From owner-freebsd-current Sat Oct 17 19:41:04 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA23634 for freebsd-current-outgoing; Sat, 17 Oct 1998 19:41:04 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA23596 for ; Sat, 17 Oct 1998 19:40:37 -0700 (PDT) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.9.1/8.9.1/Spinner) with ESMTP id KAA16030; Sun, 18 Oct 1998 10:39:14 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199810180239.KAA16030@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: Tom Jackson cc: Karl Pielorz , Chuck Robey , freebsd-current@FreeBSD.ORG Subject: Re: panic In-reply-to: Your message of "Sat, 17 Oct 1998 20:23:05 EST." <19981017202305.A12163@TOJ.org> Date: Sun, 18 Oct 1998 10:39:13 +0800 From: Peter Wemm Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Tom Jackson wrote: > nm tells me module_register (dugh) not module_init. > > For me, no amount of dumps on anything will produce a crash dump. On my > laptop it just says bad dump device and on my server (which is not crashing > now) I get a count down but no crash modules are produced. > > btw, these are 100% elf and I don't think the panic is a smp problem. > > Maybe Peter will know what the problem is. I can guess, but I need more information. Can you mail me your kernel config file[s] that are breaking? My gut reaction is that the sysinit table is getting corrupted, or that something is calling module_register too soon. module_init() initializes a couple of tailq's, and all sorts of things go wrong if something tries to add to these lists before then. As a long shot, can you add #define DIAGNOSTIC to the start of kern/kern_malloc.c and see if the fault address changes to something along the lines of 0xdeadc0de. How far does it get into the boot sequence before crashing? What are the last few messages? It is possible that some filesystem initialization that you are using that I didn't test with is causing this. > On Sun, Oct 18, 1998 at 12:15:22AM +0100, Karl Pielorz wrote: > > > > Tom Jackson wrote: > > > > > [snip] > > > > > > I'm getting this crash, not on elf(kernel+exec)/smp/scsi but on a old > > > Thinkpad with a pccard ed0, a elf(kernel+exec)/up/ide. Thought since the > > > reference to modules that pcvt instead of sc0 might work - no such luck. > > > I get the panic right after the pcic is probed and before the pccard is > > > initialized. > > > > > > I'm not getting any crash dumps on any of my elf systems. > > > > Hi, > > > > I've been talking to Chuck about this - if the panic is similar / the same as > > his, you won't get a crashdump - as it's happening too soon into the boot > > process... > > > > The only thing I could think of was to have a line similar to: > > > > config kernel root on XXX dumps on YYYY > > > > In the kernel config - Though LINT warns against this, it might be the only > > way of telling the system where to dump to - as it's panics so soon during > > boot... > > > > I'm hoping someone who knows more what there talking about will step in wit h > > "That's a good/bad idea" - and why... > > > > I seem to remember using the kernel config to specify the dump device, but > > that was from ages ago (2.2.2? :-) > > > > Regards, > > > > Karl > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-current" in the body of the message > > > > -- > Tom > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-current" in the body of the message > Cheers, -Peter -- Peter Wemm Netplex Consulting "No coffee, No workee!" :-) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message