> On Mon, Apr 02, 2001 at 05:17:57PM -0400, Roland McGrath wrote:
> > > Yes. Well, I now noticed that I could provoke it crashing badly when
> > > I wait for proc to get E_BAD_ACCESS, and then run
> > >
> > > print mach_thread_self()
> >
> > Well, that is not surprising. You are telling gdb to call a function in
> > the inferior process.
>
> Ok. I am just happy that it seems to work in this simple trap w/o args,
> because I didn't manage to access _hurd_thread_self in gdb and wouldn't know
> how to get the thread port number any other way.
I think making the system call might be the most reasonable thing.
_hurd_thread_self does nothing but access data on the stack, so that is
sure to lose in the case you've got. You can look through the whole
_hurd_sigstates list, which has the thread port (mach_thread_self() value)
of each thread in its _hurd_sigstate structure.
_______________________________________________
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd
