On Sun, Jul 16, 2006 at 02:10:22PM +0200, dragoran wrote:
I did same benchmarks using flops.c (attached) the results are
attached
to it seems that
-march=k8 -mtune=k8 -m32 is the fastes but not by much ...
the last one (-m64) is weird (much slower!!) no -OX was used.
If you were running a recent kernel from rawhide, you may want
to re-run these tests, as there was an fpu optimisation in there
that may have been responsible for spurious sigfpe's that some
apps have been getting. If that patch was to blame, then it's
possible there may have been other side-effects.
Todays rawhide has it reverted.
Dave
--
http://www.codemonkey.org.uk