- Mpi-21 - mailman3.cels.anl.gov

Re: [Mpi-21] A couple of bugs in 2.1
by unknown＠example.com 23 Jul '08

23 Jul '08

Rajeev, Since this is a regression from MPI 2.0, I think we need to fix this. I do not see a reason why we should not be able to fix these errors before the final vote. Is there someone that disagrees with this ? Rich On 7/21/08 10:18 PM, "Rajeev Thakur" <thakur_at_[hidden]> wrote: > Sent it to the wrong list earlier... > > -----Original Message----- > From: owner-mpi-21_at_[hidden] [mailto:owner-mpi-21_at_[hidden]] On > Behalf Of Rajeev Thakur > Sent: Monday, July 21, 2008 9:12 PM > To: mpi-21_at_[hidden] > Subject: [mpi-21] A couple of bugs in 2.1 > > I found a couple of bugs in MPI 2.1: > > * pg 332, ln 16-21: "MPI_REPLACE, like the other predefined operations, is > defined only for the predefined MPI datatypes. Rationale: The rationale for > this is that, for consistency, MPI_REPLACE should have the same limitations > as the other operations. Extending it to all datatypes doesn't provide any > real benefit." > > > That is not true. Derived datatypes are allowed in MPI_Accumulate for the > predefined reduction operations as explained in the paragraph just above it. > > > * pg 237, line 1 and line 3: MPI_MAX_OBJECT should be MPI_MAX_OBJECT_NAME. > > Accordingly, the Change Log needs to be updated for both these items. > > The question is can these be fixed before the 2nd vote in Sept.? My > understanding of the voting process is that the reason for having two > separate votes in two separate meetings is to have a chance to catch such > bugs. So it should be fixable. (Otherwise, what is the use of the 2nd vote?) > > Rajeev > > > _______________________________________________ > mpi-21 mailing list > mpi-21_at_[hidden] > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-21 > *

1 0

A couple of bugs in 2.1
by unknown＠example.com 21 Jul '08

21 Jul '08

Sent it to the wrong list earlier... -----Original Message----- From: owner-mpi-21_at_[hidden] [mailto:owner-mpi-21_at_[hidden]] On Behalf Of Rajeev Thakur Sent: Monday, July 21, 2008 9:12 PM To: mpi-21_at_[hidden] Subject: [mpi-21] A couple of bugs in 2.1 I found a couple of bugs in MPI 2.1: * pg 332, ln 16-21: "MPI_REPLACE, like the other predefined operations, is defined only for the predefined MPI datatypes. Rationale: The rationale for this is that, for consistency, MPI_REPLACE should have the same limitations as the other operations. Extending it to all datatypes doesn't provide any real benefit." That is not true. Derived datatypes are allowed in MPI_Accumulate for the predefined reduction operations as explained in the paragraph just above it. * pg 237, line 1 and line 3: MPI_MAX_OBJECT should be MPI_MAX_OBJECT_NAME. Accordingly, the Change Log needs to be updated for both these items. The question is can these be fixed before the 2nd vote in Sept.? My understanding of the voting process is that the reason for having two separate votes in two separate meetings is to have a chance to catch such bugs. So it should be fixable. (Otherwise, what is the use of the 2nd vote?) Rajeev

1 0

Re: [Mpi-21] Errata - Re: Updated slides from June meeting / with URL
by unknown＠example.com 09 Jul '08

