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: SSE vs. x87 povray deathmatch [was: Re: [RFC PATCH, x86_64] Use -mno-sse[,2] to fall back to x87 FP ...]


Uros, 

> I'll run SPEC CPU2006 Povray which I have handy on SUSE 10.0 
> and post the results later.  Then I'll run 3.6.1 as well.

Here's what I got with 453.povray from SPEC CPU2006 on an Athlon 64 @ 2200MHz under SUSE 10.0 (using the benchmark suite choice out of 3 runs):

-O3 -funroll-all-loops -ffast-math -fno-inline-functions...			404s
-O3 -funroll-all-loops -ffast-math -fno-inline-functions -msse3...		404s
-O3 -funroll-all-loops -ffast-math -fno-inline-functions -mfpmath=387...	403s

I also build the head version of GLIBC (without math routines optimized for x86-64) and LD_PRELOAD'ed libm, with this result:

-O3 -funroll-all-loops -ffast-math -fno-inline-functions...			406s

All I can say is that the differences are too close to call...

-- 
_______________________________________________________
Evandro Menezes               AMD            Austin, TX




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