From owner-freebsd-current  Tue May 21 03:57:38 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id DAA16797
          for current-outgoing; Tue, 21 May 1996 03:57:38 -0700 (PDT)
Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id DAA16792
          for <current@freebsd.org>; Tue, 21 May 1996 03:57:35 -0700 (PDT)
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.5/8.6.9) id DAA00774; Tue, 21 May 1996 03:55:26 -0700 (PDT)
Date: Tue, 21 May 1996 03:55:26 -0700 (PDT)
Message-Id: <199605211055.DAA00774@silvia.HIP.Berkeley.EDU>
To: wscott@ichips.intel.com, wollman@lcs.mit.edu
CC: current@freebsd.org, nisha@cs.berkeley.edu
Subject: P6 memory copy speed
From: asami@cs.berkeley.edu (Satoshi Asami)
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Hello guys,

I would like to know more about the P6 bcopy speeds.  Specifically,
what exactly are the chipsets on your systems?  Something like

===
>> dmesg | grep chip
chip0 <Intel 82437 (Triton)> rev 2 on pci0:0
chip1 <Intel 82371 (Triton)> rev 2 on pci0:7
===

would be great.  Also, make and brand of motherboard would help too.

By the way, Wayne, you said that you can get better numbers from your
"server" system.  Can you re-run the set of tests on that system so we
can compare it side by side with the "desktop" version?

If someone out there has a P6 and would like to chime in, please look
at:

http://stampede.cs.berkeley.edu/~asami/Td/bcopy.html

There is also a link to the gzipped tarball for the test programs in
that page.

Thanks
Satoshi