Re: What architecture are we?
Brian C White <bcwhite@verisim.com> writes:
> A better alternative, I think, would be to have the GCC maintainer tweak
> GCC just a bit so its default behavior conforms to the specifications
> laid down by our illustrious leader (hi, Bruce!). I would suggest:
>
> - optimize for i486 (maybe Pentium)
> - use no instructions not available on the i386.
Code generated with -m386 code runs faster on Pentium machines than
code generated with -m486. Hence, in the Linux kernel (in addition to
-O2 and such) these optimizations are used:
i386: -m386
i486: -m486
i586: -m486 -malign-loops=2 -malign-jumps=2 -malign-functions=2
The -malign options reduce the bloating of -m486 code and speed it up
on Pentiums. I suggest that the GCC maintainer:
- add a -m586 option that is equivalent to the above
- make -m586 default if -m386, -m486, and -m586 are not specified
and make this relatively easy to change for people running i486 or
i386 machines
This may be more appropriate for the upstream GCC maintainer(s).
Dan
--
Daniel Quinlan http://www.pathname.com/~quinlan
quinlan@pathname.com quinlan@transmeta.com (at work)
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-REQUEST@lists.debian.org . Trouble? e-mail to Bruce@Pixar.com
Reply to: