From xemacs-m  Fri Feb 28 10:32:41 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 KAA07420
	for <xemacs-beta@xemacs.org>; Fri, 28 Feb 1997 10:32:39 -0600 (CST)
Received: by crystal.WonderWorks.COM 
	id QQcewg16495; Fri, 28 Feb 1997 11:32:32 -0500 (EST)
Date: Fri, 28 Feb 1997 11:32:32 -0500 (EST)
Message-Id: <QQcewg16495.199702281632@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: somasekh@ecn.purdue.edu
Cc: XEmacs-Beta Listserv <xemacs-beta@xemacs.org>
Subject: Error while in VM xemacs-20.1-b3
In-Reply-To: <199702280714.CAA10390@ppp-x8-33.ecn.purdue.edu>
References: <199702280714.CAA10390@ppp-x8-33.ecn.purdue.edu>
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

Dinesh Somasekhar writes:
 > 
 > 
 > Had the following error while in VM.  clicked on the button in a mime message
 > and then closed the bottom window from the modeline ( Delete Window above ).
 > Got the following backtrace when I clicked on a message in VM.
 > 
 > Signaling: (error "Window height 0 too small (after splitting)")
 >   split-window(#<window on "*Backtrace*" 0x1df1> 0 nil)
 >   tapestry-apply-window-map((- (0 0 80 10) (0 10 80 40)) 80 40 #<window on "*Backtrace*" 0x1df1>)
 >   tapestry-set-window-map((- (0 0 80 10) (0 10 80 40)) nil)

This looks like one of those "this can't happen" type bugs.
split-window is being called with 0 as its second argument.  I
don't see how this can happen unless there's something rotten in
how XEmacs is doing math on your machine.

If you can reproduce the problem, use 'e' in the debugger window
to evaluate each of the following three s-expressions:

(tapestry-compute-map-height (car map))
root-window-height
map-height

I noticed in your build report that you built with gcc -O2 but
didn't use -fno-strength-reduce.  This is known to produce math
errors in some versions of GCC.

