From owner-freebsd-current@FreeBSD.ORG  Mon Oct 20 15:35:00 2003
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Delivered-To: freebsd-current@freebsd.org
Received: from green.bikeshed.org (freefall.freebsd.org [216.136.204.21])
	by hub.freebsd.org (Postfix) with ESMTP
	id 045DB16A4B3; Mon, 20 Oct 2003 15:35:00 -0700 (PDT)
Received: from green.bikeshed.org (localhost [127.0.0.1])
	by green.bikeshed.org (8.12.10/8.12.9) with ESMTP id h9KMYxcR076143;
	Mon, 20 Oct 2003 18:34:59 -0400 (EDT)
	(envelope-from green@green.bikeshed.org)
Received: from localhost (green@localhost)h9KMYwvO076139;
	Mon, 20 Oct 2003 18:34:59 -0400 (EDT)
Message-Id: <200310202234.h9KMYwvO076139@green.bikeshed.org>
X-Mailer: exmh version 2.6.3 04/04/2003 with nmh-1.0.4
To: "Peter Edwards" <pmedwards@eircom.net>
In-Reply-To: Message from "Peter Edwards" <pmedwards@eircom.net> 
	<20030910182003.BF84744001@mx1.FreeBSD.org> 
From: "Brian F. Feldman" <green@freebsd.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: green@green.bikeshed.org
cc: current@freebsd.org
Subject: Re: useful workaround and analysis of vnode-backed md deadlock 
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
	<freebsd-current.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
Date: Mon, 20 Oct 2003 22:35:00 -0000
X-Original-Date: Mon, 20 Oct 2003 18:34:58 -0400
X-List-Received-Date: Mon, 20 Oct 2003 22:35:00 -0000

"Peter Edwards" <pmedwards@eircom.net> wrote:
> Once the running buffer space exceeds a high threshold, the scheduler
> is blocked, to be awakened when completed async writes bring it
> under the low threshold. However, a large chunk of the running buf
> space is sitting queued for the md thread to process. The md thread
> can't continue without the vnode lock, so the running buffer space
> will not fall, and the bufdaemon cannot continue without running
> buffer space, so will never release the vnode lock.

This coincides with what I just discovered on my system, as well; did you 
try the patch from my post several days ago that just excludes the buffer 
flushing daemons from being used to wait for runningbufspace starvation to 
end?

-- 
Brian Fundakowski Feldman                           \'[ FreeBSD ]''''''''''\
  <> green@FreeBSD.org                               \  The Power to Serve! \
 Opinions expressed are my own.                       \,,,,,,,,,,,,,,,,,,,,,,\