From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 5 Feb 2002 03:09:36 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 5 Feb 2002 03:09:27 -0500 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:2316 "EHLO www.linux.org.uk") by vger.kernel.org with ESMTP id ; Tue, 5 Feb 2002 03:09:25 -0500 Message-ID: <3C5F9312.DB8766B9@zip.com.au> Date: Tue, 05 Feb 2002 00:08:50 -0800 From: Andrew Morton X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.18-pre7 i686) X-Accept-Language: en MIME-Version: 1.0 To: spider@darkmere.wanfear.com CC: linux-kernel@vger.kernel.org Subject: Re: [BUG] Oops booting linux-2.4.18-pre7-ac3 In-Reply-To: <20020205021933.34aff42f.spider@darkmere.wanfear.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Spider wrote: > > This is pretty much the same as "oops booting 2.4.18-pre7-ac3" by "Todd M. Roy" but I post it as well. > > I'm not subscribed, so if you wish more information (.config, other info on hardware/software installed) please cc' me. > > //Spider > > PCI: PCI BIOS revision 2.10 entry at 0xf1150, last bus=1 > PCI: Using configuration type 1 > PCI: Probing PCI hardware > PCI: Using IRQ router VIA [1106/0686] at 00:04.0 > PCI: Disabling Via external APIC routing > PnPBIOS: Found PnP BIOS installation structure at 0xc00fc2b0. > PnPBIOS: PnP BIOS version 1.0, entry 0xf0000:0xc2e0, dseg 0xf0000. > PnPBIOS: 12 nodes reported by PnP BIOS; 12 recorded by driver. > PnPBIOS: PNP0c02: ioport range 0xe400-0xe47f has been reserved. > PnPBIOS: PNP0c02: ioport range 0xe800-0xe83f could not be reserved. > Linux NET4.0 for Linux 2.4 > Based upon Swansea University Computer Society NET3.039 > Initializing RT netlink socket > Unable to handle kernel NULL pointer dereference at virtual address 0000002c Yes, sorry - I didn't test against Alan's pnpbios driver. The problem is this, in init/main.c: #ifdef CONFIG_PNPBIOS pnpbios_init(); #endif .... start_context_thread(); do_initcalls(); pnpbios_init() launches a kernel thread which calls the kernel/user.c code before the user.c's initcall has had a chance to initialise things. I've asked Alan to back out the rather unimportant set_user-in-reparent_to_init fix while we contemplate this. I suggest you disable the pnpbios driver in kernel config in -ac3. -