This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PING][PATCH/RFT] Fix PR middle-end/PR28690, modify swap_commutative_operands_p
- From: "H. J. Lu" <hjl at lucon dot org>
- To: Peter Bergner <bergner at vnet dot ibm dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, Paolo Bonzini <paolo dot bonzini at lu dot unisi dot ch>
- Date: Fri, 8 Jun 2007 15:26:17 -0700
- Subject: Re: [PING][PATCH/RFT] Fix PR middle-end/PR28690, modify swap_commutative_operands_p
- References: <1177614080.20709.34.camel@otta> <20070428153118.GA515@lucon.org> <1177795225.20709.42.camel@otta> <20070429235551.GA12511@lucon.org> <1177935400.20709.44.camel@otta> <20070506202116.GA12898@lucon.org> <20070604194005.GA19014@vnet.ibm.com>
On Mon, Jun 04, 2007 at 02:40:05PM -0500, Peter Bergner wrote:
> On Sun, May 06, 2007 at 01:21:16PM -0700, H. J. Lu wrote:
> > Here are what I got on Core 2 Duo with -O2 -ffast-math:
> >
> > 2. SPEC CPU 2000 cores
> [snip]
> > 168.wupwise -6.56276%
>
> Well, we found a Core 2 Duo box here and we're not seeing as big
> a degradation as you are (-3.9%), but we seem to have narrowed
> down one slowdown and created a C testcase that seems to mimic the
> slowdown. Can you confirm that the testcase below runs slower
> for you too. Also, we don't know the x86/x86_64 arch that well,
> so can you tell us why the code is running slower so I can attempt
> to modify the patch so it doesn't? Thanks.
>
With revision 125027, I got the following on Core 2 Duo 64bit:
(r125027 + patch - r125027)/125027
400.perlbench 0%
401.bzip2 0.613497%
403.gcc 0%
429.mcf 0%
445.gobmk 0.581395%
456.hmmer 0%
458.sjeng 0.571429%
462.libquantum -1.0582%
464.h264ref 0%
471.omnetpp -0.869565%
473.astar 0.869565%
483.xalancbmk 0.211416%
SPECint(R)_base2006 0.666667%
410.bwaves 0%
416.gamess -1.70455%
433.milc 0%
434.zeusmp -0.653595%
435.gromacs 0.866739%
436.cactusADM -2.83019%
437.leslie3d -0.757576%
444.namd -0.675676%
447.dealII -4.28016%
450.soplex 0%
453.povray 0%
454.calculix -5.28402%
459.GemsFDTD 0.961538%
465.tonto 0%
470.lbm 0.70922%
481.wrf 0.101937%
482.sphinx3 0%
SPECfp(R)_base2006 -0.757576%
164.gzip 0%
175.vpr -0.21645%
176.gcc -2.14153%
181.mcf -0.43692%
186.crafty 0.141343%
197.parser 0.285919%
252.eon -0.481303%
253.perlbmk 1.71658%
254.gap 0.0880669%
255.vortex -1.04854%
256.bzip2 -1.51237%
300.twolf -0.991853%
SPECint_base2000 -0.366133%
168.wupwise 0.24948%
171.swim -0.796312%
172.mgrid 0.796813%
173.applu 0.71582%
177.mesa 0.365497%
178.galgel -1.30807%
179.art -5.57125%
183.equake -0.513479%
187.facerec -0.0365364%
188.ammp -2.66921%
189.lucas 0.107991%
191.fma3d -2.39833%
200.sixtrack -2.21675%
301.apsi -0.929752%
SPECfp_base2000 -1.01523%
I think we can ignoe 179.art since it changes from run to run. Does
anyone have numbers on Opteron?
H.J.