This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Various GCC compared (2.95.3 - 3.5-20040523)
- From: Rene Rebe <rene at rocklinux-consulting dot de>
- To: gcc at gcc dot gnu dot org
- Date: Fri, 28 May 2004 17:05:49 +0200 (CEST)
- Subject: Various GCC compared (2.95.3 - 3.5-20040523)
Hi all,
I started generating some benchmark numbers.
My first and "preleminary" results are a bit unexpected and thus here
are two links for public comments:
Various build %-buildtimes with gcc-2.95.2 as base:
http://rocklinux-consulting.de/stat-bt2.png
Various runtimes with gcc-2.95.3 as base:
http://rocklinux-consulting.de/stat-rt2.png
Some notes:
* gcc-3.0.4 was the faster gcc looking at the build-time numbers
* add recent gcc's seem to have problems with gzip (?)
Some comments:
* all run on a PowerPC G3 @800Mhz under GNU/Linux
I'll run the same set of benchmarks on a Athlon-XP today
* yes, I made sure I pass --disable-checking to all gcc builds
* all gcc builds are bootstrap builds, that is no profiledbootstrap
(if time permits I will add seperate results with profiledbootsrap)
* the runtime test are:
gzip/bzip2:
- compressing and decompressing a 64MB tar file and 12MB executeable
with -1 and -9
openssl:
- make test
gnupg:
- make check
libmad:
- decompressing a 4.5MB mp3
Tests I will add now:
linux26: (2.6.6) build time
ogg/vorbis: compression decompression
I still search for a fitting C++ testsuit I can incooperate.
Any comments welcome!
Sincerely yours,
René Rebe
--
René Rebe - Europe/Germany/Berlin
rene@rocklinux.org rene@rocklinux-consulting.de
http://www.rocklinux.org http://www.rocklinux-consulting.de