From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Squyres Subject: Re: [ofa-general] Re: [GIT PULL] please pull ummunotify Date: Tue, 15 Sep 2009 08:38:08 -0400 Message-ID: <003720AA-6630-4718-B7F9-8392A8422220@cisco.com> References: <20090911064019.GZ4973@obsidianresearch.com> <20090915155231.DB86.A69D9226@jp.fujitsu.com> Mime-Version: 1.0 (Apple Message framework v936) Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20090915155231.DB86.A69D9226@jp.fujitsu.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: general-bounces@lists.openfabrics.org Errors-To: general-bounces@lists.openfabrics.org To: KOSAKI Motohiro Cc: linux-rdma@vger.kernel.org, "Roland Dreier (rdreier)" , linux-kernel@vger.kernel.org, general@lists.openfabrics.org, Brice Goglin , akpm@linux-foundation.org, torvalds@linux-foundation.org List-Id: linux-rdma@vger.kernel.org On Sep 15, 2009, at 3:03 AM, KOSAKI Motohiro wrote: > - I guess you have your MPI implementaion w/ ummunotify, right? > - I guess you have test sevaral pattern, right? > if so, can we see your test result? > Roland's answers to the rest of these questions were spot-on, so I thought I'd just throw in a quick reply to the above questions: yes, we have a prototype Open MPI implementation with code that uses ummunotify (http://bitbucket.org/jsquyres/ummunot/). I just finished fixing a high-priority (but unrelated) bug in Open MPI, so merging the prototype ummunotify code into the upstream Open MPI repository is now at the top of my priority list. We have done quite a bit of testing with ummunotify, but since the code is not yet in the Open MPI mainline, most of the testing has been manual (not through our automated testing system). As far as we can tell, everything is working properly with Open MPI + ummunotify. We also anticipate that other MPI implementations will be able to use ummunotify, potentially using Open MPI as a reference ummunotify implementation. FWIW: we went through a bunch of design and implementation iterations with Roland to get code that everyone was happy with: - Roland likes it (and anticipated that the kernel community would be receptive to) - we like it - performs correctly Hope that helps. -- Jeff Squyres jsquyres@cisco.com From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753485AbZIOMiO (ORCPT ); Tue, 15 Sep 2009 08:38:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753165AbZIOMiJ (ORCPT ); Tue, 15 Sep 2009 08:38:09 -0400 Received: from rtp-iport-1.cisco.com ([64.102.122.148]:26766 "EHLO rtp-iport-1.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752668AbZIOMiI (ORCPT ); Tue, 15 Sep 2009 08:38:08 -0400 X-IronPort-AV: E=Sophos;i="4.44,390,1249257600"; d="scan'208";a="58174618" Cc: "Roland Dreier (rdreier)" , "Jason Gunthorpe" , "Brice Goglin" , , , , , Message-Id: <003720AA-6630-4718-B7F9-8392A8422220@cisco.com> From: Jeff Squyres To: KOSAKI Motohiro In-Reply-To: <20090915155231.DB86.A69D9226@jp.fujitsu.com> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v936) Subject: Re: [ofa-general] Re: [GIT PULL] please pull ummunotify Date: Tue, 15 Sep 2009 08:38:08 -0400 References: <20090911064019.GZ4973@obsidianresearch.com> <20090915155231.DB86.A69D9226@jp.fujitsu.com> X-Mailer: Apple Mail (2.936) Authentication-Results: rtp-dkim-1; header.From=jsquyres@cisco.com; dkim=pass ( sig from cisco.com/rtpdkim1001 verified; ); Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sep 15, 2009, at 3:03 AM, KOSAKI Motohiro wrote: > - I guess you have your MPI implementaion w/ ummunotify, right? > - I guess you have test sevaral pattern, right? > if so, can we see your test result? > Roland's answers to the rest of these questions were spot-on, so I thought I'd just throw in a quick reply to the above questions: yes, we have a prototype Open MPI implementation with code that uses ummunotify (http://bitbucket.org/jsquyres/ummunot/). I just finished fixing a high-priority (but unrelated) bug in Open MPI, so merging the prototype ummunotify code into the upstream Open MPI repository is now at the top of my priority list. We have done quite a bit of testing with ummunotify, but since the code is not yet in the Open MPI mainline, most of the testing has been manual (not through our automated testing system). As far as we can tell, everything is working properly with Open MPI + ummunotify. We also anticipate that other MPI implementations will be able to use ummunotify, potentially using Open MPI as a reference ummunotify implementation. FWIW: we went through a bunch of design and implementation iterations with Roland to get code that everyone was happy with: - Roland likes it (and anticipated that the kernel community would be receptive to) - we like it - performs correctly Hope that helps. -- Jeff Squyres jsquyres@cisco.com