> > Thank you for finding the problem. Would you mind citing that email you > found that claims that we're doing something wrong? I don't understand > it, but maybe someone else reading it would, and it'd be nice to get > feedback. The email thread in question is: https://sourceware.org/ml/pthreads-win32/2003/msg00001.html
> > Why does test_atomic_main() now need to call fatal_signal_exit()? After > I understand that, I'll carefully review the rest of the patch. test_atomic_main needs to call fatal_signal_init() because it can only be called when you have a single thread running. Later when you have multiple threads, you end up getting asserted via assert_single_threaded(). > > Thanks, > > Ben. _______________________________________________ dev mailing list [email protected] http://openvswitch.org/mailman/listinfo/dev
