From owner-freebsd-current@FreeBSD.ORG Thu Aug 18 17:11:05 2011 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 288711065676; Thu, 18 Aug 2011 17:11:05 +0000 (UTC) (envelope-from jhb@freebsd.org) Received: from cyrus.watson.org (cyrus.watson.org [65.122.17.42]) by mx1.freebsd.org (Postfix) with ESMTP id F1FA18FC1F; Thu, 18 Aug 2011 17:11:04 +0000 (UTC) Received: from bigwig.baldwin.cx (66.111.2.69.static.nyinternet.net [66.111.2.69]) by cyrus.watson.org (Postfix) with ESMTPSA id 9618C46B2E; Thu, 18 Aug 2011 13:11:04 -0400 (EDT) Received: from jhbbsd.localnet (unknown [209.249.190.124]) by bigwig.baldwin.cx (Postfix) with ESMTPSA id 347E08A02F; Thu, 18 Aug 2011 13:11:04 -0400 (EDT) From: John Baldwin To: freebsd-current@freebsd.org Date: Thu, 18 Aug 2011 11:21:54 -0400 User-Agent: KMail/1.13.5 (FreeBSD/8.2-CBSD-20110617; KDE/4.5.5; amd64; ; ) References: <20110816215953.GB35017@server.vk2pj.dyndns.org> In-Reply-To: <20110816215953.GB35017@server.vk2pj.dyndns.org> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Message-Id: <201108181121.54918.jhb@freebsd.org> X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.6 (bigwig.baldwin.cx); Thu, 18 Aug 2011 13:11:04 -0400 (EDT) Cc: pho@freebsd.org, Peter Jeremy Subject: Re: "witness_lock_list_get: witness exhausted" 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: Thu, 18 Aug 2011 17:11:05 -0000 On Tuesday, August 16, 2011 5:59:53 pm Peter Jeremy wrote: > I'm getting the above message when running Peter Holm's stress > test with INCARNATIONS=150 on a 16-core sparc. Does this mean > LOCK_CHILDCOUNT is too low or does it indicate a leak in witness > lock_list_entry's somewhere? Most likely the former. > The comment on LOCK_CHILDCOUNT indicates it is dimensioned to > allow 2048 threads to hold 5 locks each but it's not clear how > many threads will get started at a given INCARNATIONS count. Also, each lle holds a bucket of instances, so the same count would only let 1024 threads hold 6 locks each perhaps (depending on the bucket size.. it was originally 4 locks per lle). -- John Baldwin