Richard Guenther wrote:
On 6/15/07, Vladimir N. Makarov <[EMAIL PROTECTED]> wrote:
Just to summarize what we test at SUSE currently:
Richard, thanks for the information.
- SPEC2000 is tested in various variants on AMD x86_64, including
32bit results and FDO runs.
- SPEC2000 is tested on ia64
I have only one available ia64 machine (Montecito) but it is used for
development.
- SPEC2006 is tested on AMD x86_64, but results are currently not
accessible to the outside (we need to go over the SPEC reporting
rules)
so, no PPC testing from us (the old testing machine died and we don't
have a replacement for it). So, if you happen to have a machine for
PPC testing then this would be nice. Maybe IBM can set up such as well.
I have a ppc machine which IBM kindly sent me. But now it is actively
used for my register allocation project. May be in a few months, I
could start to use it.
Diego run all tests only once therefore it was natural to use available
machines over nights. Some machines were preproduction ones. I don't
think it has a sence to run the tests once (because the result is
inaccurate in this case) and to use the preproduction machines (I think
we should be oriented to new processors). So it is hard to resume
testing all machines which Diego used.
The performance tracking results is on
https://vmakarov.108.redhat.com/nonav/spec/index.html
To beat Gerald here - can you update gcc.gnu.org/benchmarks.html?
Gerald is pretty good at updating html pages. But ok, I'll do it.
Vlad
