This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Fix PR middle-end/14988
- From: Roger Sayle <roger at eyesopen dot com>
- To: Laurent GUERBY <laurent at guerby dot net>
- Cc: Eric Botcazou <ebotcazou at act-europe dot fr>, <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 30 Apr 2004 12:47:24 -0600 (MDT)
- Subject: Re: Fix PR middle-end/14988
On Fri, 30 Apr 2004, Laurent GUERBY wrote:
> Is there any reason not to commit this patch to mainline? I've
> been using it on x86-linux and powerpc-darwin without problems
> for a while, and this is silent wrong code generation.
We've just been waiting for someone to test it on more than a
single platform. It's not uncommon for risky patches to break
bootstrap on platforms that it wasn't tested on (occassionally
on all the platforms it wasn't tested on :). It's not unreasonable
to request that a patch containing a target #ifdef test is tested
on atleast two platforms that test compilation of the code modified
in either arm of the conditional compilation.
[Personally, I've been sitting on a patch that depends upon BIG_ENDIAN
waiting for mainline CVS to bootstrap again on PA, as that's the target
that I normally use to test other-endian from IA-32].
Anyway, it looks like you've done the necessary testing for Eric.
In which case, assuming you've run the testsuite x86-linux-gnu and
powerpc-apple-darwin without regressions, this patch is OK for
mainline.
Roger
--