From xemacs-m  Sat Jun 21 01:54:12 1997
Received: from iria.mines.u-nancy.fr (iria.mines.u-nancy.fr [193.49.140.100])
	by xemacs.org (8.8.5/8.8.5) with SMTP id BAA17616
	for <xemacs-beta@xemacs.org>; Sat, 21 Jun 1997 01:54:11 -0500 (CDT)
Received: (from galibert@localhost) by iria.mines.u-nancy.fr (950413.SGI.8.6.12/950213.SGI.AUTOCF) id IAA04271; Sat, 21 Jun 1997 08:53:03 +0200
Message-ID: <19970621085302.09544@iria.mines.u-nancy.fr>
Date: Sat, 21 Jun 1997 08:53:02 +0200
From: Olivier Galibert <Olivier.Galibert@mines.u-nancy.fr>
To: XEmacs Beta Test <xemacs-beta@xemacs.org>
Subject: Re: Yet more 64bits lossage
References: <19970621012424.18989@iria.mines.u-nancy.fr> <199706210604.XAA07502@xemacs.eng.sun.com> <QQcuvx10888.199706210628@crystal.WonderWorks.COM>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.74
In-Reply-To: <QQcuvx10888.199706210628@crystal.WonderWorks.COM>; from Kyle Jones on Sat, Jun 21, 1997 at 02:28:34AM -0400

On Sat, Jun 21, 1997 at 02:28:34AM -0400, Kyle Jones wrote:
> So if XEmacs can't use 64-bit pointers why compile with
> them?  (I'm assuming the person doing the build has a choice.)

This is not exactly true because of that (src/lisp.h):
#if (LONGBITS > INTBITS)
# define EMACS_INT long
# define EMACS_UINT unsigned long
#else
# define EMACS_INT int
# define EMACS_UINT unsigned int
#endif


Besides, for now and on irix the person has a choice. I don't know
for how long. The other existing 64bits OS being OSF/1, we see that
already something had to be done (src/m/alpha.h):
/* Define the type to use.  */
#define EMACS_INT long
#define EMACS_UINT unsigned long
#define SPECIAL_EMACS_INT


Since 64 bits systams are the upcoming fashion (sun Ultrasparc,
next intel/HP processor, maybe others) it's probably hard time
to clean this mess up since we have real production systems to
test on before it starts to bomb everywhere.

  OG.

