Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site brl-smoke.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!whuxl!whuxlm!akgua!gatech!seismo!brl-tgr!brl-smoke!ron
From: ron@brl-smoke.ARPA (Ron Natalie <ron>)
Newsgroups: net.arch
Subject: Re: Right shift vs. divide (change divide!)
Message-ID: <605@brl-smoke.ARPA>
Date: Wed, 15-Jan-86 12:26:37 EST
Article-I.D.: brl-smok.605
Posted: Wed Jan 15 12:26:37 1986
Date-Received: Fri, 17-Jan-86 06:22:45 EST
References: <124000005@ima.UUCP> <4772@alice.UUCP> <1016@turtlevax.UUCP> <32@calgary.UUCP> <285@mips.UUCP> <1032@mmintl.UUCP>
Organization: Ballistic Research Lab
Lines: 10

I don't know how this discussion managed to leak over from net.lang.c,
but it is not a C issue.  C makes no assumption on whether integer divides
of negative operands truncates towards zero or -inf.  If both operands are
positive trucation is toward zero.

C only guarantees that the integer divide and the modulo function are
consistant with respect to each other, i.e.

	(a/b)*a + a%b equals a
-Ron