From owner-freebsd-current@FreeBSD.ORG Wed Oct 8 10:48:48 2003 Return-Path: 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 9286916A4B3 for ; Wed, 8 Oct 2003 10:48:48 -0700 (PDT) Received: from srvdmz3.oekb.co.at (mailrelay2.oekb.co.at [143.245.5.93]) by mx1.FreeBSD.org (Postfix) with SMTP id 624FC43FE1 for ; Wed, 8 Oct 2003 10:48:42 -0700 (PDT) (envelope-from a@jenisch.at) Received: from 143.245.2.191 by srvdmz3.oekb.co.at (InterScan E-Mail VirusWall NT); Wed, 08 Oct 2003 19:48:40 +0200 Received: from jenisch.at ([143.245.83.20]) by MAIL01.oekb.co.at with Microsoft SMTPSVC(5.0.2195.5329); Wed, 8 Oct 2003 19:48:40 +0200 Message-ID: <3F844DF7.4090507@jenisch.at> Date: Wed, 08 Oct 2003 19:48:39 +0200 From: Ewald Jenisch User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030701 X-Accept-Language: en-us, en MIME-Version: 1.0 To: current@freebsd.org Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 08 Oct 2003 17:48:40.0343 (UTC) FILETIME=[68892270:01C38DC4] Subject: Upgrading 5.1 to latest -> page fault in kernel (crash) upon starting X X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 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: Wed, 08 Oct 2003 17:48:48 -0000 Hi, Seems like I'm running into a serious problem upgrading my 5.1 machine to the latest level: The symptoms: I did an upgrade (cvsup) then went through the usual build/install cycle (see below). When I boot the new kernel everything runs straight up to the moment when I start X-Windows. When starting X as a normal user I end up with a "db"-prompt (debugger?) - from there I can only "reset" :-(. (for details about the messages see below) When starting X as root the screen goes blank and after some seconds the box reboots. As far as building & installing is concerned I did everything as per chapter 21 of "the handbook", including installing binaries (userland) in single-user mode, running mergemaster etc.etc i.e. make buildworld; make buildkernel; make installkernel shutdown -r fsck -p mount -u / mount -a -t ufs swapon -a cd /usr/src make installworld When booting the old kernel (kernel.old) everything is fine again - so assume the problem is with the kernel itself. The "old" kernel is the one that has been installed when installing 5.1 from the CD. I also did a "diff" between the GENERIC file that came with the original installation and the one that was installed using cvsup (i.e. the one used in the fresh build) - no difference except the version string. I tried configuring "device mgadrm" into the new kernel (this wasn't there in the the original GENERIC though) - to no extent - same crash. Also: I've no indication about a problem in either /var/log/messages or the X-log - seems the box crashes before being able to write anything to the log. Has anybody else seen this? What can I do against it? Since I don't want to post exessive logs in the first place, here's what I've got: Asus CUSL2-mobo 512MB RAM 2 IDE HDs (80, 60GB) Matrox G400 graphics When upgrading I went from the version that comes on the 5.1-CD to the latest one available via cvsup (the "UPDATING" file has an entry of 2003-09-28) Please let me know if you need any additional details. Thanks much in advance for your help, -ewald ------------------------------< Cut here >----------------------------- Here's the output from "db" which is where I end in when I start X: (BTW - does anybody know a way how to capture this output besides "writing it down"?) Fatal trap 12: page fault while in kernel mode fault virtual address = 0x1c fault code: supervisor read, page not present Instruction pointer = 0x8: 0xc062e40e Stack pointer = 0x10: 0xdcaa0b68 framepointer = 0x10: 0xdcaa0b68 code segment = base 0x0, limit 0xfffff, type 0x1b DPL 0, pres 1, def32 1, gran 1 procesor eflags = interrupt enabled, resume, IOPL=0 current process = 595 (XFree86) kernel: type12 trap, code0 Stopped at _mtx_assert+0x4e: movl 0x1c(%ebx),%eax