Re: Reducing number of i386 linux images
On Wed, Aug 22, 2007 at 11:10:39PM -0600, dann frazier wrote:
> I hope it will be based on benchmarks
Yes. It is based on the lack of benchmarks. I did not find anything that
showed that the k7-optimization provides any not only theoretical
difference.
> I see no reason to drop any
> flavours if there maybe a significant performance benefit for a
> significant part of our userbase[1].
Until someone shows significant performance benefits, I don't see why we
should waste cpu and space to ship it. I know that gcc have different
cost tables for generic i686 and k7. The same is true for k8 and em64t,
but upstream said that distributors should not use it as it does not
provide any benefit. Also noone else ships explicit k7.
> fwiw, a coworker is looking into the availability of a benchmark that
> should be good for comparing pae/non-pae.
Someone said, mysql may be a usefull target. Can use much memory.
Bastian
--
Vulcans never bluff.
-- Spock, "The Doomsday Machine", stardate 4202.1
Reply to: