From owner-freebsd-current  Sat Apr 13 16:56:29 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id QAA08333
          for current-outgoing; Sat, 13 Apr 1996 16:56:29 -0700 (PDT)
Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id QAA08328
          for <current@FreeBSD.org>; Sat, 13 Apr 1996 16:56:27 -0700 (PDT)
Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id QAA05980; Sat, 13 Apr 1996 16:53:50 -0700
From: Terry Lambert <terry@lambert.org>
Message-Id: <199604132353.QAA05980@phaeton.artisoft.com>
Subject: Re: ed_start() panic revisited
To: scrappy@ki.net (Marc G. Fournier)
Date: Sat, 13 Apr 1996 16:53:50 -0700 (MST)
Cc: current@FreeBSD.org
In-Reply-To: <Pine.BSI.3.92.960412232310.206C-100000@freebsd.ki.net> from "Marc G. Fournier" at Apr 12, 96 11:28:39 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-current@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

> 	Just had an ed_start() panic tonight, after compiling a new
> kernel and forgetting to recompile the if_ed.c module without -O
> enabled (as per Julian's suggestion)

[ ... ]

> 316
> 317                     switch (type) {
> 318                     case T_PAGEFLT:                 /* page fault */
> 319                             (void) trap_pfault(&frame, FALSE);
> 320                             return;
> 321
> 322                     case T_PROTFLT:         /* general protection fault */
> 323                     case T_SEGNPFLT:        /* segment not present fault */
> 
> 
> 	Julian had suggested recompiling the if_ed.c module seperately from
> the rest, without -O enabled, to see if that made a difference, as the
> original core dump seemed to show everything was okay.  Without -O enabled,
> the system seems to run okay (well, other then the rlist panics, of course),
> but with -O enabled, it dies after several hours.
> 
> 	Is there anything i can do on my end to figure out the difference
> between the optimized vs non-optimized if_ed.o module in order to correct
> this problem?  Some sort of 'diff' between the object files maybe?

volatile frame? type?


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.