From owner-freebsd-hackers  Sat Apr 13 19:25:03 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id TAA14682
          for hackers-outgoing; Sat, 13 Apr 1996 19:25:03 -0700 (PDT)
Received: from uuneo.neosoft.com (root@uuneo.neosoft.com [206.109.1.3])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id TAA14673
          for <hackers@freebsd.org>; Sat, 13 Apr 1996 19:24:59 -0700 (PDT)
Received: from nmti.com (ficc@localhost) by uuneo.neosoft.com (8.7.5/8.7.4) with UUCP id VAA10779 for hackers@freebsd.org; Sat, 13 Apr 1996 21:01:06 -0500 (CDT)
Received: from sonic.nmti.com (peter@sonic.nmti.com [198.178.0.2]) by web.nmti.com (8.6.12/8.6.9) with SMTP id UAA06537 for <hackers@freebsd.org>; Sat, 13 Apr 1996 20:52:52 -0500
Received: by sonic.nmti.com; id AA14757; Sat, 13 Apr 1996 20:52:50 -0500
From: peter@nmti.com (Peter da Silva)
Message-Id: <9604140152.AA14757@sonic.nmti.com.nmti.com>
Subject: Pentium fast copy?
To: hackers@freebsd.org
Date: Sat, 13 Apr 1996 20:52:50 -0500 (CDT)
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

The following quote from a paper (sorry, I don't know what the paper is
offhand) appeared on the Firewalls mailing list. What's FreeBSD like on
this front?

    Our results show that none of the systems adequately delivers the
    Pentium's memory write performance.  For example, the Pentium can
    copy data at over 160 megabytes/second using a prefetching copy
    routine, yet none of the systems we tested have implemented such a
    routine.  As described below, the prefetching routines address the
    fact that the Pentium does not have a write-allocate cache.
    Without this optimization, the same routines copy data at about 40
    megabytes/second.