From owner-freebsd-hackers  Wed May 30  2:10:26 2001
Delivered-To: freebsd-hackers@freebsd.org
Received: from smtp6.mindspring.com (smtp6.mindspring.com [207.69.200.110])
	by hub.freebsd.org (Postfix) with ESMTP id 8CE3137B422
	for <hackers@FreeBSD.ORG>; Wed, 30 May 2001 02:10:23 -0700 (PDT)
	(envelope-from tlambert2@mindspring.com)
Received: from mindspring.com (pool0246.cvx7-bradley.dialup.earthlink.net [209.178.164.246])
	by smtp6.mindspring.com (8.9.3/8.8.5) with ESMTP id FAA24118;
	Wed, 30 May 2001 05:10:17 -0400 (EDT)
Message-ID: <3B14B912.B870E4F4@mindspring.com>
Date: Wed, 30 May 2001 02:10:42 -0700
From: Terry Lambert <tlambert2@mindspring.com>
Reply-To: tlambert2@mindspring.com
X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony}  (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Ed Hudson <elh_fbsd@spnet.com>
Cc: hackers@FreeBSD.ORG
Subject: Re: general speed differences between 4.1.1-RELEASE and 4.3-RELEASE
References: <200105260621.f4Q6L6911677@m44.spnet.com>
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?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo?subject=subscribe%20freebsd-hackers>
List-Unsubscribe: <mailto:majordomo?subject=unsubscribe%20freebsd-hackers>
X-Loop: FreeBSD.ORG

Ed Hudson wrote:
> 
> the cost of soft updates, and the cost of hw.ata.wc=0
> 
> enclosed is a .jpeg of an xgraph of the following interactive test:

[ ... ]

>         hw.ata.wc=0, soft-updates enables.
>         hw.ata.wc=0, soft-updates disabled.
>         hw.ata.wc=1, soft-updates disabled.
> 
> the 'points' in the graph are the only real data (the lines
> are xgraph's interpolations).

A methodology comment...

I'd like to see:

         hw.ata.wc=1, soft-updates enabled.

Realize that it is the write caching, not the soft updates,
which makes write caching dangerous... it's just as dangerous
without soft updates, and mounted sync.  The drive lies to us
about whether or not the data was committed to stable storage,
which is where the problem is coming from.

Also, I can't tell if you are using sync or async mounts
in the non-soft-updates case: I would suggest doing both.

FWIW: I was just as annoyed when /dev/random went in, and
shot embedded 386 performance all to hell... at least I
actually get something from the write caching tradeoff.

-- Terry

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