GMP on Pentium 2
deltatrinity at hotmail.com
Tue Oct 21 20:41:24 CEST 2003
Humm, probably yes.
Maybe something have been optimized in Athlon asm files and not done in P2
asm files. Like if when compiling for P2, a generic Pentium 1 file is used
and when compiling for Athlin, it's using athlon-specific files. Well, I
can only guess... :)
>From: Carl Devore <devore at math.udel.edu>
>To: delta trinity <deltatrinity at hotmail.com>
>CC: <gmp-discuss at swox.com>
>Subject: Re: GMP on Pentium 2
>Date: Tue, 21 Oct 2003 12:47:13 -0400 (EDT)
>On Tue, 21 Oct 2003, delta trinity wrote:
> > Well, this may be due to several things. First, I assume that both
> > processors are running at the same speed
> > First, we would have to see how much cache is present in both systems.
> > the P2 have less L1/L2 cache than the athlon
>I think the OP was saying that both tests were run on the **same** system.
>The only difference is that in one case he compiled the code **as if** it
>were on an Athlon, even though it was on a Pentium 2. Thus, the OP
>wonders why optimizing for an Athlon actually works better on a Pentium 2
>than optimizing for a Pentium 2.
See when your friends are online with MSN Messenger 6.0. Download it now
More information about the gmp-discuss