This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [BENCHMARK] Povray official benchmark scores for 387 and SSE insn sets
- From: Jakub Jelinek <jakub at redhat dot com>
- To: Richard Guenther <richard dot guenther at gmail dot com>
- Cc: Richard Henderson <rth at redhat dot com>, Uros Bizjak <uros at kss-loka dot si>, gcc at gcc dot gnu dot org
- Date: Tue, 25 Jan 2005 10:22:12 +0100
- Subject: Re: [BENCHMARK] Povray official benchmark scores for 387 and SSE insn sets
- References: <41F5260D.4070606@kss-loka.si> <20050124230356.GA29545@redhat.com> <84fc9c0005012500553c50e546@mail.gmail.com>
- Reply-to: Jakub Jelinek <jakub at redhat dot com>
On Tue, Jan 25, 2005 at 09:55:51AM +0100, Richard Guenther wrote:
> On Mon, 24 Jan 2005 15:03:56 -0800, Richard Henderson <rth@redhat.com> wrote:
> > On Mon, Jan 24, 2005 at 05:45:01PM +0100, Uros Bizjak wrote:
> > > As a last thought - is there a reason not to create an ABI similar to
> > > something like x86_64 has for its FP math?
> >
> > You'd have to get libc by-in. Which is probably impossible.
>
> At least we could change calling conventions for local functions, like
> I think is done
> with cgraph and register passing conventions atm.
Definitely, that's IMHO the way to go.
I already suggested this on gcc-patches in November, just did not get to
code that up yet (though, it wouldn't be appropriate for 4.0 anyway).
Jakub