From owner-freebsd-ppc@FreeBSD.ORG Tue May 31 06:17:20 2005 Return-Path: X-Original-To: freebsd-ppc@freebsd.org Delivered-To: freebsd-ppc@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 109D316A41C for ; Tue, 31 May 2005 06:17:20 +0000 (GMT) (envelope-from tilman@arved.at) Received: from 21322530218.direct.eti.at (21322530218.direct.eti.at [213.225.30.218]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4B1CA43D1D for ; Tue, 31 May 2005 06:17:18 +0000 (GMT) (envelope-from tilman@arved.at) Received: from jim.arved.de (localhost [127.0.0.1]) by 21322530218.direct.eti.at (8.13.1/8.13.1) with ESMTP id j4V6HHcd082049; Tue, 31 May 2005 08:17:17 +0200 (CEST) (envelope-from tilman@arved.at) Received: (from arved@localhost) by jim.arved.de (8.13.3/8.13.1/Submit) id j4V6HGc1082048; Tue, 31 May 2005 08:17:16 +0200 (CEST) (envelope-from tilman@arved.at) X-Authentication-Warning: jim.arved.de: arved set sender to tilman@arved.at using -f Date: Tue, 31 May 2005 08:17:16 +0200 From: Tilman Linneweh To: peterg@ptree32.com.au Message-ID: <20050531061716.GP92909@arved.at> References: <818615e1.4e07ff3.819d600@dommail.onthenet.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <818615e1.4e07ff3.819d600@dommail.onthenet.com.au> User-Agent: Mutt/1.4.2.1i Cc: freebsd-ppc@freebsd.org Subject: Re: panic pmap_pvo_to pte: pvo 0x129e0800 pte does not match pte0x817bc0 in pmap_pteg_table X-BeenThere: freebsd-ppc@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the PowerPC List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 31 May 2005 06:17:20 -0000 * peterg@ptree32.com.au [2005-05-31 05:28]: > >> What was going on at the time it occurred ? > > > >Three to five package building processes, so load around 3 to 4 and > >a bit of Disk I/O. > >The box has 512 MB RAM but unfortunately no Swap, so not enough RAM > >could be a reason. I will try to add a Swap partition to see if > >it keeps the box more stable. > > I'd say not: it's probably more to do with the concurrency. Does it take long > to trigger ? What if you kick off 2/3x that many package builds ? Usually it happens after a few hours. I tried kicking the load above 10 by additional running make buildworld -j 10 and the machine froze within a few minutes, without breaking into ddb. Unrelated, i am getting different panic, which does not look ppc-specific: It happens reproducable when i try to compile the nss port (or the one included in mozilla) panic: mutex Giant not owned at /usr/src/sys/kern/kern_mutex.c:299 KDB: enter: panic [thread pid 20485 tid 100084 ] Stopped at 0x251a5c: lwz r11, r1, 0x0 db> where Tracing pid 20485 tid 100084 td 0x15d074b0 0xdd131a10: at panic+0x134 0xdd131ab0: at _mtx_assert+0x74 0xdd131ad0: at _mtx_unlock_flags+0xec 0xdd131b00: at do_execve+0x98c 0xdd131c70: at kern_execve+0xb4 0xdd131ca0: at execve+0x58 0xdd131ce0: at syscall+0x29c .. regards tilman