From owner-sc22wg5@open-std.org  Thu Jan 22 18:57:30 2009
Return-Path: <owner-sc22wg5@open-std.org>
X-Original-To: sc22wg5-dom7
Delivered-To: sc22wg5-dom7@www2.open-std.org
Received: by www2.open-std.org (Postfix, from userid 521)
	id 67CC8CA5FE6; Thu, 22 Jan 2009 18:57:30 +0100 (CET)
X-Original-To: sc22wg5@open-std.org
Delivered-To: sc22wg5@open-std.org
Received: from nspiron-1.llnl.gov (nspiron-1.llnl.gov [128.115.41.81])
	by www2.open-std.org (Postfix) with ESMTP id 9AE96CA3439
	for <sc22wg5@open-std.org>; Thu, 22 Jan 2009 18:57:28 +0100 (CET)
X-Attachments: None
Received: from cyrus2.llnl.gov ([128.15.97.105])
  by nspiron-1.llnl.gov with ESMTP; 22 Jan 2009 09:57:27 -0800
From: Aleksandar Donev <donev1@llnl.gov>
Organization: LLNL
To: mpi3-fortran@lists.mpi-forum.org
Subject: Re: [MPI3 Fortran] (j3.2006) (SC22WG5.3886) =?iso-8859-1?q?=5Bukfortran=5D=09=5BMPI3=09Fortran=5D=09MPI=09non-blocking?=
 =?iso-8859-1?q?_transfers?=
Date: Thu, 22 Jan 2009 09:57:25 -0800
User-Agent: KMail/1.9.4
Cc: fortran standards email list for J3 <j3@j3-fortran.org>,
	WG5 <sc22wg5@open-std.org>
References: <Prayer.1.3.1.0901211104060.5654@hermes-2.csi.cam.ac.uk> <20090122100652.C31E9CA3434@www2.open-std.org> <4978A8A0.8090407@cray.com>
In-Reply-To: <4978A8A0.8090407@cray.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200901220957.25811.donev1@llnl.gov>
Sender: owner-sc22wg5@open-std.org
Precedence: bulk

On Thursday 22 January 2009 09:10, Bill Long wrote:

> If we use a different name, like
> ASYNC_EXTERNAL, is there any difference between that and VOLATILE? =A0I
> suspect not.
Yes, I explained this a long time ago, many times, but we go in circles=20
again.
=46or one thing, VOLATILE has some ill-defined semantics.
=46or two, VOLATILE disables all optimization. ASYNCHRONOUS only disables=20
code motion accross waits and certain argument associations (such as=20
more care with copy in/out). The rest of the code can be optimized as=20
usual. Surely this is important for a high-performance library like=20
MPI.

> The other is that it incorrectly suggests that MPI calls
> have something to do with I/O which is not the case (except for the
> actual MPI I/O calls
And I already explained this a long time ago, and even quoted you the=20
Wikipedia definition of I/O. But if you care not to listen, that is=20
fine, let's continue going in circles.

Best,
Aleks