On Sun, 2006-07-30 at 18:48 +0200, Gilles Chanteperdrix wrote:
> Philippe Gerum wrote:
> > The second bug causing spurious RT32 signals to be notified is more of a
> > GDB issue
>
> When running the attached program inside gdb, the RT32 signal seems to
> be used as the asynchronous cancellation signal. Or at least when
> running with libthread_db.so.
>
Mm, ok. This would also correlate with GDB going south when ptracing the
asynchronous cancellation handler defined by the pthread library.
> plain text document attachment (cancel.c)
> #include <pthread.h>
>
> void *routine(void *cookie)
> {
> pthread_setcanceltype(PTHREAD_CANCEL_ASYNCHRONOUS, NULL);
>
> for (;;)
> ;
>
> return cookie;
> }
>
> int main(int argc, const char *argv[])
> {
> pthread_t tid;
> pthread_create(&tid, NULL, routine, NULL);
> pthread_cancel(tid);
> pthread_join(tid, NULL);
>
> return 0;
> }
--
Philippe.
_______________________________________________
Xenomai-core mailing list
Xenomai-core@gna.org
https://mail.gna.org/listinfo/xenomai-core
