Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site sdcsvax.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!harvard!talcott!panda!genrad!decvax!ittatc!dcdwest!sdcsvax!darrell
From: darrell@sdcsvax.UUCP (Darrell Long)
Newsgroups: net.arch
Subject: Re: Re: RISC (really on multiplication d
Message-ID: <1003@sdcsvax.UUCP>
Date: Tue, 23-Jul-85 11:30:49 EDT
Article-I.D.: sdcsvax.1003
Posted: Tue Jul 23 11:30:49 1985
Date-Received: Fri, 26-Jul-85 04:13:35 EDT
References: <149@mips.UUCP> <600005@pbear.UUCP> <493@mmintl.UUCP> <233@weitek.UUCP>
Reply-To: darrell@sdcsvax.UUCP (Darrell Long)
Organization: EECS Dept. U.C. San Diego
Lines: 24

In article <233@weitek.UUCP> mahar@weitek.UUCP (mahar) writes:
>In article <493@mmintl.UUCP>, franka@mmintl.UUCP (Frank Adams) writes:
>> Which is that integer multiplies are hardly ever (never, in my
>> experience) defined in a manner suitable for their most common use.
>> I would guess, conservatively, that well over 90% of the integer multiplies
>> that are done generate a result with no more bits than the larger of the
>> two operands.  That is, one multiplies two words and wants a word result.
>> But hardware multiplies invariably generate a two word result, leaving the
>> high-order word to be allowed for and/or disposed of.
>
>The Decsystem-10 & 20 has an instruction IMUL which does exactly what you
>want. It Multiplies a 36bit value by another 36 bit value and gives a 36
>bit result. They also have a divide which divides a 36 bit number.

This is also true for the VAX-11 and the AT&T 3B series.  If you want
a two word result then you must ask for it explicitly (if you can get
it at all).
-- 
Darrell Long
Department of Electrical Engineering and Computer Science
University of California, San Diego

USENET: sdcsvax!darrell
ARPA:   darrell@sdcsvax