On Mon, May 18, 2015 at 04:22:25PM +0000, Michael van Elst wrote: > >As long as the key has been created - which it has not in the case at hand, > >or pt_havespecific would be true! > > How do you know that the key has not been created?
We established off-list that the key has been created but pthread_setspecific() has not been called. This means that pt_havespecific is false (I misread), and pthread_getspecific() should return NULL. So the assertion is wrong, and I'll have to debug further why the access to pt_specific[0] causes a crash (or wether something else goes wrong) in the netbsd32 compat case. Will dig further... Martin
