[PATCH] fix for PR 1687, plus a small cleanup

[Richard Henderson]

On Wed, May 07, 2003 at 10:36:48PM +0200, Steven Bosscher wrote:
> I would appreciate if somebody could see if the new longjmp check is
> OK.  We used to check for BUILT_IN_LONGJMP; with this patch we use
> special_function_p.

It is not ok.  __builtin_longjmp is different from longjmp
in this case.


r~