Re: [committed] Skip gcc.dg/Warray-bounds.c and g++.dg/warn/Warray-bounds.C

[Janis Johnson <janis187 at us dot ibm dot com>]

On Mon, Apr 16, 2007 at 10:55:37AM -0700, Mark Mitchell wrote:
> Janis Johnson wrote:
> 
> > Yes, it was a conscious decision so that we could notice a new ICE for
> > a test that was already an expected failure for some other reason.
> > However, H-P once pointed out to me that an ICE like this can still be
> > XFAILed with:
> > 
> >   /* { dg-do compile } */
> >   /* { dg-xfail-if "PR 31500" { hppa*-*-* && { ! hppa*64*-*-* } } } */
> 
> OK, that's a good hack.  If that works, John, I think that change can be
> safely considered pre-approved -- unless Janis objects, of course.

No objections from me.  If it doesn't work I can look into general
XFAIL of ICEs.

Janis