From owner-sc22wg5+sc22wg5-dom8=www.open-std.org@open-std.org Tue May 14 21:15:17 2013 Return-Path: X-Original-To: sc22wg5-dom8 Delivered-To: sc22wg5-dom8@www.open-std.org Received: by www.open-std.org (Postfix, from userid 521) id C8C45356E85; Tue, 14 May 2013 21:15:17 +0200 (CEST) Delivered-To: sc22wg5@open-std.org X-Greylist: delayed 1031 seconds by postgrey-1.34 at www5.open-std.org; Tue, 14 May 2013 21:15:16 CEST Received: from ppsw-33.csi.cam.ac.uk (ppsw-33.csi.cam.ac.uk [131.111.8.133]) by www.open-std.org (Postfix) with ESMTP id F128B356929 for ; Tue, 14 May 2013 21:15:01 +0200 (CEST) X-Cam-AntiVirus: no malware found X-Cam-ScannerInfo: http://www.cam.ac.uk/cs/email/scanner/ Received: from hermes-2.csi.cam.ac.uk ([131.111.8.54]:43453) by ppsw-33.csi.cam.ac.uk (smtp.hermes.cam.ac.uk [131.111.8.157]:25) with esmtpa (EXTERNAL:nmm1) id 1UcKPz-00063x-gi (Exim 4.80_167-5a66dd3) (return-path ); Tue, 14 May 2013 19:57:47 +0100 Received: from prayer by hermes-2.csi.cam.ac.uk (hermes.cam.ac.uk) with local (PRAYER:nmm1) id 1UcKPz-0007L9-5z (Exim 4.72) (return-path ); Tue, 14 May 2013 19:57:47 +0100 Received: from [91.125.161.29] by webmail.hermes.cam.ac.uk with HTTP (Prayer-1.3.5); 14 May 2013 19:57:47 +0100 Date: 14 May 2013 19:57:47 +0100 From: "N.M. Maclaren" To: sc22wg5@open-std.org, Mark Batty Cc: Bill Long , "Lionel, Steve" , Lorri Menard , Daniel C Chen Subject: Existing support for uses of atomics in Fortran coarray codes Message-ID: X-Mailer: Prayer v1.3.5 Mime-Version: 1.0 Content-Type: text/plain; format=flowed; charset=ISO-8859-1 Sender: owner-sc22wg5@open-std.org Precedence: bulk Mark Batty, Peter Sewell and I had a discussion about atomic semantics and Fortran this morning, but there were a couple of things that were rather important and I didn't have a feel for the answers. Specifically, what semantics are provided by existing coarray atomics, and how they are used in real programs. We are definitely going to have to decide what to say about this at Delft, and the problem isn't simple :-( In particular: Do implementations guarantee coherence of access to a single atomic location and, if not, what do they guarantee? What, if anything, do they guarantee about the consistency of accesses to two different atomic locations? We know what POWER and x86 hardware guarantee, but have no idea of what (more? less?) is guaranteed in coarray implementations. Even with using MPI as a basis, it could be anything from sequential consistency to nothing, depending on the details of the implementation. And the fancy RDMA networks are another matter entirely! It would also be useful if there were some examples of how they are used for ordering (i.e. in combination with SYNC MEMORY) and running totals etc. Specifically, any use where the consistency semantics matter to the program. Any feedback appreciated. Thanks. Regards, Nick Maclaren.