Xref: utzoo sci.math:9425 comp.arch:13351 comp.lang.c:25326 comp.sources.wanted:10248
Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!crdgw1!crdos1!davidsen
From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr)
Newsgroups: sci.math,comp.arch,comp.lang.c,comp.sources.wanted
Subject: Re: Integer Multiply/Divide on Sparc
Message-ID: <2042@crdos1.crd.ge.COM>
Date: 24 Jan 90 16:30:14 GMT
References: <84768@linus.UUCP> <15418@vlsisj.VLSI.COM> <1543@milton.acs.washington.edu>
Reply-To: davidsen@crdos1.crd.ge.com (bill davidsen)
Followup-To: sci.math
Organization: GE Corp R&D Center, Schenectady NY
Lines: 16


  Actually memory is cheap enough to use 64k for a lookup table. Make a
16 bit address from the two bytes to multiply as
	0      78      15
	|_______|______|
	| byte1 | byte2|
	|_______|______|
and just pull the answer out of memory.

  Actually, the Intel practice of treating a 32 bit register as lots of
little registers, with the 8 and 16 bit portions addressible by name,
would be handy for some of this stuff, eliminating a shift. It should
still be quite fast.
-- 
bill davidsen	(davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen)
            "Stupidity, like virtue, is its own reward" -me