09 Jul '08

Done. On Jul 9, 2008, at 3:29 AM, Rolf Rabenseifner wrote: > Jeff and all, > > I corrected the error on the "implementation-estimates" slide > (slide no. 8) in > > http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf > and > http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview_4to1.p… > > I always asked at the meetings about the implemementation > of the 33 MPI-2.1 issues. I explicitely stated that I do > not ask about implementation of the MPI-2.0 Dynamic Spawn chapter. > > I corrected the text on the slide to reflect this discussions > on the meetings. > > Jeff, please can you copy my both .pdf to your protocol page. > > Thanks and best regards > Rolf > > > On Thu, 03 Jul 2008 18:31:04 +0200 > "Rolf Rabenseifner" <rabenseifner_at_[hidden]> wrote: >> Jeff and all, >> I've updated my slides of June meeting at (user and pw is mpi21) >> http://www.hlrs.de/mpi/mpi21/doc/ >> See section "Protocol of June 30 - July 2, 2008 Meeting". >> Jeff, please can you copy them to the meetings web: >> http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf >> http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview_4to1.p… >> Best regards >> Rolf > > Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] > High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 > University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 > Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner > Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30) > _______________________________________________ > mpi-21 mailing list > mpi-21_at_[hidden] > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-21 -- Jeff Squyres Cisco Systems

1 0

Errata - Re: Updated slides from June meeting / with URL
by unknown＠example.com 09 Jul '08

09 Jul '08

Jeff and all, I corrected the error on the "implementation-estimates" slide (slide no. 8) in http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf and http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview_4to1.p… I always asked at the meetings about the implemementation of the 33 MPI-2.1 issues. I explicitely stated that I do not ask about implementation of the MPI-2.0 Dynamic Spawn chapter. I corrected the text on the slide to reflect this discussions on the meetings. Jeff, please can you copy my both .pdf to your protocol page. Thanks and best regards Rolf On Thu, 03 Jul 2008 18:31:04 +0200 "Rolf Rabenseifner" <rabenseifner_at_[hidden]> wrote: > Jeff and all, > > I've updated my slides of June meeting at (user and pw is mpi21) > > http://www.hlrs.de/mpi/mpi21/doc/ > > > See section "Protocol of June 30 - July 2, 2008 Meeting". > > Jeff, please can you copy them to the meetings web: > > http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf > > http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview_4to1.p… > > > Best regards > Rolf Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)

1 0

Re: [Mpi-21] Updated slides from June meeting / with URL
by unknown＠example.com 03 Jul '08

03 Jul '08

On Thu, 03 Jul 2008 18:12:33 +0200 "Rolf Rabenseifner" <rabenseifner_at_[hidden]> wrote: > Jeff and all, > > I've updated my slides of June meeting at (user and pw is mpi21) http://www.hlrs.de/mpi/mpi21/doc/ > > See section "Protocol of June 30 - July 2, 2008 Meeting". > > Jeff, please can you copy them to the meetings web: http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview_4to1.p… > > Best regards > Rolf > > > Dr. Rolf Rabenseifner . . . . . . . . . .. email >rabenseifner_at_[hidden] > High Performance Computing Center (HLRS) . phone >++49(0)711/685-65530 > University of Stuttgart . . . . . . . . .. fax ++49(0)711 / >685-65832 > Head of Dpmt Parallel Computing . . . >www.hlrs.de/people/rabenseifner > Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30) > _______________________________________________ > mpi-21 mailing list > mpi-21_at_[hidden] > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-21 Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)

1 0

Updated slides from June meeting
by unknown＠example.com 03 Jul '08

03 Jul '08

Jeff and all, I've updated my slides of June meeting at See section "Protocol of June 30 - July 2, 2008 Meeting". Jeff, please can you copy them to the meetings web: mpi_21_at_MPIforum_2008-06_overview.pdf mpi_21_at_MPIforum_2008-06_overview_4to1.pdf Best regards Rolf Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)

