From owner-freebsd-current@FreeBSD.ORG Fri Jul 3 01:24:32 2009 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 94765106564A for ; Fri, 3 Jul 2009 01:24:32 +0000 (UTC) (envelope-from gcr+freebsd-current@tharned.org) Received: from roadkill.tharned.org (roadkill.tharned.org [75.145.12.185]) by mx1.freebsd.org (Postfix) with ESMTP id 513278FC1B for ; Fri, 3 Jul 2009 01:24:32 +0000 (UTC) (envelope-from gcr+freebsd-current@tharned.org) Received: from blue.tharned.org (blue.tharned.org [10.10.10.8]) (authenticated bits=0) by roadkill.tharned.org (8.14.3/8.14.3) with ESMTP id n631OUkF086578 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Thu, 2 Jul 2009 20:24:31 -0500 (CDT) (envelope-from gcr+freebsd-current@tharned.org) Date: Thu, 2 Jul 2009 20:24:30 -0500 (CDT) From: Greg Rivers To: Kostik Belousov In-Reply-To: <20090702194444.GK2884@deviant.kiev.zoral.com.ua> Message-ID: References: <20090701192154.GW2884@deviant.kiev.zoral.com.ua> <20090702084149.GE2884@deviant.kiev.zoral.com.ua> <20090702194444.GK2884@deviant.kiev.zoral.com.ua> User-Agent: Alpine 2.00 (BSF 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; format=flowed; charset=US-ASCII X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.2 (roadkill.tharned.org [75.145.12.185]); Thu, 02 Jul 2009 20:24:31 -0500 (CDT) Cc: alc@freebsd.org, freebsd-current@freebsd.org Subject: Re: Panic during shutdown (cause identified) X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 03 Jul 2009 01:24:32 -0000 On Thu, 2 Jul 2009, Kostik Belousov wrote: > The information you provided about amd(8) causing the problem was > crusial. The issue is that amd locks its pages with mlockall(2), and > the code neglected to account the wired mappings, but did not forgot to > decrease their swap share on unmapping. > > Patch below fixed the issue for me. > [snip] > Confirmed: your patch fixes the issue for me as well. Thanks! -- Greg Rivers