From xemacs-m  Sun Mar 23 13:32:54 1997
Received: from xemacs.cs.uiuc.edu (localhost [127.0.0.1])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id NAA08536;
	Sun, 23 Mar 1997 13:32:53 -0600 (CST)
Message-Id: <199703231932.NAA08536@xemacs.org>
To: Kyle Jones <kyle_jones@wonderworks.com>
cc: XEmacs Developers <xemacs-beta@xemacs.org>
Subject: Re: Buffer-specific display-tables 
In-reply-to: Your message of "Sun, 23 Mar 1997 14:10:28 EST."
             <QQcido20311.199703231910@crystal.WonderWorks.COM> 
Date: Sun, 23 Mar 1997 13:32:51 -0600
From: Chuck Thompson <cthomp@xemacs.org>

    >> There is no probably about it.  The call to get_display_table
    >> is in what is pretty much the heart of the redisplay engine.
    >> The Second Rule of Redisplay is definitely in effect.  Perhaps
    >> more accurately the First Corollary of the Second Rule which is
    >> 'Thou Shalt Not GC

    Kyle> Something has got to give.  The display code need to
    Kyle> reference specifiers.  specifier_instance can call eval,
    Kyle> which means it can GC.  So I think the instantiators that
    Kyle> can call eval have got to go.

Actually, no it can't.  I can't believe how rapidly one can forget how
everything works.  Garbage collection is inhibited while redisplay is
active.  Check out the variable in_display.


			-Chuck

