From owner-freebsd-stable@FreeBSD.ORG Tue Apr 24 07:23:31 2012 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id B7B1C1065672 for ; Tue, 24 Apr 2012 07:23:31 +0000 (UTC) (envelope-from dave@jetcafe.org) Received: from hugeraid.jetcafe.org (hugeraid.jetcafe.org [205.147.26.109]) by mx1.freebsd.org (Postfix) with ESMTP id 9B2A88FC0C for ; Tue, 24 Apr 2012 07:23:31 +0000 (UTC) Received: from hugeraid.jetcafe.org (localhost [127.0.0.1]) by hugeraid.jetcafe.org (8.13.8/8.13.8) with ESMTP id q3O7AQaQ075735 for ; Tue, 24 Apr 2012 00:10:26 -0700 (PDT) Message-Id: <201204240710.q3O7AQaQ075735@hugeraid.jetcafe.org> X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.0.4 To: freebsd-stable@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 24 Apr 2012 00:10:26 -0700 From: Dave Hayes Subject: Attempting to boot into ramdisk on 8.3 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 24 Apr 2012 07:23:31 -0000 I have a build process (which worked at release 7.3) that makes a bootable ISO using a ramdisk image as the boot volume. At release 8 it panics right after reporting the real memory size with: kernel trap 12 with interrupts disabled Fatal trap 12: page fault while in kernel mode ... [thread pid 0 tid 0 ] Stopped at pmap_enter+0x19a: moveq (%rcx),%r14 (I have the text frozen on another screen if more data is needed, I've literally spent weeks trying to get this panic to appear on the screen so it's not going anywhere.) I build the world with this system without cross compiling. # uname -mrs FreeBSD 8.3-RC1 amd64 The system that paniced is very similar: a 64-bit amd4 system. What am I doing wrong, if anything? How can I do a successful large ramdisk boot? My process is to copy the entire OS to a ramdisk and boot off of that. I use the following ideas for this: /boot/loader.conf: mfsroot_type="mfs_root" mfsroot_name="/mfsboot" vfs.root.mountfrom="ufs:md0" vfs.root.mountfrom.options="rw" ## Tunables kern.ipc.nmbclusters=32768 net.inet.tcp.tcbhashsize=16384 vm.pmap.pg_ps_enabled=1 accf_http_load="YES" net.inet.tcp.syncache.hashsize=1024 net.inet.tcp.syncache.bucketlimit=100 The size of mfsboot is 600M. The diff between GENERIC and my kernel config, comments left in so people can see what I was doing 7.3ish: 28.43d26 < ### FBCD64 SPECIFIC < #options MD_ROOT_SIZE="524288" < options GEOM_UZIP < #options INCLUDE_CONFIG_FILE < #options ZERO_COPY_SOCKETS < #options HZ=1000 < #options DEVICE_POLLING < #options NKPT=600 # not set up on amd64 < # < # Debugging < options KDB < options DDB < options BREAK_TO_DEBUGGER < options ALT_BREAK_TO_DEBUGGER Thanks in advance for any assistance! -- Dave Hayes - Consultant - Altadena CA, USA - dave@jetcafe.org >>> The opinions expressed above are entirely my own <<< One of the most common defenses against really learning something is to believe that one knows it already.