|
From: Bjoern K. <Bjo...@un...> - 2004-03-22 07:17:02
|
Hi! Am Freitag, 19. M=E4rz 2004 18:40 schrieb Melchior FRANZ: > * Bjoern Kahl -- Friday 19 March 2004 18:20: > > I am trying to valgrind (memcheck) a multithreaded QT-Application. > > > > The program works well (more or less :-) ) without valgrind, but fails > > horribly with valgrind, even with the "none" tool. > > So *that* is my problem! (See the thread "2.1.1. and later: crash ..."). > I got the crash with all Qt applications (libqt-mt is linked with > libpthread) and with FlightGear ... when I had the latter compiled with > pthread. After your message I recompiled FlightGear without threads, and > viola: valgrind works again! Definitely a thread bug in valgrind then. I am glad, I am nht the only one with this problem, as I started to think may have made some really stupid things with my code, somehow breaking the thread-synchronisation. What I do not understand: How could valgrind prevent QT from working with pthreads? At least it creates new threads, just mutexes dont work. And I can't see any "illegal" memory acess in QT. (I thought valgrind is used by KDE, so how did they check there KDE-apps, without working QThreads ?!?) If I can do something to help sort this out, please let me know what and how. Bjoern =2D- +-----------------------------------------------------------------+ | Dipl.-Phys. Bjoern Kahl +++ Lehrstuhl Angewandte Informatik III | | (Robotik und Eingebettete Systeme) +++ Universitaet Bayreuth | | phone: +49-921-55-5164 +++ www: http://ai3.inf.uni-bayreuth.de | +-----------------------------------------------------------------+ |