Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site umcp-cs.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!umcp-cs!chris
From: chris@umcp-cs.UUCP (Chris Torek)
Newsgroups: net.lang.c
Subject: Re: Division With Shifts
Message-ID: <2711@umcp-cs.UUCP>
Date: Thu, 9-Jan-86 23:01:42 EST
Article-I.D.: umcp-cs.2711
Posted: Thu Jan  9 23:01:42 1986
Date-Received: Sat, 11-Jan-86 06:15:45 EST
References: <1395@brl-tgr.ARPA>
Organization: U of Maryland, Computer Science Dept., College Park, MD
Lines: 12

I think everyone agrees that for nonnegative integers,
(i >> 1) == (i / 2).  So let us ignore those.  For negative
integers, (i >> 1) tends to round towards -infinity instead
of towards zero.  To fix this all you need to do is add one
before shifting.

I was going to show why, but it hardly seems worth the effort.
It just works.
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 4251)
UUCP:	seismo!umcp-cs!chris
CSNet:	chris@umcp-cs		ARPA:	chris@mimsy.umd.edu