2010/4/3 John M McIntosh <johnmci@smalltalkconsulting.com>
On 2010-04-03, at 9:27 AM, laurent laffont wrote:
Hi,
Thanks to people who send the results of tinyBenchmarks for various CPU, here's the report.
What I want to know is whether these results are crap or "normal". It seems that all people succeed in compiling the VM with the source provided. So it may be a candidate for Pharo 1.0 ?
The procedure to build the VM was:
wget http://lolgzs.free.fr/pharo/squeak-vm/Squeak-3.11.3.2135-pharo-src.tar.gz tar -xvzf Squeak-3.11.3.2135-pharo-src.tar.gz cd Squeak-3.11.3.2135-pharo-src/ mkdir build && cd build ../unix/cmake/configure --CFLAGS="-O2" make
The results:
Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz 64 bits =============================================== 432 M bytecodes/sec; 13 M sends/sec
Optimal on a Intel(r) Core(TM)2 duo CPU @ 2.33Ghz is 587 M bytecodes/sec; 14.7 M sends/sec
Is "Optimal" means special CFLAGS, GCC version / Intel compiler, or source code changes ? I understand you cannot tell how you optimize the VM. But with the VM source I provide, maybe you can tell that compiling it with your "secret weapons" you obtain near optimal results ? Laurent Laffont
-- =========================================================================== John M. McIntosh <johnmci@smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project