From owner-freebsd-fs@FreeBSD.ORG  Wed Dec  3 13:51:44 2003
Return-Path: <owner-freebsd-fs@FreeBSD.ORG>
Delivered-To: freebsd-fs@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id 4D1AA16A4CE
	for <freebsd-fs@FreeBSD.org>; Wed,  3 Dec 2003 13:51:44 -0800 (PST)
Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163])
	by mx1.FreeBSD.org (Postfix) with ESMTP id AC66043FCB
	for <freebsd-fs@FreeBSD.org>; Wed,  3 Dec 2003 13:51:42 -0800 (PST)
	(envelope-from truckman@FreeBSD.org)
Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2])
	by gw.catspoiler.org (8.12.9p2/8.12.9) with ESMTP id hB3LpReF032726;
	Wed, 3 Dec 2003 13:51:31 -0800 (PST)
	(envelope-from truckman@FreeBSD.org)
Message-Id: <200312032151.hB3LpReF032726@gw.catspoiler.org>
Date: Wed, 3 Dec 2003 13:51:27 -0800 (PST)
From: Don Lewis <truckman@FreeBSD.org>
To: kmarx@vicor.com
In-Reply-To: <3FCD4396.1040100@vicor.com>
MIME-Version: 1.0
Content-Type: TEXT/plain; charset=us-ascii
cc: freebsd-fs@FreeBSD.org
cc: mckusick@beastie.mckusick.com
Subject: Re: 4.8 ffs_dirpref problem
X-BeenThere: freebsd-fs@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Filesystems <freebsd-fs.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-fs>,
	<mailto:freebsd-fs-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-fs>
List-Post: <mailto:freebsd-fs@freebsd.org>
List-Help: <mailto:freebsd-fs-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-fs>,
	<mailto:freebsd-fs-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 03 Dec 2003 21:51:44 -0000

On  2 Dec, Ken Marx wrote:
> 
> 
> Don Lewis wrote:

>> I went ahead and spun a new version of my patch with the new multiplier,
>> one other tweak to the formula, and updated comments.

> 
> Hi, Sorry for the delay. Tested, then realized a stupid mistake,
> started over, panic'd the box, and finally tested. Panic was
> a bit unsettling. Happened when I tried to call my dump routine
> from the debugger. Might have been a typo for all I know(!?) -
> was doing quickly to keep benchmark script from bailing due to
> "Profiling timer expired" killing my timer. (That's another kettle
> of wax, I guess.)
> 
> Other than that, the newer hashkey stuff does seem to improve
> the shallowness of the table. Below are some samples while doing
> the 1.4gb untar commands on 98% and 99%+ full disk. Again, 
> running w/ 75% avgbfree, avgifree so as to hit the table harder.

It looks good enough to me.  I just committed it.