This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Fix PR37053: Move tweaks of commutative precedence to target hook
- From: Peter Bergner <bergner at vnet dot ibm dot com>
- To: Maxim Kuvyrkov <maxim at codesourcery dot com>
- Cc: gcc-patches <gcc-patches at gcc dot gnu dot org>, Paolo Bonzini <bonzini at gnu dot org>, Ian Lance Taylor <iant at google dot com>, Luis Machado <luisgpm at linux dot ibm dot com>
- Date: Fri, 26 Jun 2009 10:45:28 -0500
- Subject: Re: [PATCH] Fix PR37053: Move tweaks of commutative precedence to target hook
- References: <4A44D817.firstname.lastname@example.org>
On Fri, 2009-06-26 at 18:15 +0400, Maxim Kuvyrkov wrote:
> The attached patch fixes PR37053
> The problem was introduced by patch for PR28690
> (http://gcc.gnu.org/ml/gcc-patches/2006-12/msg00268.html, rtlanal.c
> hunk); this hunk causes ICE on m68k (and, possibly, other) architectures.
> My first impulse was to remove the pointer tweak; but, well, if PowerPC
> really benefits from it,
I will say we worked for a long time coming up with the patch we did so
that it didn't affect other arches. In particular, HJ ran a lot of
performance tests to make sure it didn't degrade performance on x86.
I will also note, that without this patch, SPEC2000 performance degrades
by about 30% across the entire benchmark suite and 500% on galgel on
POWER6, so yeah, we benefit from it. :)
> ..., let's make it a hook.
> The attached patch does just that. Bootstrapped on x86_64-linux-gnu and
> cross-build to powerpc-linux-gnu (with no new warnings in logs); ok to
I'll let others comment on whether they want a target hook solution or not.
If they don't, I'm willing to work with you to come up with something that
works for both of us.
In the mean time, I have asked Luis to test your patch on POWER6 to ensure
there are no performance degradations. It doesn't look like there should