new syscall stub support for ia64 libc

[Ulrich Drepper]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

David Mosberger wrote:

> OK, it looks to me like the unwinder is failing.  I suspect it's
> choking either on the .altrp directive or on the unwind directives for
> the signal trampoline.

But we already did unwind through signal handlers.  It must be something
new your patch adds.  So likely the code in _dl_sysinfo_break, the
.altrp etc.  If a signal is received the thread is usually in a syscall
so we have to unwind through the signal handler frame, the sigreturn
stuff, and then the frame around the break instruction.  It must be the
step from sigreturn to _dl_sysinfo_break, everything else is the same.

- -- 
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQE/uQ+G2ijCOnn/RHQRAo1NAJ0aF0EYnOiEfGCJId1GEzRd83bvDwCeMZFY
ZQR4Jc+29Se2JNc/wCu4Jlg=
=Gvc+
-----END PGP SIGNATURE-----