Re: 486 Optimizations...

teg(at)redhat(dot)com (Trond Eivind Glomsrød) writes:

> Larry Rosenman <ler(at)lerctr(dot)org> writes:
>
> > * Trond Eivind Glomsr?d <teg(at)redhat(dot)com> [001114 15:43]:
> > > Larry Rosenman <ler(at)lerctr(dot)org> writes:
> > >
> > > > Anyone care if I build a patch to kill the -m486 type options in the
> > > > following files:
> > > >
> > > > $ grep -i -- 486 *
> > > > bsdi: i?86) CFLAGS="$CFLAGS -m486";;
> > > > freebsd:CFLAGS='-O2 -m486 -pipe'
> > > > univel:CFLAGS='-v -O -K i486,host,inline,loop_unroll -Dsvr4'
> > >
> > > Why would you want to? Not all gccs support -mpentium/mpentiumpro etc.
> > I would remove it entirely.
> >
> > Why should a package FORCE 486 or any other processor optimizations?
>
> It doesn't force, it only uses those if you don't specify any yourself
> AFAIK (at least, that's the normal way to do it)
>
> FWIW, this was what Red Hat Linux used up to and including 6.2 - it
> increases performance on almost every chip, it runs everywhere, it
> goes with any gcc.

To clarify - this is on IA32 only, of course. OSes know running on
more chips should get separate entries, like Linux does.
--
Trond Eivind Glomsrød
Red Hat, Inc.