From owner-freebsd-current  Fri Apr 14 17:34:50 1995
Return-Path: current-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.10/8.6.6) id RAA19450
          for current-outgoing; Fri, 14 Apr 1995 17:34:50 -0700
Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241])
          by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id RAA19444
          for <current@FreeBSD.org>; Fri, 14 Apr 1995 17:34:42 -0700
Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id RAA02037; Fri, 14 Apr 1995 17:31:38 -0700
From: "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com>
Message-Id: <199504150031.RAA02037@gndrsh.aac.dev.com>
Subject: Re: Interesting (and odd) effect in -current
To: cmf@ins.infonet.net (Carl M. Fongheiser)
Date: Fri, 14 Apr 1995 17:31:37 -0700 (PDT)
Cc: current@FreeBSD.org
In-Reply-To: <199504142209.RAA00277@kf0yn.ampr.org> from "Carl M. Fongheiser" at Apr 14, 95 05:10:08 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1851      
Sender: current-owner@FreeBSD.org
Precedence: bulk

> 
> Thanks to hardware trouble, I haven't had a chance to boot a reasonably
> current -current kernel until a few days ago.  In particular, this is
> the first kernel that didn't have the memory test in it.  The first time
> I booted it (warm boot), it booted fine.  I ran with it for a couple days.
> Wednesday night, we had a power failure, and I discovered that the thing
> wouldn't cold boot with that kernel.  It gets all the way through the
> autoboot, but it trapped when it started up a getty due to a kernel reference
> to address 0.  It failed again on the reboot.  My old kernel (from late Jan.)
> booted fine.  A perusal of the crash dump showed the trap happened inside
> of scopen().  Yesterday, I supped again, and built another kernel.  This
> one falls over the same way, but slightly earlier in stty.  This time it
> looks like it happened in spec_ioctl().
> 
> Just to summarize, my late January kernel runs flawlessly, but my 2 most
> recent kernels sometimes fall over late in the boot sequence.  I don't know
> if this could be uncovering some flakiness in my hardware, or if there's
> something that isn't getting initialized properly.  My hardware configuration:
> 
> PCI/I-486SP3G motherboard, 256k L2 cache, 8M main memory.
        ^^^^^^^              ^^^^^^^^^^^^^

Make sure your L2 cache is wet to write through mode on that motherboard,
also disable GAT in the chipset BIOS setup screen.

> 2 Maxtor 7345AT drives
> 1 Quantum LT730S drive
> STB 4COM (4-port serial card)
> 3COM 3C503 ethernet card
> #9 GXE64 PCI video card
> 
> Any help would be appreciated, and I'd be happy to supply further details
> on request.
> 
> Carl Fongheiser
> cmf@ins.infonet.net
> 


-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                   Custom computers for FreeBSD