Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/5/84; site yetti.UUCP
Path: utzoo!utcs!mnetor!yetti!oz
From: oz@yetti.UUCP (Ozan Yigit)
Newsgroups: net.emacs,net.lang,net.wanted.sources
Subject: Re: Info wanted on Y compiler
Message-ID: <248@yetti.UUCP>
Date: Tue, 3-Sep-85 09:32:46 EDT
Article-I.D.: yetti.248
Posted: Tue Sep  3 09:32:46 1985
Date-Received: Tue, 3-Sep-85 16:19:37 EDT
References: <2455@mit-hermes.ARPA> <496@tymix.UUCP>
Reply-To: oz@yetti.UUCP (Ozan Yigit)
Distribution: net.emacs,net.lang
Organization: York University Computer Science
Lines: 34
Xref: utcs net.emacs:1221 net.lang:1741 net.wanted.sources:1288
Summary: 

In article <496@tymix.UUCP> granvold@tymix.UUCP (Tom Granvold) writes:
>-
>    OK, I'll bite. What is a Y compiler?
>
>Tom Granvold
>Tymnet
	Y is a language designed in the Dept. of Computer Science,
	University of Arizona. It is meant to replace Ratfor for
	software tools type applications. Y is a true compiler, which
	looks very much like ratfor. It lacks structures and pointers.
	Data structures: int, char, real.

	I installed it. It is very easy to re-target to a new architecture,
	although I have not done any re-targeting.

	With regards to the compiler technology: Y compiler is a standard
	recursive-descent compiler. No fancy technology there. The really
	interesting part is "po", a superp peephole-optimizer that can
	be set to work with Y compiler. "po" uses *architecture descriptions*
	for re-targeting. The typical output of "po" is just as good, or
	better in most part, than your standard C compiler in terms of
	code quality.

Oz



-- 
Usenet: [decvax|allegra|linus|ihnp4]!utzoo!yetti!oz
Bitnet: oz@[yusol|yuyetti]
	You see things; and you say "WHY?"
	But I dream things that never were;
	and say "WHY NOT?"
			G. Bernard Shaw (Back to Methuselah)