From xemacs-m  Sun Jun 22 20:56:57 1997
Received: from crystal.WonderWorks.COM (crystal.WonderWorks.com [192.203.206.1])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id UAA21679
	for <xemacs-beta@xemacs.org>; Sun, 22 Jun 1997 20:56:55 -0500 (CDT)
Received: by crystal.WonderWorks.COM 
	id QQcvcp25247; Sun, 22 Jun 1997 21:56:56 -0400 (EDT)
Date: Sun, 22 Jun 1997 21:56:56 -0400 (EDT)
Message-Id: <QQcvcp25247.199706230156@crystal.WonderWorks.COM>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
From: Kyle Jones <kyle_jones@wonderworks.com>
To: XEmacs Developers <xemacs-beta@xemacs.org>
Subject: `skip_debugger'
In-Reply-To: <kigu3iqpqr4.fsf@jagor.srce.hr>
References: <kigu3iqpqr4.fsf@jagor.srce.hr>
X-Mailer: VM 6.33 under 20.3 "Copenhagen" XEmacs Lucid (beta8)
X-Face: /cA45WHG7jWq>(O3&Z57Y<"WsX5ddc,4c#w0F*zrV#=M
        0@~@,s;b,aMtR5Sqs"+nU.z^CSFQ9t`z2>W,S,]:[+2^
        Nbf6v4g>!&,7R4Ot4Wg{&tm=WX7P["9%a)_da48-^tGy
        ,qz]Z,Zz\{E.,]'EO+F)@$KtF&V

This looks like a compiler bug.  Looking at the stacktrace,
0xbfffd058 is a plausible address for a stack position at
that point.  The compiler could have pushed the pointer to a
value instead of the value on the stack as the argument to
Ferror_message_string.

My advice is to back off on the optimizations, or maybe back off
the bleeding edge compiler version.  I plugged in Hrvoje's
version of skip_debugger and there is no crash here.

	config: i386-pc-bsdi2.0
      compiler: gcc 2.6.3
compiler flags: -g -O4 -m486 -fforce-mem -fforce-addr
		-funroll-loops -fno-strength-reduce
	     X: X11R6
    scrollbars: Lucid
      menubars: Lucid
       dialogs: Athena
	 sound: no
	   XPM: yes
	X-Face: yes
	   PNG: yes
	   GIF: yes
          JPEG: yes
     rel-alloc: no

