This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH][RFC] Move IVOPTs closer to RTL expansion
- From: Richard Biener <rguenther at suse dot de>
- To: Pat Haugen <pthaugen at linux dot vnet dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Thu, 12 Sep 2013 09:56:53 +0200 (CEST)
- Subject: Re: [PATCH][RFC] Move IVOPTs closer to RTL expansion
- Authentication-results: sourceware.org; auth=none
- References: <alpine dot LNX dot 2 dot 00 dot 1309041115000 dot 20077 at zhemvz dot fhfr dot qr> <5230E33B dot 1080001 at linux dot vnet dot ibm dot com>
On Wed, 11 Sep 2013, Pat Haugen wrote:
> On 09/04/2013 04:20 AM, Richard Biener wrote:
> > Any help with benchmarking this on targets other than x86_64
> > is appreciated (I'll re-do x86_64).
> I ran CPU2000 and CPU2006 on PowerPC comparing the patch. CPU2000 had 3
> benchmarks degrade in the 4%-6%range (254.gap, 168.wupwise, 173.applu).CPU2006
> showed one benchmark, 410.bwaves, degrading 9%.I have dug into applu/bwaves
> and saw nothing directly attributable to your patch, degradations looked to be
> caused by idioms/oddities of the PowerPC architecture related to minor
> scheduling differences and/or data location.
Ok, I won't pursue the patch further at the moment unless a compelling
reason comes around. The idea I had back in time I proposed this
first was that we'd like to expose the effects of PROMOTE_MODE
to IVOPTs which would be easiest if the GIMPLE IL would be lowered
(we may be better optimizing required truncations for example
by VRP). Of course I didn't get around to playing with this
(mainly because x86 doesn't care ...).
Thanks for doing the measurement,