From owner-freebsd-current@FreeBSD.ORG  Mon Nov  8 18:15:03 2004
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Delivered-To: freebsd-current@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id 5764A16A4CE
	for <current@freebsd.org>; Mon,  8 Nov 2004 18:15:03 +0000 (GMT)
Received: from peedub.jennejohn.org (Jbd81.j.pppool.de [85.74.189.129])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 69AD843D48
	for <current@freebsd.org>; Mon,  8 Nov 2004 18:15:02 +0000 (GMT)
	(envelope-from garyj@jennejohn.org)
Received: from jennejohn.org (localhost [127.0.0.1])
	by peedub.jennejohn.org (8.13.1/8.11.6) with ESMTP id iA8IEsgd039802;
	Mon, 8 Nov 2004 19:15:00 +0100 (CET)
	(envelope-from garyj@jennejohn.org)
Message-Id: <200411081815.iA8IEsgd039802@peedub.jennejohn.org>
X-Mailer: exmh version 2.7.0 06/18/2004 with nmh-1.0.4
To: Divacky Roman <xdivac02@stud.fit.vutbr.cz>
In-Reply-To: Message from Divacky Roman <xdivac02@stud.fit.vutbr.cz> 
	<20041108173007.GA93246@stud.fit.vutbr.cz> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Mon, 08 Nov 2004 19:14:54 +0100
From: Gary Jennejohn <garyj@jennejohn.org>
cc: current@freebsd.org
Subject: Re: [HEADSUP] Please help test device vnode bypass! 
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
	<freebsd-current.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Mon, 08 Nov 2004 18:15:03 -0000


Divacky Roman writes:
> > >  On a 700MHz K7 machine this doubles the speed of
> > >          dd if=/dev/zero of=/dev/null bs=1 count=1000000
> > >  
> > >  This roughly translates to shaving 2usec of each read/write syscall.
> 
> I tried it (no problem yet) and did a bit testing...
> (cp -R /sys/ /tmp/s, dd if=/dev/zero of=smth count=1000000, cp smth smth1 and
> so on)
> 
> and it seems like its actually a bit slower (but everything works properly) -
> did I get anything wrong? should the results be a bit better (I did three run
> s
> every time of each test)
> 

You evidently missed the point that this change benefits accesses to
_devices_. All your tests are using files.

---
Gary Jennejohn / garyj[at]jennejohn.org gj[at]freebsd.org garyj[at]denx.de