1 0

Re: [Mpi-21] Implementation of MPI-2.1
by unknown＠example.com 26 Jun '08

26 Jun '08

Hi everybody, I'm going to miss this meeting, so let me use this reply as a chance to express my opinion ahead of the vote: - We need to produce official versions of the MPI-1.3 and MPI-2.1 documents with change mark-up, either in change bars or in color coding. At the moment, as far as I understand, only the official documents are going to be published, and they won't have clear indication, apart from the MPI-2.1 change log, of what was actually changed in the text. Many standard bodies I know routinely publish marked-up versions of their new documents to facilitate smooth transition. If we don't follow this practice, we risk implementors overlooking some small but important changes, and then claiming MPI-1.3 and MPI-2.1 compliance by mistake. - The desire to have only Forum members edit the standards is a good one, but it may not be practical. In the particular case of the language binding Annex, we'd not be able to complete the massive chances on time without delegating this to someone else. This would make the standard process slip, which was considered a bigger risk at the time. What is necessary here is rather more careful supervision and review of the results, which I think we should take into account going forward. - If there are any material changes needed to Intel MPI implementation to make it MPI-1.3 and MPI-2.1 compliant, we'll target 2009. If we can claim MPI-1.3 and MPI-2.1 compliance in the current code without noticeable changes, either to the code or to the current draft during the pending voting, we may decide to go for this year's fall release. Best regards. Alexander -----Original Message----- From: mpi-21-bounces_at_[hidden] [mailto:mpi-21-bounces_at_[hidden]] On Behalf Of Rolf Rabenseifner Sent: Thursday, June 26, 2008 3:29 PM To: MPI 2.1 Mailing List Subject: [Mpi-21] Implementation of MPI-2.1 Dear MPI-implementor, if you are not yet on the list at slide 8 in http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf or if you are on the list and you have not yet given an estimate when your MPI-2.1 implemenatation will be available, then please give us this estimate at June meeting in San Francisco (or with a short reply to this list before the meeting). Thanks and best regards Rolf Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30) _______________________________________________ mpi-21 mailing list mpi-21_at_[hidden] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-21 --------------------------------------------------------------------- Intel GmbH Dornacher Strasse 1 85622 Feldkirchen/Muenchen Germany Sitz der Gesellschaft: Feldkirchen bei Muenchen Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer Registergericht: Muenchen HRB 47456 Ust.-IdNr. VAT Registration No.: DE129385895 Citibank Frankfurt (BLZ 502 109 00) 600119052 This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.

1 0

Implementation of MPI-2.1
by unknown＠example.com 26 Jun '08

26 Jun '08

Dear MPI-implementor, if you are not yet on the list at slide 8 in http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf or if you are on the list and you have not yet given an estimate when your MPI-2.1 implemenatation will be available, then please give us this estimate at June meeting in San Francisco (or with a short reply to this list before the meeting). Thanks and best regards Rolf Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)

1 0

Preparation of MPI-2.1 at June 2008 meeting
by unknown＠example.com 26 Jun '08

26 Jun '08

Hi all, all files for preparing the MPI-2.1 / MPI-1.3 sessions at June 2008 MPI Forum meeting in San Francisco are done. Files are available at (username and pw are mpi21) http://www.hlrs.de/mpi/mpi21/doc/ Only eleven Slides: ------------------- http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview.pdf or print-version: http://www.hlrs.de/mpi/mpi21/doc/mpi_21_at_MPIforum_2008-06_overview_4to1.p… On the web page you will find the final documents of the MPI-2.1 and MPI-1.3 standards - waiting for the first votes. And you find all issues that will be handed over to MPI-2.2: http://www.hlrs.de/mpi/mpi21/doc/MPI-2.1-reviews_open-issues.txt or print-version: http://www.hlrs.de/mpi/mpi21/doc/MPI-2.1-reviews_open-issues.pdf Best regards Rolf Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)

1 0

Fwd: Final MPI-2.1 (June 23, 2008) is available on the web
by unknown＠example.com 23 Jun '08

23 Jun '08

It's hopefully done. --- the forwarded message follows --- <strong>attached mail follows:</strong> Hi all, the final MPI-2.1 (June 23, 2008) is available via http://www.hlrs.de/mpi/mpi21/doc/ (username and pw is mpi21) This is the final version. All further corrections will go to MPI-2.2. Thank you to you all for all your work you put into this common project! Best regards Rolf PS: All open issues that are handed over from MPI-2.1 to MPI-2.2 can found at http://www.hlrs.de/mpi/mpi21/doc/MPI-2.1-reviews_open-issues.txt and http://www.hlrs.de/mpi/mpi21/doc/MPI-2.1-reviews_open-issues.pdf Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)

1 0