Path: utzoo!utgpu!water!watmath!clyde!rutgers!rochester!PT.CS.CMU.EDU!andrew.cmu.edu!rs4u+
From: rs4u+@andrew.cmu.edu (Richard Siegel)
Newsgroups: comp.sys.mac
Subject: Re: Code Optimization
Message-ID: <gW09A1y00Uo1ktk0mF@andrew.cmu.edu>
Date: 29 Jan 88 13:56:17 GMT
Organization: Carnegie Mellon University
Lines: 38


gary@fairlight.oz (Gary Evesson) writes:

"It doesn't seem that any of the manufacturers of compilers for these machines 
are interested (I would welcome a contradiction from anyone) as far as I can 
tell."

	You're going to get your contradiction. :-)

THINK Technologies, for one, is interested in producing high-quality 
compilers. However, your must realize the following: when producing a 
revolutionary language system such as LightspeedC or Lightspeed Pascal, it's 
of the first order of importance to release a product that works correctly: 
there should be as few bugs as possible in the environment and in the 
compiler.
	Writing an optimizing code generator is, as you say, an art. It's damned 
difficult, and it takes a long time to debug such an animal. Had we chosen to 
write LightspeedC or Lightspeed Pascal as multi-pass optimizing compilers, the 
release of both products very well could have been delayed by some 
(indefinite) amount of time. And these products might or might not have had 
compiler bugs.

	As it is now, both compilers have a minimum of codegen bugs, and both work 
fairly reliably. Since I'm no longer on the spot at THINK, I don't know what 
kind of codegenerator work is going on, but I can tell you that we're 
committed to producing high-quality compilers for the Mac.

(I know, it sounds like an advertising plug, but it's true. :-))

		--Rich

===================================================================
Richard Siegel
THINK Technologies, QA Technician (on leave)

Arpa: rs4u@andrew.cmu.edu
UUCP: {decvax,ucbvax,sun}!andrew.cmu.edu!rs4u
==================================================================