Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site mmintl.UUCP
Path: utzoo!linus!philabs!pwa-b!mmintl!franka
From: franka@mmintl.UUCP (Frank Adams)
Newsgroups: net.lang.c
Subject: Re: Right shift vs. divide
Message-ID: <937@mmintl.UUCP>
Date: Fri, 27-Dec-85 00:50:26 EST
Article-I.D.: mmintl.937
Posted: Fri Dec 27 00:50:26 1985
Date-Received: Sat, 28-Dec-85 13:15:58 EST
References: <974@brl-tgr.ARPA>
Reply-To: franka@mmintl.UUCP (Frank Adams)
Organization: Multimate International, E. Hartford, CT
Lines: 17

In article <974@brl-tgr.ARPA> Schauble@mit-multics.arpa (Paul Schauble) writes:
>This optimization only works
>if the numbers are known to be non-negative. A Pascal compiler may make
>this optimization, becausethe programmer may specify the range for
>variables, but not C.

And you were doing so well up to this point! :-)  Of course a C compiler
can make this optimization, because the variables may be declared unsigned.

By the way, the FORTRAN compiler for the DEC-10 would process an integer
divide by two by checking the sign, incrementing if negative, and then
doing a right shift.  Since the divide took special setup (the single vs
double length operands problem), this didn't necessarily even take more
space than a divide instruction.

Frank Adams                           ihpn4!philabs!pwa-b!mmintl!franka
Multimate International    52 Oakland Ave North    E. Hartford, CT 06108