From owner-freebsd-current@FreeBSD.ORG  Wed Sep 28 07:26:41 2005
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
X-Original-To: current@FreeBSD.org
Delivered-To: freebsd-current@FreeBSD.ORG
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id 2881416A41F
	for <current@FreeBSD.org>; Wed, 28 Sep 2005 07:26:41 +0000 (GMT)
	(envelope-from truckman@FreeBSD.org)
Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163])
	by mx1.FreeBSD.org (Postfix) with ESMTP id B94B843D49
	for <current@FreeBSD.org>; Wed, 28 Sep 2005 07:26:40 +0000 (GMT)
	(envelope-from truckman@FreeBSD.org)
Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2])
	by gw.catspoiler.org (8.13.3/8.13.3) with ESMTP id j8S7QX2V000750
	for <current@freebsd.org>; Wed, 28 Sep 2005 00:26:37 -0700 (PDT)
	(envelope-from truckman@FreeBSD.org)
Message-Id: <200509280726.j8S7QX2V000750@gw.catspoiler.org>
Date: Wed, 28 Sep 2005 00:26:33 -0700 (PDT)
From: Don Lewis <truckman@FreeBSD.org>
To: current@FreeBSD.org
In-Reply-To: <200509280609.j8S68vZ3000590@gw.catspoiler.org>
MIME-Version: 1.0
Content-Type: TEXT/plain; charset=us-ascii
Cc: 
Subject: Re: analysis of snapshot-related system deadlock
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.5
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>
X-List-Received-Date: Wed, 28 Sep 2005 07:26:41 -0000

On 27 Sep, Don Lewis wrote:

> Probably the easiest fix would be to call runningbufwakeup() from
> ffs_copyonwrite() before grabbing "snaplk", and increase runningbufspace
> again before returning from ffs_copyonwrite().  The bufs waiting for
> "snaplk" aren't yet async writes currently running, to borrow from the
> comment on waitrunningbufspace().

Sleeping on "wdrain" is probably also undesirable if a thread holds
"snaplk" because of the potential impact on the activity of other
threads.  A possible fix would be for bufwrite() to skip the call to
waitrunningbufspace() if the thread holds "snaplk".  This might cause
problems with queuing an excessive amount of I/O, but I don't think so
because the outer bufwrite() will still call waitrunningbufspace().