From owner-freebsd-hackers  Sun Feb 24  7:37:29 2002
Delivered-To: freebsd-hackers@freebsd.org
Received: from scaup.prod.itd.earthlink.net (scaup.mail.pas.earthlink.net [207.217.120.49])
	by hub.freebsd.org (Postfix) with ESMTP id 3B43B37B402
	for <hackers@freebsd.org>; Sun, 24 Feb 2002 07:37:19 -0800 (PST)
Received: from pool0068.cvx40-bradley.dialup.earthlink.net ([216.244.42.68] helo=mindspring.com)
	by scaup.prod.itd.earthlink.net with esmtp (Exim 3.33 #1)
	id 16f0i6-0000rk-00; Sun, 24 Feb 2002 07:37:10 -0800
Message-ID: <3C79089B.10AD892C@mindspring.com>
Date: Sun, 24 Feb 2002 07:36:59 -0800
From: Terry Lambert <tlambert2@mindspring.com>
X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony}  (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Andrew Mobbs <andrewm@chiark.greenend.org.uk>
Cc: Matthew Dillon <dillon@apollo.backplane.com>, hackers@FreeBSD.ORG
Subject: Re: Test patch for msync/object-flushing performance (for stable)
References: <15478.31998.459219.178549@chiark.greenend.org.uk>
		<200202222042.g1MKg4u22700@apollo.backplane.com>
		<200202222201.g1MM1f431236@apollo.backplane.com> <15481.61.57511.222531@chiark.greenend.org.uk>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
List-ID: <freebsd-hackers.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20freebsd-hackers>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20freebsd-hackers>
X-Loop: FreeBSD.ORG

Andrew Mobbs wrote:
>               vm.msync_flush_flags
>        |   0   |   1   |   2   |   3   |
> -------+-------+-------+-------+-------|
> write  |  519  |  517  | 1632  |  519  |
> sync   | 2227  |  176  |  848  |  177  |
> -------+-------+-------+-------+-------|
> write  |  514  |  517  |  518  |  516  |
> sync   | 2215  |  175  | 2219  |  176  |
> -------+-------+-------+-------+-------|
> write  |  511  |  649  |  517  |  513  |
> sync   | 2209  |  125  | 2223  |  176  |
> -------+-------+-------+-------+-------|
> write  |  514  |  518  |  515  |  517  |
> sync   | 2217  |  176  | 2209  |  176  |
> -------+-------+-------+-------+-------|
> write  |  516  |  516  |  517  |  518  |
> sync   | 2219  |  176  | 2222  |  177  |
> 
> Nearly 13 times improvement in sync times, very nice.
                             ^       ^
Wow.  Pessimial... ----------'-------'

Looks like most (but not all) gets recovered if you also
set the other flag.  This argues against the "optimization"
enabled by bit 1, but OK's the one for bit 0.

Matt... is there a situation which you can provide a
small test for that doesn't make the bit 1 change look like
a total lose?

You might see it be a good idea in an unpack of the ports
tree over itself (for example), but right now it looks like
a no-brainer to leave it out, and default the bit 0 to on.

-- Terry

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message