From owner-freebsd-hardware  Thu May 29 08:03:07 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id IAA17957
          for hardware-outgoing; Thu, 29 May 1997 08:03:07 -0700 (PDT)
Received: from lariat.lariat.org (ppp0.lariat.org@[129.72.251.2])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA17952
          for <hardware@FreeBSD.ORG>; Thu, 29 May 1997 08:03:03 -0700 (PDT)
Received: from solo.lariat.org ([129.72.251.10])
	by lariat.lariat.org (8.8.5/8.8.5) with SMTP id JAA04761;
	Thu, 29 May 1997 09:01:17 -0600 (MDT)
Message-Id: <3.0.1.32.19970529080242.006b44ec@lariat.org>
X-Sender: brett@lariat.org
X-Mailer: Windows Eudora Pro Version 3.0.1 (32)
Date: Thu, 29 May 1997 08:02:42 -0600
To: bde@zeta.org.au, asami@vader.cs.berkeley.edu, roberto@keltia.freenix.fr
From: Brett Glass <brett@lariat.org>
Subject: Re: Intel Pentium II released
Cc: hardware@FreeBSD.ORG
In-Reply-To: <882564A6.0037ED8A.00@IWND1.infoworld.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

Something to test: would an MMX copy be faster than the FP copy? The K6 is
claimed to have particularly fast transfers when used for MMX.

--Brett


At 07:51 PM 5/29/97 +1100, bde@zeta.org.au wrote:
 
>>Bruce, do you think you can add a code to disable the FP-copy when the
>>chip is a 586-class but not from "GenuineIntel" (or whatever)?  The
>>only 586-class non-Intel CPU out there is the K6 (K5 too?), right?
>
>CPU_M2 also has class CPU_586.  I don't know if the K6 is a CPU_586 or
>a CPU_M2.  It's not really either.  The FP-copy attachment code shouldn't
>know about this.  It should run some tests and pick the fastest method.
>
>Bruce