From owner-freebsd-questions@FreeBSD.ORG  Fri Dec 16 02:05:19 2005
Return-Path: <owner-freebsd-questions@FreeBSD.ORG>
X-Original-To: freebsd-questions@freebsd.org
Delivered-To: freebsd-questions@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id A8B6016A41F
	for <freebsd-questions@freebsd.org>;
	Fri, 16 Dec 2005 02:05:19 +0000 (GMT)
	(envelope-from freebsd@sopwith.solgatos.com)
Received: from schitzo.solgatos.com (c-67-168-241-176.hsd1.or.comcast.net
	[67.168.241.176])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 489A343D5F
	for <freebsd-questions@freebsd.org>;
	Fri, 16 Dec 2005 02:05:18 +0000 (GMT)
	(envelope-from freebsd@sopwith.solgatos.com)
Received: from sopwith.solgatos.com (uucp@localhost)
	by schitzo.solgatos.com (8.11.6/8.11.6) with UUCP id jBG25Uf18072
	for freebsd-questions@freebsd.org; Thu, 15 Dec 2005 18:05:30 -0800
Received: from localhost by sopwith.solgatos.com (8.8.8/6.24)
	id BAA11562; Fri, 16 Dec 2005 01:59:57 GMT
Message-Id: <200512160159.BAA11562@sopwith.solgatos.com>
To: freebsd-questions@freebsd.org
Date: Thu, 15 Dec 2005 17:59:57 +0000
From: Dieter <freebsd@sopwith.solgatos.com>
Subject: NCQ   Re: Disk write caching, hw.ata.wc and atacontrol disagree)
X-BeenThere: freebsd-questions@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: freebsd@sopwith.solgatos.com
List-Id: User questions <freebsd-questions.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-questions>, 
	<mailto:freebsd-questions-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-questions>
List-Post: <mailto:freebsd-questions@freebsd.org>
List-Help: <mailto:freebsd-questions-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-questions>, 
	<mailto:freebsd-questions-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Fri, 16 Dec 2005 02:05:19 -0000

>> So hw.ata.wc thinks that the disk write cache is off,
>> but atacontrol cap thinks the disk write cache is on?

> I remember this is working but there was a reporting problem
> in atacontrol.

Looks like "atacontrol cap" reports the state of the disk *before*
hw.ata.wc took effect.

It would be less confusing if "atacontrol cap" reported the state
of the disk at the time atacontrol is run.

> You should easily be able to tell whether it is on or off by the
> write performance.  It should plummet by a factor of 4 or so.

I'm seeing only ~10% of the performance I get with the write cache on.

Which leads to:

Feature                      Support  Enable    Value           Vendor
Native Command Queuing (NCQ)   yes       -      31/0x1F

Where is the knob to turn on NCQ?  My controller and disks claim to support
it, but grepping the 6.0 man pages finds nothing.