From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-93380-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 30047 invoked by alias); 25 Mar 2004 12:37:31 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 30013 invoked from network); 25 Mar 2004 12:37:29 -0000
Received: from unknown (HELO nile.gnat.com) (205.232.38.5)
  by sources.redhat.com with SMTP; 25 Mar 2004 12:37:29 -0000
Received: from localhost (localhost [127.0.0.1])
	by nile.gnat.com (Postfix) with ESMTP
	id E7F04F2CB5; Thu, 25 Mar 2004 07:37:28 -0500 (EST)
Received: from nile.gnat.com ([127.0.0.1])
 by localhost (nile.gnat.com [127.0.0.1]) (amavisd-new, port 10024) with LMTP
 id 23120-01-8; Thu, 25 Mar 2004 07:37:28 -0500 (EST)
Received: from gnat.com (hoosic.gnat.com [205.232.38.102])
	by nile.gnat.com (Postfix) with ESMTP
	id 8BCD8F281B; Thu, 25 Mar 2004 07:37:28 -0500 (EST)
Message-ID: <4062D287.7050100@gnat.com>
Date: Thu, 25 Mar 2004 14:45:00 -0000
From: Robert Dewar <dewar@gnat.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040113
MIME-Version: 1.0
To: Joost VandeVondele <jv244@hermes.cam.ac.uk>
Cc: gcc@gcc.gnu.org
Subject: Re: (a+b)+c should be replaced by a+(b+c)
References: <Pine.SOL.4.58.0403250711490.18567@yellow.csi.cam.ac.uk>
In-Reply-To: <Pine.SOL.4.58.0403250711490.18567@yellow.csi.cam.ac.uk>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: by amavisd-new at nile.gnat.com
X-SW-Source: 2004-03/txt/msg01461.txt.bz2

Joost VandeVondele wrote:

> good compilers (e.g. xlf90) will (at -O4) do higher order transforms of
> the loop to introduce blocking, independent FMAs, ... that makes this
> little piece of code about 100 times faster at O4 than O2 (what about
> LNO/SSA?). This can only be done if you allow (a+b)+c -> a+(b+c). It is
> basically what any optimized blas routine will do. Matrix multiply is a
> trivial example, if you want blas performance, call blas. There are many
> other kernels like this in e.g. scientific code that are not blas. You
> can't expect a scientist to hand unroll and block any kernel to the
> appropriate depth for any machine. There need to be a compiler option to
> do this. This can only be done if you allow (a+b)+c -> a+(b+c).

Can you really deduce this freedom from later versions of the Fortran
standard?