From mboxrd@z Thu Jan  1 00:00:00 1970
Received: by margaux.inria.fr, Thu, 4 Feb 93 10:27:05 +0100
Received: from concorde.inria.fr by margaux.inria.fr, Wed, 3 Feb 93 23:46:02 +0100
Received: from Tamuz.Stanford.EDU by concorde.inria.fr; Wed, 3 Feb 1993 23:44:02 +0100
Received:  by Tamuz.Stanford.EDU (5.61/25-theory-eef) id AA13395; Wed, 3 Feb 93 14:43:56 -0800
From: Xavier Leroy <xavier@Theory.Stanford.EDU>
Message-Id: <9302032243.AA13395@Tamuz.Stanford.EDU>
Subject: Port of Caml Light to the Dec Alpha
To: caml-list@margaux, sml-list@cs.cmu.edu
Date: Wed, 3 Feb 1993 14:43:54 -0800 (PST)
Content-Type: text/plain
Sender: weis@margaux

I have ported Caml Light to the Dec Alpha under OSF/1.  This is the
first implementation of any ML dialect to run on a 64-bits
architecture.  Integers wrap around at 4611686018427387904, the size
of heap objects can go up to 64 million gigabytes, and the system
happily bootstraps itself in 36 seconds, which is three times faster
than on a Decstation 5000.

The patches are available by anonymous FTP from ftp.inria.fr (128.93.1.26),
directory lang/caml-light, file cl5-64bits.tar.Z. These patches are to
be untarred on top of the standard Caml Light 0.5 distribution (same
host, same directory, file cl5unix.tar.Z).

The patches might even work for other 64-bits architectures; who knows?

- Xavier Leroy