Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site alice.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!alice!ark
From: ark@alice.UucP (Andrew Koenig)
Newsgroups: net.lang.c,net.arch
Subject: Re: Integer division
Message-ID: <4947@alice.uUCp>
Date: Thu, 6-Feb-86 23:11:12 EST
Article-I.D.: alice.4947
Posted: Thu Feb  6 23:11:12 1986
Date-Received: Sun, 9-Feb-86 04:46:08 EST
References: <731@abic.UUCP>
Organization: Bell Labs, Murray Hill
Lines: 6
Xref: watmath net.lang.c:7792 net.arch:2477

To add some more fuel to the fire, consider extending a%b to
floating-point operands.  If we do this, we find that defining
a%b to have the sign of a in all cases allows a%b to be represented
exactly as a floating-point number, whereas giving a%b the sign
of b does not.  Consider, for instance, the case where b is huge
and positive and a is tiny and negative.