This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
2.96 (20000525) benchmarks
- To: gcc at gcc dot gnu dot org
- Subject: 2.96 (20000525) benchmarks
- From: German Jose Gomez Garcia <german at pinon dot ccu dot uniovi dot es>
- Date: Thu, 25 May 2000 19:42:57 +0200 (METDST)
Hello,
Although I followed the development of egcs/gcc up to 2.95.2,
I had forgot it for a long time. I decided that it was time to check it up
again, so I downloaded latest from CVS and compiled it (with make
bootstrap).
It compiles flawlessly (even with CFLAGS=-O2 -fomit-frame-pointer)
but I have to say that I was badly surprise by the results.
I compiled crafty both with 2.95.2 and with 2.96 (25-05-2000) and
run the benchmark command (I know this may be the worse benchmark ever
done by anybody but...)
The results were:
gcc-2.95.2
Total nodes: 82884181
Raw nodes per second: 343917
Total elapsed time: 241
SMP time-to-ply measurement: 2.655602
gcc-2.96 (20000525)
Total nodes: 76470799
Raw nodes per second: 315995
Total elapsed time: 242
SMP time-to-ply measurement: 2.644628
they correspond to crafty-17.10 compiled with
-march=i686 -O3 -fomit-frame-pointer
I run it on a Dual PII 450 running 2.2.16-pre4 with
glibc-2.1.3
binutils-2.9.5.0.41
the only change from one version to the other was the -V 2.95.2
and the -V 2.96 specified to the gcc driver.
I wonder if that is normal/expected or by the contrary I have made
some fatal error?
Thanks and keep the good work, everybody is waiting for the 3.0 :-)
- german
PS: Please CC to me as I'm not subscribed to this list.
<>-------------------------------------+-----------------------------------<>
One O.S. to rule them all, | German Gomez Garcia
One O.S. to find them. | german@pinon.ccu.uniovi.es
One O.S. to bring them all |
and in the darkness bind them. | "Wur Qanar Wur Stilor Wur Kas"
<>-------------------------------------+-----------------------------------<>