From owner-freebsd-bugs  Fri Mar 24 08:44:06 1995
Return-Path: bugs-owner
Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA08788 for bugs-outgoing; Fri, 24 Mar 1995 08:44:06 -0800
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id IAA08770 for <bugs@FreeBSD.ORG>; Fri, 24 Mar 1995 08:44:00 -0800
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id CAA26032; Sat, 25 Mar 1995 02:38:47 +1000
Date: Sat, 25 Mar 1995 02:38:47 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199503241638.CAA26032@godzilla.zeta.org.au>
To: emb@herman.tiac.net, wollman@halloran-eldar.lcs.mit.edu
Subject: Re: New Kernel, pppd crashing system.
Cc: bugs@FreeBSD.org
Sender: bugs-owner@FreeBSD.org
Precedence: bulk

>It's in an infinite loop.  If you had hit Ctl-Alt-Esc to get into the
>debugger, the `tr[ace]' command would have told you that you were in
>rn_walktree_from().  From there, you could have said `panic', which
>would put you back into the debugger, and then `cont' to take a crash
>dump.  (Unfortunately, GDB for some reason would not have seen the
>rn_walktree_from() stack frame.)

I think `panic' gets done in debugger context.  gcc doesn't understand
the debugger trap frame.  `call panic' might work better.  If it doesn't
then you can set $eip to somewhere that pushes suitable args and calls
`panic'.

Bruce