At that time I ddin't have an Athlon so someone made some measures for me and it looked like there was with gcc 3.2 there was a non-negligble difference (these things evolve from version to version) between the -O2 -mcpu=i686 used in most RedHat packages and the -O2 -march=athlon you would use in a package built specifically for the athlon. Another point is RedHat doesn't use -mmmx for normal packages, probably due to the PentiumPro (the only processor in the family who didn't support MMX). I were Redhat I would tell to he with PPro (there were never many of them) and I would use MMX. However on early MMX-enabled CPUs shifting from FP to MMX modes is slow so you only get real benefits when sequence is long. For an athlon-optimized package you could use -m3dnow and, if it is an athlon XP, -msse. Use of -mmmx on the lmbench test (memory bandwidth) brought significant speedups on my old K6. JFM On Tue, 2003-05-27 at 06:10, Jesse Keating wrote: > On Monday 26 May 2003 18:18, Steve Snyder wrote: > > Also: shouldn't you have a "-O2" (or something) in amongst those compile > > switches? > > Thats automatic in the rpmbuilding. > > Another thing you might be seeing, is that it may take longer to compile > athlon binaries than i686 binaries. The funny thing is that you will gain > very very small amounts of speed increase, if any at all. You've got to ask > yourself, why are you recompiling all the packages? The ones that really do > benifit from athlon/i686 like opts. are already provided by Red Hat compiled > for this. > > See the many many discussions we've already had on this list about this issue. > > -- > Jesse Keating RHCE MCSE > http://geek.j2solutions.net > Mondo DevTeam (www.mondorescue.org) > > Was I helpful? Let others know: > http://svcs.affero.net/rm.php?r=jkeating > > > -- > Shrike-list mailing list > Shrike-list@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/shrike-list >
