[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: athlon configuration impossibilty
From: |
Aaron Lehmann |
Subject: |
Re: athlon configuration impossibilty |
Date: |
Sun, 14 Oct 2001 10:22:58 -0700 |
User-agent: |
Mutt/1.3.20i |
On Sun, Oct 14, 2001 at 01:52:59PM +0200, Torbjorn Granlund wrote:
> Aaron Lehmann <address@hidden> writes:
>
> Running ./configure --target=athlon-pc-linux-gnu will result in the p6
> functions being used, because config.sub remaps athlon-* to i686-...
>
> Really? Using what GMP version?
Oh, I was looking at 3.1.1 from the Debian packages. It looks like
they patched config.{sub,guess} and replaced them with newer versions.
That would explain it.
BTW, preliminary analysis suggests that using gcc 3.0.2-pre makes the
library much slower than with gcc 2.95.4-pre. I'll see if I can get
more conclusive results. My tested consisted of running 'primes', and
it took a lot more CPU time when gmp was compiled with gcc 3.x.
However, I may have messed up some of the CFLAGS inadvertently, so I
will try to verify these results and present benchmarks.