Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!cbatt!ihnp4!ptsfa!lll-lcc!styx!ames!ucbcad!ucbvax!decvax!decwrl!pyramid!prls!philabs!pwa-b!mmintl!franka
From: franka@mmintl.UUCP
Newsgroups: comp.lang.c
Subject: Re: short circuit evaluation - some comments
Message-ID: <2010@mmintl.UUCP>
Date: Thu, 26-Feb-87 15:21:55 EST
Article-I.D.: mmintl.2010
Posted: Thu Feb 26 15:21:55 1987
Date-Received: Sun, 1-Mar-87 16:26:36 EST
References: <813@cullvax.UUCP> <1287@ucbcad.berkeley.edu>
Reply-To: franka@mmintl.UUCP (Frank Adams)
Organization: Multimate International, E. Hartford, CT
Lines: 11

In article <1287@ucbcad.berkeley.edu> faustus@ucbcad.UUCP writes:
>I think if you're being really careful, you can't optimize away any
>arithmetic operations, because there is always the possible side effect
>of an arithmetic exception or memory error.

Not quite true.  Consider "x // 4 + y // 4", where x and y have the same
integral type.  The "+" operation can neither overflow nor cause a memory
error.

Frank Adams                           ihnp4!philabs!pwa-b!mmintl!franka
Ashton-Tate          52 Oakland Ave North         E. Hartford, CT 06108