Re: Loop Unroll

"Mike Dimmick" <mike@dimmick.demon.co.uk>
21 May 2001 02:01:53 -0400

          From comp.compilers
Related articles
Loop Unroll lxh@arch.cs.pku.edu.cn (Alkaid) (2001-05-18)
Re: Loop Unroll mike@dimmick.demon.co.uk (Mike Dimmick) (2001-05-21)
Re: Loop Unroll gvmt@bgl.vsnl.net.in (Venkatesha Murthy G.) (2001-05-21)
Re: Loop Unroll samiam@cisco.com (Scott Moore) (2001-05-21)
Re: Loop Unroll rkrayhawk@aol.com (2001-05-21)
Re: Loop Unroll christian.bau@isltd.insignia.com (Christian Bau) (2001-05-22)
| List of all articles for this month |
From: "Mike Dimmick" <mike@dimmick.demon.co.uk>
Newsgroups: comp.compilers
Date: 21 May 2001 02:01:53 -0400
Organization: Compilers Central
References: 01-05-046
Keywords: optimize, comment
Posted-Date: 21 May 2001 02:01:53 EDT

"Alkaid" <lxh@arch.cs.pku.edu.cn> wrote in message
news:01-05-046@comp.compilers...
> Hi,
>
> When we do ILP optimization, loop unrolling is a kind of method.
> But there is a question: if loop time is a big prime number, how to unroll
> the loop?
> e.g. when loop time is 100, I can unroll the loop 4 times to get a new one
> which has 25 iterations. What shall I do if the loop time is 101?


The obvious solution is to unroll it four times, to get 25 iterations, then
follow the loop with another copy of the contents of the loop, to be
executed one further time.


This increases the size of the program code, but should still increase its
speed (fewer jumps).


HTH,


--
Mike Dimmick
[Many other people made similar suggestions. -John]

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.