Message-Id: <m0xp81S-000S2XC@inti.gov.ar>
Comments: Authenticated sender is <salvador AT natacha DOT inti DOT gov DOT ar>
From: "Salvador Eduardo Tropea (SET)" <salvador AT inti DOT gov DOT ar>
Organization: INTI
To: Geza Herman <geza AT inf DOT bme DOT hu>, djgpp AT delorie DOT com
Date: Mon, 5 Jan 1998 11:41:01 +0000
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: Loop unrolling
In-reply-to: <Pine.GSO.3.95.980105091641.16302D-100000@kempelen.inf.bme.hu>
Precedence: bulk

Geza Herman <geza AT inf DOT bme DOT hu> wrote:
> 
> 	Why does the info of the GNU compiler say( in article 
> 	-funroll-all-loops ): 'unrolling usually makes the program
> 	slower'? I think unrolling loops makes the program faster!
 
Because the code is much greater making less efective the cache.
 
> PS: I know that's not a DJGPP specific question( but GNU C ).

But I think that it applies more for architectures like i386, not sure.

SET
------------------------------------ 0 --------------------------------
Visit my home page: http://set-soft.home.ml.org/
or
http://www.geocities.com/SiliconValley/Vista/6552/
Salvador Eduardo Tropea (SET). (Electronics Engineer)
Alternative e-mail: set-sot AT usa DOT net - ICQ: 2951574
Address: Curapaligue 2124, Caseros, 3 de Febrero
Buenos Aires, (1678), ARGENTINA
TE: +(541) 759 0013