From owner-freebsd-current  Mon Nov 24 23:00:17 1997
Return-Path: <owner-freebsd-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id XAA18444
          for current-outgoing; Mon, 24 Nov 1997 23:00:17 -0800 (PST)
          (envelope-from owner-freebsd-current)
Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA18437
          for <freebsd-current@FreeBSD.ORG>; Mon, 24 Nov 1997 23:00:11 -0800 (PST)
          (envelope-from mike@word.smith.net.au)
Received: from word (localhost [127.0.0.1])
	by word.smith.net.au (8.8.7/8.8.5) with ESMTP id RAA00340;
	Tue, 25 Nov 1997 17:25:09 +1030 (CST)
Message-Id: <199711250655.RAA00340@word.smith.net.au>
X-Mailer: exmh version 2.0zeta 7/24/97
To: Nate Williams <nate@mt.sri.com>
cc: shimon@simon-shapiro.org, freebsd-current@FreeBSD.ORG
Subject: Re: Compiler Bug??? 
In-reply-to: Your message of "Mon, 24 Nov 1997 23:03:02 PDT."
             <199711250603.XAA25821@mt.sri.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Tue, 25 Nov 1997 17:25:08 +1030
From: Mike Smith <mike@smith.net.au>
Sender: owner-freebsd-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> Guess what, you have bad memory!  Seriously, this is not a compiler bug,
> but something in your hardware that is hosed up.  Either the memory is
> bad, the timings are bad, the cache is bad, or some combination of the
> above is bad.  Compiler bugs don't go away on reboots, but corrupted
> memory does.

... just to clarify here, as I am reading Simon slightly differently 
from you;

 - code built while running UP works.
 - code built while running SMP fails.

Simon, are you using the -pipe option with gcc?  Have you tried without 
it?  What optimisation level are you using?

I'm inclined to agree with Nate in that it's unlikely that you have 
found a compiler bug, more likely an environment (eg. kernel) bug that 
is causing you some serious grief.

mike