Version:           kdenetwork-3.5.7 (using KDE KDE 3.5.7)
Installed from:    Gentoo Packages
Compiler:          gcc (GCC) 4.1.1 (Gentoo 4.1.1) 
OS:                Linux

Hi,  A few months ago I installed KDE 3.5.6 (full suite) on a Lemote Fulong miniPC which runs a MIPS-derivative CPU.  Kopete on this system was quite stable, until I upgraded the whole desktop to KDE 3.5.7.

Since then, Kopete starts up, attempts a connection to the Jabber server, then dies with the following crash dump.  I've tried strace on the kopete binary without success (it forks into the background).

Using host libthread_db library "/lib/libthread_db.so.1".
[Thread debugging using libthread_db enabled]
[New Thread 716137280 (LWP 19948)]
[KCrash handler]
#6  0x2d7ebcd8 in raise () from /lib/libc.so.6
#7  0x2d7ed7e8 in abort () from /lib/libc.so.6
#8  0x2d7e365c in __assert_fail () from /lib/libc.so.6
#9  0x2d44b288 in __pthread_tpp_change_priority () from /lib/libpthread.so.0
#10 0x2d4404b4 in pthread_mutex_lock () from /lib/libpthread.so.0
#11 0x2e1a196c in cricket::PhysicalSocketServer::Add ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#12 0x2e1a2374 in cricket::PhysicalSocketServer::PhysicalSocketServer ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#13 0x2e1a9400 in cricket::MessageQueue::MessageQueue ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#14 0x2e1a6294 in cricket::Thread::Thread ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#15 0x2e1a6d28 in cricket::ThreadManager::ThreadManager ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#16 0x2e1a6d98 in __static_initialization_and_destruction_0 ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#17 0x2e1b8818 in __do_global_ctors_aux ()
   from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#18 0x2dfbc060 in _init () from /usr/kde/3.5/lib/kde3/kopete_jabber.so
#19 0x2aab7208 in call_init () from /lib/ld.so.1
#20 0x2aab737c in _dl_init_internal () from /lib/ld.so.1
#21 0x2aabc0d4 in dl_open_worker () from /lib/ld.so.1
#22 0x2aab6d1c in _dl_catch_error () from /lib/ld.so.1
#23 0x2aabb7c0 in _dl_open () from /lib/ld.so.1
#24 0x2d520ee0 in dlopen_doit () from /lib/libdl.so.2
#25 0x2aab6d1c in _dl_catch_error () from /lib/ld.so.1
#26 0x2d5214c8 in _dlerror_run () from /lib/libdl.so.2
#27 0x2d520e00 in dlopen@@GLIBC_2.2 () from /lib/libdl.so.2
#28 0x2c1965d8 in sys_dl_open () from /usr/kde/3.5/lib/libkdecore.so.4
#29 0x2c194c38 in tryall_dlopen () from /usr/kde/3.5/lib/libkdecore.so.4
#30 0x2c195e34 in lt_dlopen () from /usr/kde/3.5/lib/libkdecore.so.4
#31 0x2c0b1d2c in KLibLoader::library () from /usr/kde/3.5/lib/libkdecore.so.4
#32 0x2abffc00 in Kopete::PluginManager::loadPluginInternal ()
   from /usr/kde/3.5/lib/libkopete.so.1
#33 0x2ac002cc in Kopete::PluginManager::slotLoadNextPlugin ()
   from /usr/kde/3.5/lib/libkopete.so.1
#34 0x2ac015b0 in Kopete::PluginManager::qt_invoke ()
   from /usr/kde/3.5/lib/libkopete.so.1
#35 0x2c6dbd88 in QObject::activate_signal () from /usr/qt/3/lib/libqt-mt.so.3
#36 0x2cb29854 in QSignal::signal () from /usr/qt/3/lib/libqt-mt.so.3
#37 0x2c7017dc in QSignal::activate () from /usr/qt/3/lib/libqt-mt.so.3
#38 0x2c70b67c in QSingleShotTimer::event () from /usr/qt/3/lib/libqt-mt.so.3
#39 0x2c658390 in QApplication::internalNotify ()
   from /usr/qt/3/lib/libqt-mt.so.3
#40 0x2c659814 in QApplication::notify () from /usr/qt/3/lib/libqt-mt.so.3
#41 0x2bfdb4e0 in KApplication::notify ()
   from /usr/kde/3.5/lib/libkdecore.so.4
#42 0x2c649130 in QEventLoop::activateTimers ()
   from /usr/qt/3/lib/libqt-mt.so.3
#43 0x2c5ed1f0 in QEventLoop::processEvents ()
   from /usr/qt/3/lib/libqt-mt.so.3
#44 0x2c677408 in QEventLoop::enterLoop () from /usr/qt/3/lib/libqt-mt.so.3
#45 0x2c677150 in QEventLoop::exec () from /usr/qt/3/lib/libqt-mt.so.3
#46 0x004350b4 in main ()

I should try the same package on big-endian MIPS and see if it's an endianness thing, but anyway.  Suffice to say it's known to be broken on little-endian.  I've tried recompiling kdenetwork without success, the build succeeds but Kopete is still broken.  Other KDE apps aren't affected.  The behaviour has been 100% reproducable.  Does anyone have some pointers where I should start looking?

Regards,
Stuart Longland