Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!ames!ptsfa!ihnp4!inuxc!pur-ee!uiucdcs!uiucdcsm!grunwald
From: grunwald@uiucdcsm.cs.uiuc.edu
Newsgroups: comp.arch
Subject: Re: Optimization vs. the programmer
Message-ID: <3300004@uiucdcsm>
Date: Sun, 19-Apr-87 16:15:00 EST
Article-I.D.: uiucdcsm.3300004
Posted: Sun Apr 19 16:15:00 1987
Date-Received: Tue, 21-Apr-87 01:22:56 EST
References: <479@danews.ATT.COM>
Lines: 11
Nf-ID: #R:danews.ATT.COM:479:uiucdcsm:3300004:000:430
Nf-From: uiucdcsm.cs.uiuc.edu!grunwald    Apr 19 15:15:00 1987


Yesszh, lambastization for a typo: yes, it's supposed to be 0x7, not 0x3.

As for my definition of ``reduction in strength'' -- this is the defn. as
put forth in the Dragon book. Then example they give is the PL/I-ish

	length(S1 || S2)	-> length(s1) + length(s2)

In the case where an expression involving a loop iterand is being reduced in
strength, I think that the term is really induction variable expansion and
elimination.