Version: 1.9.7 (using KDE KDE 3.5.7) Installed from: Gentoo Packages Compiler: GCC 4.2.0 i686-pc-linux-gnu-4.2.0 CFLAGS="-march=athlon-xp -pipe -O2 -ggdb" OS: Linux I try to delete letter and kmail crash. Not always. Using host libthread_db library "/lib/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1257670304 (LWP 6249)] [New Thread -1287251056 (LWP 6263)] [New Thread -1278858352 (LWP 6262)] [New Thread -1270465648 (LWP 6261)] [New Thread -1262072944 (LWP 6260)] [KCrash handler] #6 0xb5df8552 in QGList::findRef (this=0x80d7580, d=0x83481e0, fromStart=true) at tools/qglist.cpp:725 #7 0xb5df93fc in QGList::removeRef (this=0x80d7580, d=0x83481e0) at tools/qglist.cpp:553 #8 0xb5a2587e in QPtrList<QPostEvent>::removeRef (this=0x80d7580, d=0x83481e0) at ../include/qptrlist.h:86 #9 0xb5a1c460 in QApplication::sendPostedEvents (receiver=0x0, event_type=0) at kernel/qapplication.cpp:3247 #10 0xb5a1c690 in QApplication::sendPostedEvents () at kernel/qapplication.cpp:3181 #11 0xb59b7d91 in QEventLoop::processEvents (this=0x8085f48, flags=4) at kernel/qeventloop_x11.cpp:144 #12 0xb5a3a108 in QEventLoop::enterLoop (this=0x8085f48) at kernel/qeventloop.cpp:198 #13 0xb5a39e06 in QEventLoop::exec (this=0x8085f48) at kernel/qeventloop.cpp:145 #14 0xb5a1d105 in QApplication::exec (this=0xbf943dd0) at kernel/qapplication.cpp:2758 #15 0x0804abe7 in main (argc=0, argv=0x0) at main.cpp:110
How can the problem be reproduced? I don't get crashes when deleting messages. This backtrace looks useless, it seems that the crash is deep inside Qt, it should not crash like this. Maybe the crash is caused be invalid compiler optimizations? Please try to recompile Qt and KMail without any optimizations. Also try to get a better backtrace, see http://techbase.kde.org/Development/Tutorials/Debugging/How_to_create_useful_crash_reports
I also delete letters normal without any crash, but some times, when I try to delete - Kmail - crash. Thats all. I not always but some times. If it was be always, I'll be run it thought gdb...
Is this bug still present ? If so, could you install the debug packages and provide a backtrace ?
No feedback (comment #3) nor useful backtrace (comment #1). Closing.