From owner-freebsd-scsi@FreeBSD.ORG  Mon Jun  2 01:29:58 2003
Return-Path: <owner-freebsd-scsi@FreeBSD.ORG>
Delivered-To: freebsd-scsi@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id EDE8E37B401
	for <freebsd-scsi@freebsd.org>; Mon,  2 Jun 2003 01:29:57 -0700 (PDT)
Received: from matou.sibbald.com (matou.sibbald.com [195.202.201.48])
	by mx1.FreeBSD.org (Postfix) with ESMTP id DCDCB43F85
	for <freebsd-scsi@freebsd.org>; Mon,  2 Jun 2003 01:29:55 -0700 (PDT)
	(envelope-from kern@sibbald.com)
Received: from [192.168.68.112] (rufus [192.168.68.112])
	by matou.sibbald.com (8.11.6/8.11.6) with ESMTP id h528Tlv04599;
	Mon, 2 Jun 2003 10:29:47 +0200
From: Kern Sibbald <kern@sibbald.com>
To: mjacob@feral.com
In-Reply-To: <20030601165751.H97138@beppo>
References: <1054490081.1582.1685.camel@rufus>
	 <2846020000.1054498114@aslan.scsiguy.com>
	 <1054503893.1578.1723.camel@rufus>  <20030601165751.H97138@beppo>
Content-Type: text/plain
Organization: 
Message-Id: <1054542587.1578.1772.camel@rufus>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.2.4 
Date: 02 Jun 2003 10:29:47 +0200
Content-Transfer-Encoding: 7bit
cc: freebsd-scsi@freebsd.org
Subject: Re: SCSI tape data loss
X-BeenThere: freebsd-scsi@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: SCSI subsystem <freebsd-scsi.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-scsi>,
	<mailto:freebsd-scsi-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-scsi>
List-Post: <mailto:freebsd-scsi@freebsd.org>
List-Help: <mailto:freebsd-scsi-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-scsi>,
	<mailto:freebsd-scsi-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Mon, 02 Jun 2003 08:29:58 -0000

Yes, after a bit more thought, I realized this was the
case, thanks.

In any case, both buffered and async writes are turned off by
default.

On Mon, 2003-06-02 at 02:00, Matthew Jacob wrote:
> Of course linux has async && buffered. Linux has to copy the data from
> user space to kernel buffers and *then* write them. This leads to an
> obvious desire to overlap such writes. The same feature was available in
> Solaris 2.5 as well.
> 
> 'Buffering' as we talk about here typically means the device buffers
> themselves. You don't want to turn this off. You don't want to turn this
> off. You don't want to turn this off. The only device that I know of
> that really *has* to have this off is the old M4 1/2" reel drive because
> it would discard buffered data when it saw the early warning marker.
> 
> I have a longer answer to the previous mail about to go out.
> 
> -matt