Version: 3.1.90 (using KDE 3.1.9) Compiler: gcc version 3.2.2 OS: Linux (i686) release 2.6.0-test3 1. I started new konqueror. 2. Went to slashdot.org. 3. Opened a new tab. 4. Noticed that I still have some GUI effects turned on so I went to Control Center via that special button in Panel, turned off the GUI effects and pressed Ok which of course applies these changes I just made. 5. Konqueror crashed. As you can probably see I compiled everything else than kdelibs with --enable-debug=yes (I forgot kdelibs ;). Qt is the qt-copy with ./apply-patches less than week old. I also had another konqueror process running, so this might had had something to do with the empty tab? I'll try reproducing it after sending this report and comment it with the results. KDE (all modules) are checked out 08.08.2003 and built clean. Here is the KDE Crash handler backtrace: [New Thread 16384 (LWP 3865)] 0x41500b79 in __wait4 () at __wait4:-1 in __wait4 #0 0x41500b79 in __wait4 () at __wait4:-1 #1 0x41581414 in __DTOR_END__ () from /lib/libc.so.6 #2 0x41309cd3 in waitpid (pid=-512, stat_loc=0xfffffe00, options=-512) at wrapsyscall.c:181 #3 0x407662c0 in KCrash::defaultCrashHandler(int) (sig=11) at kcrash.cpp:240 #4 <signal handler called> #5 0x40731942 in KIconLoader::loadIcon(QString const&, KIcon::Group, int, int, QString*, bool) const (this=0x839f818, _name=@0xfffffe00, group=Desktop, size=1, state=70, path_store=0xbfffec88, canReturnNull=false) at kiconloader.cpp:630 #6 0x4073623d in KIconFactory::createPixmap(QIconSet const&, QIconSet::Size, QIconSet::Mode, QIconSet::State) (this=0x839b348, mode_P=4294966784) at kiconloader.cpp:1149 #7 0x40afdb26 in QIconSetPrivate::icon(QIconSet const*, QIconSet::Size, QIconSet::Mode, QIconSet::State) () from /usr/src/qt-copy/lib/libqt-mt.so.3 #8 0x40afe4a7 in QIconSet::pixmap(QIconSet::Size, QIconSet::Mode, QIconSet::State) const () from /usr/src/qt-copy/lib/libqt-mt.so.3 #9 0x40c4d6e0 in QPopupMenu::updateSize() () from /usr/src/qt-copy/lib/libqt-mt.so.3 #10 0x40c52a4b in QPopupMenu::styleChange(QStyle&) () from /usr/src/qt-copy/lib/libqt-mt.so.3 #11 0x40ad1561 in QApplication::setStyle(QStyle*) () from /usr/src/qt-copy/lib/libqt-mt.so.3 #12 0x406c751f in KApplication::applyGUIStyle() (this=0xbffff700) at kapplication.cpp:1648 #13 0x406c88f8 in KApplication::kdisplaySetStyle() (this=0x8116a18) at kapplication.cpp:1822 #14 0x406c702f in KApplication::x11EventFilter(_XEvent*) (this=0xbffff700, _event=0xbffff030) at kapplication.cpp:1524 #15 0x40a56f76 in qt_x11EventFilter(_XEvent*) () from /usr/src/qt-copy/lib/libqt-mt.so.3 #16 0x40a61d1c in QApplication::x11ProcessEvent(_XEvent*) () from /usr/src/qt-copy/lib/libqt-mt.so.3 #17 0x40a79f9c in QEventLoop::processEvents(unsigned) () from /usr/src/qt-copy/lib/libqt-mt.so.3 #18 0x40ae9cc6 in QEventLoop::enterLoop() () from /usr/src/qt-copy/lib/libqt-mt.so.3 #19 0x40ae9b68 in QEventLoop::exec() () from /usr/src/qt-copy/lib/libqt-mt.so.3 #20 0x40ad4611 in QApplication::exec() () from /usr/src/qt-copy/lib/libqt-mt.so.3 #21 0x416be75e in kdemain () from /opt/kde-head-1/lib/libkdeinit_konqueror.so #22 0x40854906 in kdeinitmain () from /opt/kde-head-1/lib/kde3/konqueror.so #23 0x0804d022 in launch (argc=2, _name=0x805f2b4 "konqueror", args=0x805f2c7 "\001", cwd=0x0, envc=1, envs=0xbffff860 "
I wasn't able to reproduce this.
Interesting. Thanks for the report.. There were some similar crashes I saw before, but they are not reproducible, and don't make any sense... I'll try to read the code a bit more during the evening, since you have debug info, and so the trace offers a few more clues, but no guarantees I can figure anything out -- this is something that probably should be Valgrinded, but if it's not reproducible...
*** Bug 62872 has been marked as a duplicate of this bug. ***
*** Bug 63380 has been marked as a duplicate of this bug. ***
This is a rather major bug. It seems to be reproduced by many people and shows lots of memory corruption. I also see kicker crash like this.
*** Bug has been marked as fixed ***.
*** Bug 71250 has been marked as a duplicate of this bug. ***
Hi, I see that this bug has resolution RESOLVED. Shouldn't this be changed to OPEN or something then ?