Article View: gnu.gcc.help
Article #49553Re: Is there a better way than this to multiply 16.16 fixed-point on i386?
From: Eyal Lebedinsky
Date: Tue, 04 Jan 1994 22:40
Date: Tue, 04 Jan 1994 22:40
15 lines
1374 bytes
1374 bytes
In <2gbr49$sem@agate.berkeley.edu> benrg@uclink.berkeley.edu (Ben Aaron Rudiak-Gould) writes: >I'm writing a program in DJGPP that uses fixed-point math. It needs to >run as quickly as possible. Currently I'm implementing multiplication >and division with the following functions: [stuff deleted] One suggestion: create a mutiply/divide function too. You will be surprised how often these come together, and it is very efficient to do the multiply 16x16->32 the immediately do the 32/16->16. And most of the time you will find the same fixed point on both sides of the divide so there will be no need for any shifting. -- Regards Eyal Lebedinsky eyal@ise.canberra.edu.au
Technical Details
Message-ID:
<eyal.757723251@fir>Path:
arc1-ca.pugleaf.net!gmd.de!xlink.net!sol.ctr.columbia.edu!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!munnari.oz.au!newshost.anu.edu.au!csc.canberra.edu.au!fir!eyalReferences:
<2gbr49$sem@agate.berkeley.edu>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 11318c,4f1ae26683e618bc X-Google-Attributes: gid11318c,public X-Google-ArrivalTime: 1994-01-04 15:21:59 PST Newsgroups: gnu.gcc.help Path: arc1-ca.pugleaf.net!gmd.de!xlink.net!sol.ctr.columbia.edu!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!munnari.oz.au!newshost.anu.edu.au!csc.canberra.edu.au!fir!eyal From: eyal@fir.canberra.edu.au (Eyal Lebedinsky) Subject: Re: Is there a better way than this to multiply 16.16 fixed-point on i386? Message-ID: <eyal.757723251@fir> Sender: news@csc.canberra.edu.au Organization: University of Canberra References: <2gbr49$sem@agate.berkeley.edu> Date: 4 Jan 94 22:40:51 GMT Lines: 15