This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] insn priority adjustments in scheduler and rs6000 port
- From: David Edelsohn <dje at watson dot ibm dot com>
- To: "Vladimir N. Makarov" <vmakarov at redhat dot com>
- Cc: Dorit Naishlos <DORIT at il dot ibm dot com>, Ayal Zaks <ZAKS at il dot ibm dot com>, gcc-patches at gcc dot gnu dot org
- Date: Fri, 03 Oct 2003 19:41:42 -0400
- Subject: Re: [PATCH] insn priority adjustments in scheduler and rs6000 port
>>>>> Vladimir N Makarov writes:
Vlad> You mentioned SPEC but the results are absent in the message. As I
Vlad> understood, the results mentioned in
Vlad> http://gcc.gnu.org/ml/gcc-patches/2003-09/msg02105.html
Vlad> are for the 1st patch 'insn priority adjustments in scheduler and
Vlad> rs6000 port'. Is it true? Or the results were for the two patches?
The SPEC performance results reported in the message you reference
give results both for the first patch and for the first and second patch
(scheduling of queued instructions) combined.
First patch:
SPECfp improvement: 0.3% average
SPECint improvement: 0.2% average
Max improvement: 3.2%
Max degradation: 1.2%
First and Second patch (scheduling of queued instructions):
SPECfp improvement: 0.6% average
SPECint improvement: 0.6% average
Max improvement: 4.1%
Max degradation: 1.0%
David