Sender:	nate AT cartsys DOT com
Message-ID:	<35BD6187.A151342E@cartsys.com>
Date:	Mon, 27 Jul 1998 22:28:39 -0700
From:	Nate Eldredge <nate AT cartsys DOT com>
MIME-Version:	1.0
To:	"Salvador Eduardo Tropea (SET)" <salvador AT inti DOT gov DOT ar>
CC:	EDWARD CASHIN <cashin AT pollux DOT cs DOT uga DOT edu>, djgpp AT delorie DOT com
Subject:	Re: implementation of modulo operator in djgpp
References:	<m0z0rEP-000S4rC AT inti DOT gov DOT ar>

Salvador Eduardo Tropea (SET) wrote:
> 
> EDWARD CASHIN <cashin AT pollux DOT cs DOT uga DOT edu> wrote:

> > ...which begs the question, "How fast is the modulo?"
> > and more specifically, how is it implemented?
> 
> Modulo is just the rest of a division. So if last is a variable (not a macro
> and not a constant) you'll get a div or idiv and that's sllloooowwww.
> But I think you must call gcc with -S to take a look at the generated
> assembler ;-)

True.  Suggested optimization:

Make the buffer size a power of 2.  Then do:

front = (front + 1) & (size - 1);
-- 

Nate Eldredge
nate AT cartsys DOT com

- Raw text -