Re: SUMMARY: pentium optimized debian discussion
On Tue, 28 Jul 1998, Jason Gunthorpe wrote:
>
> On Tue, 28 Jul 1998, Joey Hess wrote:
>
> > I'd suggest, if anyone can find a noticible speed increase on a pentium
> > clone, we bypass the whole problem and call it 586. It may be worthwhile to
> > call it 586 anyhow so we don't have to worry about intel.
>
> I think the main speed gains are less from instruction choice and more
> from more extensive efforts to decouple and pair instructions to keep the
> dual pipelines full. ANY x86 system that uses more than one pipeline (ie
> Cyrix, AMD, PII) will benifit to some degree.
But only if the pipelines are sufficiently similar. But I guess they are.
> I also think that if we are serious about this then we should use pegcs
> and not just egcs where possible..
And make a debian package out of it? Should 'update-alternative' the
compilers from egcc, g++ and gcc? (and oldg++ ?).
Maarten
_____________________________________________________________________________
| TU Delft, The Netherlands, Faculty of Information Technology and Systems |
| Department of Electrical Engineering |
| Computer Architecture and Digital Technique section |
| M.Boekhold@et.tudelft.nl |
-----------------------------------------------------------------------------
--
To UNSUBSCRIBE, email to debian-devel-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: