From owner-freebsd-hardware  Mon Feb 24 23:33:05 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id XAA04905
          for hardware-outgoing; Mon, 24 Feb 1997 23:33:05 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA04900
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 23:33:00 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id SAA00656; Tue, 25 Feb 1997 18:30:56 +1100
Date: Tue, 25 Feb 1997 18:30:56 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702250730.SAA00656@godzilla.zeta.org.au>
To: asami@vader.cs.berkeley.edu, jin@george.lbl.gov, mark@quickweb.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>Do not waste time to play this game. The "dd" is O.S. dependent code.

No, dd is very machine-independent.  It just loops calling read() and
write() with the specified block size.  However, the implementation
of /dev/zero is very machine-dependent.  FreeBSD happens to have an
implementation that copies memory in a straightforward way, so the speed
reported by dd is closely related to the memory write bandwidth.  The
read bandwidth doesn't matter much because most reads are from the
cache.

>It does not give you what is real memory speed on your system. The result
>from dd is really depended on the O.S. you are running. If you run 2.2 or
>higher, you will get much better performance than 2.1.x.

There isn't much difference unless you have a P5 and the P5-optimized
copyout routine is not disabled.

>440FX does have worse memory speed than Triton-{I,II}; even though P6 has
>much better CPU speed, but the PCI controller (440FX) is worse.

That would explain why a P6 is more than twice as slow as a P5 for large
writes.  It is inherently about twice as slow because P6's read cache
lines before writing to them.  Read-before-write is useful if the memory
is read after writing to it, but the dd benchmark never does that.

Bruce