new syscall stub support for ia64 libc

[David Mosberger]

>>>>> On Tue, 18 Nov 2003 11:29:27 -0800, Ulrich Drepper <drepper@redhat.com> said:

  Uli> David Mosberger wrote:

  >> Hans Boehm pointed out that pthread_setcancelstate() isn't
  >> asynch-signal-safe,

  Uli> Where does he get this from?

I assume from here (table of async-safe routines in the "Signal
Concepts" section):

 http://www.opengroup.org/onlinepubs/007904975/functions/xsh_chap02_09.html

  Uli> pthread_setcancelstate() is one of the three functions which
  Uli> are guaranteed to be async-cancel safe.

Yes, that's why I _assumed_ it would be async-signal-safe, but it
seems now that was an incorrect assumption.

	--david