Re: another life-cycle/timing bug

2015-08-13 Thread Stephan Bergmann
On 08/13/2015 11:06 AM, Tor Lillqvist wrote: Trying to "exit cleanly" is so last century. Apps should just _exit(). In a multithreaded app, in general there needs to be communication among the threads about process termination, no matter whether termination is done via exit() or _exit(). ___

Re: another life-cycle/timing bug

2015-08-13 Thread Tor Lillqvist
Trying to "exit cleanly" is so last century. Apps should just _exit(). (Yes, I know what you are going to say about valgrind and its cousins.) --tml ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/list

Re: another life-cycle/timing bug

2015-08-13 Thread Stephan Bergmann
On 08/12/2015 11:52 PM, Norbert Thiebaud wrote: Thread 4 Crashed: 0 libupdchklo.dylib 0x000122446611 rtl::OUString::equalsAscii(char const*) const + 33 1 libupdchklo.dylib 0x0001224401cf NamedValueByNameAccess::getValue(char const*) + 95 2 libupdchklo

another life-cycle/timing bug

2015-08-12 Thread Norbert Thiebaud
I ran into this one twice.. it is pretty hard to hit... but still. saddly I do not have a core for it... Process: cppunittester [13018] Path: /Volumes/VOLUME/*/cppunittester Identifier:cppunittester Version: 0 Code Type: X86-64