Google Groups Home
Help | Sign in
Auto vectorizing compiler ?
There are currently too many topics in this group that display first. To make this topic appear first, remove this option from another topic.
There was an error processing your request. Please try again.
flag
  2 messages - Collapse all
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
Neal  
View profile
 More options Jul 3, 5:52 pm
Newsgroups: comp.compilers
From: Neal <nealcr...@gmail.com>
Date: Thu, 3 Jul 2008 14:52:09 -0700 (PDT)
Local: Thurs, Jul 3 2008 5:52 pm
Subject: Auto vectorizing compiler ?
I've just started working on programming models for many-core
machines, and was wondering what the state of the art in terms of
parallelization of serial code is. I'm most interested in SIMD
vectorization right now... does anyone have a point to articles,
websites, research etc on this topic? Thanks!

Neal


    Reply to author    Forward  
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
kamal  
View profile
 More options Jul 5, 12:36 pm
Newsgroups: comp.compilers
From: kamal <kama...@hp.com>
Date: Sat, 5 Jul 2008 09:36:29 -0700 (PDT)
Local: Sat, Jul 5 2008 12:36 pm
Subject: Re: Auto vectorizing compiler ?
On Jul 4, 2:52 am, Neal <nealcr...@gmail.com> wrote:

> I've just started working on programming models for many-core

many-core = MIMD and not SIMD

> machines, and was wondering what the state of the art in terms of

the state of the art is somewhat dated.

> parallelization of serial code is. I'm most interested in SIMD
> vectorization right now... does anyone have a point to articles,
> websites, research etc on this topic? Thanks!

http://portal.acm.org/citation.cfm?id=30396

this book describes how one can create a hypothetical architecture of
processors -optimized for certain algorithms eg kalman filters/fast
fourier transforms etc..
 you can use the same thing in reverse fashion to decompose a program
into fragments that execute on different processors in an SIMD
machine.

regards
-kamal


    Reply to author    Forward  
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
End of messages
« Back to Discussions « Newer topic     Older topic »

Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
©2008 Google