Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 SMI; site sun.uucp
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!petrus!bellcore!decvax!decwrl!sun!wmb
From: wmb@sun.uucp (Mitch Bradley)
Newsgroups: net.lang.forth
Subject: Re: Forth standards
Message-ID: <3429@sun.uucp>
Date: Tue, 1-Apr-86 17:34:01 EST
Article-I.D.: sun.3429
Posted: Tue Apr  1 17:34:01 1986
Date-Received: Sat, 5-Apr-86 04:15:27 EST
References: <337@tut.UUCP>
Organization: Sun Microsystems, Inc.
Lines: 23

> jty@tut says:
>    Anyway, I heard Mitch Bradley (wmb@sun) is writing a portable UNIX forth
>    in C, can somebody (wmb?) verify this?

Well, yes, I have written a portable UNIX forth in C.  I guess I better
post it.  I have hesitated to do so because I don't have time to support it.

>    It is possible to write a HUGE switch-statement (in C) to
>    implement the virtual Forth machine, but this is VERY S L O W W w w ..

In fact that is the way it is implemented, but the good news is that
it is NOT slow.  On a Sun, it is only about 2.5 times slower than my
68000 assembly language Forth.  I don't think I would want to use it
on an 8080, but on most modern processors, it is quite acceptable.

>    The question is, will it support "separate I and D space"?

Well, sort of.  The basic Forth kernel is separate text and data, but
when you incrementally compile stuff on top of that and save the
executable image, the incrementally compiled stuff is all in the data
segment.

Mitch