemain (argc=<value optimized out>, argv=<value optimized out>) at /usr/src/debug/kdebase-workspace-4.3.80/kwin/main.cpp:526
#22 0x00007f1b0fb19a7d in __libc_start_main (main=<value optimized out>, argc=<value optimized out>, ubp_av=<value optimized out>, init=<value optimized out>, fini=<value optimized out>, 
    rtld_fini=<value optimized out>, stack_end=0x7fffdbd97cf8) at libc-start.c:220
#23 0x0000000000400699 in _start () at ../sysdeps/x86_64/elf/start.S:113
Warning: the current language does not match this frame.
The current source language is "auto; currently c".

Unfortunately I changed the code at exactly the lines which are mentioned in the backtrace :-(

Are you able to reproduce it reliable. If yes can you provide a pattern how to let desktop grid crash? I am unable to crash it, but sometimes desktop grid get's kind of stuck. I think it is related but up to now I have not yet found a way to reproduce.

Sadly it happens sporadic while using desktop grid.

I have started kwin in gdb, but I don't know which values are important to you I've selected a few:

(gdb) info locals
geometry = {xp = 4.9406564584124654e-324, yp = 6.953347831980716e-310, w = 6.9533558071188068e-310, h = 0}

(gdb) info args
this = 0xb20f70
w = 0x110a750

(gdb) print *w
$6 = {_vptr.EffectWindow = 0x1}

(gdb) print *this
$7 = {m_useGlobalAnimationModifier = true, m_managedWindows = {{d = 0x11049d0, e = 0x11049d0}}, m_movingWindows = 0}

I hope this can help. I can provide more debugging-information if needed.

As the relevant code changed, please try again with beta 2 and please provide a new backtrace if you are still able to reproduce.

(In reply to comment #4)
> As the relevant code changed, please try again with beta 2 and please provide a
> new backtrace if you are still able to reproduce.
Any news? If this crash still occurs in beta 2 please add a new backtrace, if not please add a comment that it is fixed. I would like to have such a crasher fixed before RC tagging (5th of January) but therefore I need to know if it is still valid.

Sry that I forgot to answer... This seems to be fixed with beta2 :)

(In reply to comment #6)
> Sry that I forgot to answer...
no problem :-)
> This seems to be fixed with beta2 :)
great thanks for the info