From owner-freebsd-hardware  Fri Sep 12 09:42:27 1997
Return-Path: <owner-freebsd-hardware>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id JAA16332
          for hardware-outgoing; Fri, 12 Sep 1997 09:42:27 -0700 (PDT)
Received: from www2.shoppersnet.com (shoppersnet.com [204.156.152.112])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA16326
          for <freebsd-hardware@FreeBSD.ORG>; Fri, 12 Sep 1997 09:42:24 -0700 (PDT)
Received: (from hlew@localhost)
	by www2.shoppersnet.com (8.8.5/8.8.5) id JAA16957;
	Fri, 12 Sep 1997 09:42:03 -0700 (PDT)
Date: Fri, 12 Sep 1997 09:42:03 -0700 (PDT)
From: Howard Lew <hlew@www2.shoppersnet.com>
To: Satoshi Asami <asami@cs.berkeley.edu>
cc: roberto@keltia.freenix.fr, freebsd-hardware@FreeBSD.ORG
Subject: Re: IBM/Cyrix PR200+
In-Reply-To: <199709120431.VAA28321@silvia.HIP.Berkeley.EDU>
Message-ID: <Pine.BSF.3.91.970912092726.15696D-100000@www2.shoppersnet.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Thu, 11 Sep 1997, Satoshi Asami wrote:

>  * > My too -- however, i suppose that Pentium-oriented copy/zero block
>  * > procedures don't work.
>  * 
>  * Oh they do but much slower then the regular ones.
>  * 
>  * Always use "flags 0x7" for npx0 in your config. file for both Cyrix and
>  * AMD's processors.
> 
> Speaking of which, can anyone with an *MX processor check the gus mmx
> extension PR (filed under "gnu", I think) and try and see if copying
> through the mmx registers instead of FP would be faster for some of
> the CPU's?
> 
> Satoshi
> 

Didn't find the PR, but the flags do help a Cyrix 6x86MX get quite a bit 
more MB/sec with dd.