Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!ll-xn!mit-eddie!killer!pollux!dalsqnt!rpp386!jfh
From: jfh@rpp386.UUCP (John F. Haugh II)
Newsgroups: comp.lang.c
Subject: Re: Shifting question
Message-ID: <4246@rpp386.UUCP>
Date: 22 Jul 88 00:49:49 GMT
References: <705@bnr-rsc.UUCP> <11556@steinmetz.ge.com> <60290@sun.uucp> <1818@spar.SPAR.SLB.COM> <60670@sun.uucp>
Reply-To: jfh@rpp386.UUCP (The Beach Bum)
Organization: Big "D" Home for Wayward Hackers
Lines: 35

In article <60670@sun.uucp> alanf%smile@Sun.COM (Alan Fargusson) writes:
>In article <1818@spar.SPAR.SLB.COM>, hunt@spar.SPAR.SLB.COM (Neil Hunt) writes:
>> We are talking factors of two in execution time for these functions.
>> What a pain !!
>
>What is your solution?

the solution which neil missed was to sacrifice code size.  he wrote

>> 	rshift = count > 0 ? count : 0;
>> 	lshift = count < 0 ? -count : 0;
>> 
>> 	for(j = 0; j < image->rows; j++)
>> 		for(i = 0; i < image->cols; i++)
>> 			image->pixels[i][j] =

where the test could have been

	if (count < 0) {
		count = - count;
		for ... 
			image->pixels[i][j] = (image->pixels[i][j] >> count);
	} else {
		for ...
			image->pixels[i][j] = (image->pixels[i][j] << count);
	}

this does not suffer a significant decrease in performance, the only added
overhead being the initial test and the possible negation.

- john.
-- 
John F. Haugh II                 +--------- Cute Chocolate Quote ---------
HASA, "S" Division               | "USENET should not be confused with
UUCP:   killer!rpp386!jfh        |  something that matters, like CHOCOLATE"
DOMAIN: jfh@rpp386.uucp          |             -- with my apologizes