From owner-freebsd-current@FreeBSD.ORG Mon Oct 29 17:07:43 2007 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 ADE7016A468 for ; Mon, 29 Oct 2007 17:07:43 +0000 (UTC) (envelope-from gkozyrev@gmail.com) Received: from nf-out-0910.google.com (nf-out-0910.google.com [64.233.182.188]) by mx1.freebsd.org (Postfix) with ESMTP id 4361D13C4AC for ; Mon, 29 Oct 2007 17:07:31 +0000 (UTC) (envelope-from gkozyrev@gmail.com) Received: by nf-out-0910.google.com with SMTP id b2so1281409nfb for ; Mon, 29 Oct 2007 10:07:00 -0700 (PDT) Received: by 10.78.138.6 with SMTP id l6mr4039713hud.1193657818386; Mon, 29 Oct 2007 04:36:58 -0700 (PDT) Received: by 10.78.148.14 with HTTP; Mon, 29 Oct 2007 04:36:58 -0700 (PDT) Message-ID: Date: Mon, 29 Oct 2007 13:36:58 +0200 From: "Gleb Kozyrev" To: "Kris Kennaway" In-Reply-To: <4725B7EA.5080807@FreeBSD.org> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <200710251435.58984.jhb@freebsd.org> <200710261222.28656.jhb@freebsd.org> <20071026184536.GE37471@deviant.kiev.zoral.com.ua> <4725B7EA.5080807@FreeBSD.org> Cc: freebsd-current@freebsd.org Subject: Re: Deadlock, exclusive sx so_rcv_sx, amd64 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: Mon, 29 Oct 2007 17:07:43 -0000 On 29/10/2007, Kris Kennaway wrote: > > It now just takes longer until freeze with shpgperproc = 5000 (originally 200). > > > > So how about vm.pmap.pv_entry_max? Have you obtained the usual > debugging information (DDB backtraces, etc) when the system freezes? vm.pmap.pv_entry_max and vm.pmap.shpgperproc turn out to be related and raising one of them automatically raises the other. If there's a bug those sysctls won't help, will they? If there's no bug what values can you suggest? Defaults are vm.pmap.pv_entry_max: 1479365 vm.pmap.shpgperproc: 200 Backtraces, etc are provided in this thread. -- With best regards, Gleb Kozyrev.