This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Re: Rewrite i386 string operation expansion


On 11/30/06, Jan Hubicka <jh@suse.cz> wrote:

thanks for the results!  I've filled in the tables (and noticed that
nocona doesn't have 32bit codegen filled either), so I would be curious
if you can benchmark any speedups now.

Yes, with attached patch, povray now benchmarks a ~0.6% faster. The average of three benchmark runs (povray -display=NONE abyss.pov) are:

povray-29.nov: 14.033s
povray-current-inline: 13.941s
povray-current: 13.949s

(povray-current-inline represents povray, compiled with -minline-all-stringops)

As expected, -minline-all-strngops has negligible effect on the benchmark.


2006-11-30  Jan Hubicka  <jh@suse.cz>
            Uros Bizjak  <uros@kss-loka.si>

BTW: This is my old e-mail address. I'd rather use ubizjak at gmail dot com.


Thanks for your efforts,
Uros.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]