This is the mail archive of the
mailing list for the GCC project.
Re: fix eh tests on x86-darwin
- From: Andrew Pinski <pinskia at physics dot uc dot edu>
- To: geoffk at apple dot com (Geoffrey Keating)
- Cc: gkeating at apple dot com (Geoffrey Keating), gcc-patches at gcc dot gnu dot org, aoliva at redhat dot com, iant at google dot com, pinskia at physics dot uc dot edu (Andrew Pinski)
- Date: Fri, 25 Aug 2006 18:59:15 -0400 (EDT)
- Subject: Re: fix eh tests on x86-darwin
> On 24/08/2006, at 9:21 PM, Andrew Pinski wrote:
> > On Thu, 2006-08-24 at 18:26 -0700, Geoffrey Keating wrote:
> >> I'm running one last bootstrap & testrun and will wait for that and
> >> for a day or so for people to comment before I commit.
> > Why check in this patch for 4.2? You know x86-darwin full support is
> > new for GCC 4.2 anyways so it cannot be a regression. There are other
> > regressions (and bugs) which are more important which can be fixed
> > instead of working on something that does not effect any primary or
> > secondary target.
> I've verified that the bug is indeed a regression by building the 4.1
> branch on x86-darwin and confirming that if you compile gcc.dg/
> brendan/eh1.C with it, the resulting executable works:
> PASS: g++.old-deja/g++.brendan/eh1.C execution test
> I consider that x86-darwin support is important, because within the
> lifespan of the 4.2 release I expect there will be more use of 4.2 on
> x86-darwin than on ppc-darwin.
Why do you say that, there are so many more PowerPC Mac OS X boxes there
than x86 boxes and plus x86-darwin is a new target. x86-darwin was broken
in GCCs before 4.2 really so it should not be considered a regression.
Also you don't make the primary/secondary targets, the SC decides those
and right now Powerpc-darwin is a secondary target while x86-darwin is not.