Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!cbatt!ihnp4!inuxc!pur-ee!uiucdcs!uiucdcsm!mccaugh
From: mccaugh@uiucdcsm.UUCP
Newsgroups: comp.lang.c
Subject: Re: code optimizers (was Re: induction
Message-ID: <4700005@uiucdcsm>
Date: Sat, 21-Feb-87 00:53:00 EST
Article-I.D.: uiucdcsm.4700005
Posted: Sat Feb 21 00:53:00 1987
Date-Received: Mon, 23-Feb-87 02:28:24 EST
References: <745@bucsb.bu.edu.UUCP>
Lines: 13
Nf-ID: #R:bucsb.bu.edu.UUCP:745:uiucdcsm:4700005:000:746
Nf-From: uiucdcsm.cs.uiuc.edu!mccaugh    Feb 20 23:53:00 1987


 Just for the record:
 1) Having extensively used FTN-IV-G along with assembler programs, I can
    safely say that the "G" version was not the "optimized" FORTRAN;

 2) On the other hand, the "H" version was definitely optimized: unlike
    many compilers of its vintage (which were very memory-conscious), the
    "H"-optimizer bothered to keep the entire parse-tree intact for optimi-
    zation, which accounts for its particular success as a FTN-optimizer
    (the more info at your disposal -- even at the expense of memory -- the
    better position you are in to generate "optimal" code). For more details,
    consult "Compiler Construction for Digital Computers" by David Gries (2nd
    edition -- most readable if you speak IBM-360).