Path: utzoo!attcan!utgpu!watmath!att!mtuxo!mtgzz!drutx!druhi!kam
From: kam@druhi.ATT.COM (Karen A. Morrissey)
Newsgroups: comp.lang.prolog
Subject: Re: searching for C-Prolog
Summary: everything but Apollo
Keywords: C-Prolog
Message-ID: <4512@druhi.ATT.COM>
Date: 16 Aug 89 19:45:08 GMT
References: <740@uniol.UUCP>
Distribution: comp
Organization: AT&T, Denver, CO
Lines: 27


I haven't ported to Apollo, but when we were using C-PROLOG (several
versions), I had it running on:

	VAX-11
	Amdahl (S/370 clones)
	AT&T 3B20
	AT&T 3B2 (and other WE32xxx based machines)
	AT&T 7300

C-Prolog 1.5 and later will compile and run on most any machine with
32-bit longs (C data type) and 16-bit shorts (C data type).

Pointers and tagged values are merged, using the assumption that the
high bit isn't used for a valid address (untrue on some machines).
This assumption isn't hard to remove. (I don't have the source anymore
-- not my job -- and it's been 3 years, so I can't tell you where to
look other than that it is in one of the main header files. I also
recall having to fix some place where 0 was considered a valid DB
reference.)

Questions to:

Karen Morrissey
AT&T Bell Labs
303-538-4587
att!druhi!kam