This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Atom: Enabling unroll at O2 optimization level
- From: Richard Guenther <richard dot guenther at gmail dot com>
- To: Igor Zamyatin <izamyatin at gmail dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Wed, 11 Apr 2012 10:39:38 +0200
- Subject: Re: [PATCH] Atom: Enabling unroll at O2 optimization level
- References: <CAKdSQZn_t_YS96yxC1gYON=RT1uzmP0FMqEvucbZNPAVx-_c5Q@mail.gmail.com>
On Tue, Apr 10, 2012 at 8:43 PM, Igor Zamyatin <email@example.com> wrote:
> Hi All!
> Here is a patch that enables unroll at O2 for Atom.
> This gives good performance boost on EEMBC 2.0 (~+8% in Geomean for 32
> bits) with quite moderate code size increase (~5% for EEMBC2.0, 32
5% is not moderate. Your patch does enable unrolling at -O2 but not -O3,
why? Why do you disable register renaming? check_imull requires a function
This completely looks like a hack for EEMBC2.0, so it's definitely not ok.
-O2 is not supposed to give best benchmark results.
> Tested for i386 and x86-64, ok for trunk?
> 2012-04-10 ?Yakovlev Vladimir ?<firstname.lastname@example.org>
> ? ? ? ?* gcc/config/i386/i386.c (check_imull): New routine.
> ? ? ? ?(ix86_loop_unroll_adjust): New target hook.
> ? ? ? ?(ix86_option_override_internal): Enable unrolling on Atom at -O2.
> ? ? ? ?(TARGET_LOOP_UNROLL_ADJUST): New define.