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: Possible store motion tweak


law@redhat.com writes:
> If the proprietary benchmark license is anything like proprietary testsuite
> licenses, then we really can't give lots of detail -- those licenses are
> insanely restrictive.

Yup.

>  >This patch is Ok for mainline if bootstrapping and regression testing
>  >on two more platforms is Ok.  Once committed, if the automated SPEC
>  >testers don't show any serious regressions, it can stay.  I think its
>  >unreasonable to ask all contributors to run SPEC, but if you have
>  >access to it, those numbers would provide confidence before the commit.
>  >Reasonable?

OK, I bootstrapped & regression tested on i686-pc-linux-gnu and
alpha-linux-gnu.  (powerpc was suggested, but I don't have access
to a suitable box.)

> How about testing + SPEC on i686-pc-linux-gnu and call it good?  Richard has
> already tested mips64-pc-linux-gnu and seen good benchmarks on that
> platform as well.

I tried running SPEC on a Pentium III box and got the results below.
Not sure how meaningful they are.  Although I tried to quieten the
machine as much as I could (by paring down the number of system
services, etc.), I still had to keep the network stuff running.

Do the results look OK?  Is the patch good to go?

(The variations look suspiciously like noise to me, but I'm not
really used to SPEC, so I thought I'd better ask...)

Richard


Optimisation options were plain old -O3.

                                     Estimated                     Estimated
                   Base      Base      Base      Peak      Peak      Peak
   Benchmarks    Ref Time  Run Time   Ratio    Ref Time  Run Time   Ratio
   ------------  --------  --------  --------  --------  --------  --------
   164.gzip          1400   443       316          1400   452       310    *
   164.gzip          1400   452       310          1400   443       316     
   164.gzip          1400   451       311    *     1400   452       309     
   175.vpr           1400   540       259          1400   542       259    *
   175.vpr           1400   542       258    *     1400   540       259     
   175.vpr           1400   542       258          1400   543       258     
   176.gcc           1100        --          X     1100        --          X
   181.mcf           1800   819       220          1800   814       221     
   181.mcf           1800   819       220    *     1800   817       220     
   181.mcf           1800   815       221          1800   815       221    *
   186.crafty        1000   298       335    *     1000   297       337     
   186.crafty        1000   298       336          1000   296       337     
   186.crafty        1000   299       335          1000   297       337    *
   197.parser        1800   702       256          1800   701       257     
   197.parser        1800   705       255          1800   702       257    *
   197.parser        1800   703       256    *     1800   704       256     
   252.eon           1300        --          X     1300        --          X
   253.perlbmk       1800        --          X     1800        --          X
   254.gap           1100   392       280          1100   395       279     
   254.gap           1100   392       281    *     1100   394       279    *
   254.gap           1100   392       281          1100   394       279     
   255.vortex        1900   594       320          1900   599       317     
   255.vortex        1900   597       318          1900   588       323     
   255.vortex        1900   596       319    *     1900   592       321    *
   256.bzip2         1500   620       242          1500   623       241    *
   256.bzip2         1500   616       243          1500   623       241     
   256.bzip2         1500   617       243    *     1500   622       241     
   300.twolf         3000  1115       269          3000  1108       271     
   300.twolf         3000  1097       274          3000  1103       272    *
   300.twolf         3000  1100       273    *     3000  1102       272     
   ========================================================================
   164.gzip          1400   451       311    *     1400   452       310    *
   175.vpr           1400   542       258    *     1400   542       259    *
   176.gcc                                   X                             X
   181.mcf           1800   819       220    *     1800   815       221    *
   186.crafty        1000   298       335    *     1000   297       337    *
   197.parser        1800   703       256    *     1800   702       257    *
   252.eon                                   X                             X
   253.perlbmk                               X                             X
   254.gap           1100   392       281    *     1100   394       279    *
   255.vortex        1900   596       319    *     1900   592       321    *
   256.bzip2         1500   617       243    *     1500   623       241    *
   300.twolf         3000  1100       273    *     3000  1103       272    *


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