From owner-freebsd-hackers  Wed Mar  6  9:56:53 2002
Delivered-To: freebsd-hackers@freebsd.org
Received: from bingnet2.cc.binghamton.edu (bingnet2.cc.binghamton.edu [128.226.1.18])
	by hub.freebsd.org (Postfix) with ESMTP id 3B29B37B400
	for <freebsd-hackers@FreeBSD.ORG>; Wed,  6 Mar 2002 09:56:50 -0800 (PST)
Received: from onyx (onyx.cs.binghamton.edu [128.226.140.171])
	by bingnet2.cc.binghamton.edu (8.11.6/8.11.6) with ESMTP id g26HuXm06314;
	Wed, 6 Mar 2002 12:56:33 -0500 (EST)
Date: Wed, 6 Mar 2002 12:54:48 -0500 (EST)
From: Zhihui Zhang <zzhang@cs.binghamton.edu>
X-Sender: zzhang@onyx
To: Peter Pentchev <roam@ringlet.net>
Cc: Peter Edwards <peter.edwards@openet-telecom.com>,
	"Brian T.Schellenberger" <bts@babbleon.org>,
	Lars Eggert <larse@ISI.EDU>,
	"Rogier R. Mulhuijzen" <drwilco@drwilco.net>,
	Julian Elischer <julian@elischer.org>, freebsd-hackers@FreeBSD.ORG
Subject: Re: A weird disk behaviour
In-Reply-To: <20020306192418.G14052@straylight.oblivion.bg>
Message-ID: <Pine.SOL.4.21.0203061252180.6633-100000@onyx>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
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



On Wed, 6 Mar 2002, Peter Pentchev wrote:

> On Wed, Mar 06, 2002 at 12:12:44PM -0500, Zhihui Zhang wrote:
> > 
> > 
> > On Wed, 6 Mar 2002, Peter Edwards wrote:
> > 
> > > Zhihui Zhang wrote:
> > > 
> > > <snip>
> > > 
> > > > ... I also do not read anything during the partial block write,
> > > > and I think the disk controller should not do that either.
> > > 
> > > If you do a partial block write, surely at some point the block must be read
> > > in order to preserve that segment of data you are _not_ overwriting?
> > 
> > First off, I am not writing through any file system. I access the raw
> > device directly.  Secondly, the bytes written are always a multiple of 512
> > bytes. If one sector is the I/O unit of a disk controller, why should it
> > read anything to prevent overwritten?
> 
> I think Peter was referring to the (more common IMHO) case when one sector
> was not quite the I/O unit of the disk controller, especially WRT caching.
> That is, the disk controller does not actually do a physical disk write
> for each and every sector, but only in larger blocks.

This makes sense. So you are saying that the read-modify-write cycle
actually happens within the disk controller? If so, I would love to know
the unit of the disk controller I/O to avoid it!

-Zhihui


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