September 2020 - mpi-forum - mailman3.cels.anl.gov

Announcing second vote for issue #137 - The Embiggenment ("BigCount")
by Bangalore, Purushotham 14 Sep '20

14 Sep '20

Hi All, Here are the details related to this announcement: Issue: https://github.com/mpi-forum/mpi-issues/issues/137 PR: https://github.com/mpi-forum/mpi-standard/pull/268 PDF: mpi40-report-ticket137-cfed9a5-14Sep2020.pdf<https://github.com/mpi-forum/mpi-issues/files/5219835/mpi40-report-ticket13…> Highlighted PDF that shows the changes since first vote: mpi40-report-ticket137-cfed9a5-14Sep2020-highlighted.pdf<https://github.com/mpi-forum/mpi-issues/files/5219837/mpi40-report-ticket13…> We would like to propose a no-no-vote on the changes made since the first vote. There was some concern that there are a lot of changes based on the number of files changes in the PR. I wanted to emphasize that we had to create a new pull request to address the merge conflicts that were created after several new chapters and sections were merged. As a result, the new PR includes several changes that were already made in the original PR. The actual differences between the two versions are not that many and most of these changes are editorial. I have listed the explicit changes made since the first vote below (these changes are also highlighted in the PDF file above): 0. the change bar latex macros have been removed 1. change all new function extensions from _l to _c and the text " - an underscore followed by a lower-case L " has been removed, where used 2. addition of the following sentence in Section 2.5.8: "Even though the \type{MPI\_Count} type is large enough to encode address locations, \type{MPI\_Count} type shall not be used to represent an \mpitermni{absolute address}\mpitermindex{addresses!absolute}." 3. delete the following sentence that was repeated: "The support for large count and displacement in Fortran is available when using newer \MPI/ Fortran bindings \code{(USE mpi_f08)}." 4. embiggen MPI_TYPE_GET_ENVELOPE and MPI_TYPE_GET_CONTENTS 5. change error class from MPI_ERR_VALUE_TOO_LARGE to MPI_ERR_TYPE for functions MPI_TYPE_GET_ENVELOPE and MPI_TYPE_GET_CONTENTS 6. addition of separate MPI_Op_create_c and MPI_Register_datarep_c functions for F08 bindings 7. following text changes in Section 18.2: (a) replace "The following were used prior to MPI-4.0:" with "The following types, which were used prior to MPI-4.0, have been deemed too small to hold values that applications wish to use:" and remove the sentence "These types have been deemed too small to hold values that applications wish to use." in the next paragraph (b) addition of "(e.g., in constructors of MPI datatypes that can be used with files)" to the end of third bullet (c) addition of the following new text: For the large count versions of three datatype constructors, MPI_TYPE_CREATE_HINDEXED, MPI_TYPE_CREATE_HINDEXED_BLOCK, and MPI_TYPE_CREATE_STRUCT, absolute addresses shall not be used to specify byte displacements since the parameter is of type MPI_COUNT instead of type MPI_AINT (see Section 2.5.8). (d) addition of "MPI_Offset for parameters that represent byte displacement in files, " in the C bindings description (e) addition of "INTEGER(KIND=MPI_OFFSET_KIND) for parameters that represent byte displacement in files, " in the Fortran bindings description ====== Thanks, Puri on behalf of the large counts WG

1 0

Announcing the reading of the standalone semantics terms summary document
by Bangalore, Purushotham 13 Sep '20

13 Sep '20

Hi All, Here are the details related to the reading of the standalone document for the summary of the semantic terms: Issue: https://github.com/mpi-forum/mpi-issues/issues/304 PR: https://github.com/mpi-forum/mpi-standard/pull/279 PDF: addendum-Semantics.pdf<https://github.com/mpi-forum/mpi-issues/files/5215968/addendum-Semantics.pdf> Thanks, Puri

1 0

Announcing the second vote for Issue #96
by Bangalore, Purushotham 13 Sep '20

13 Sep '20

Hi All, Here is the announcement for the second vote for issue #96. Issue: https://github.com/mpi-forum/mpi-issues/issues/96 PR: https://github.com/mpi-forum/mpi-standard/pull/116 PDF file of the first vote: mpi40-report-ticket96-5ccf3e4-18Aug2020.pdf<https://github.com/mpi-forum/mpi-standard/files/5093050/mpi40-report-ticket…> Thanks, Puri

1 0

Announcing Tools WG Errata
by Marc-André Hermanns 13 Sep '20

13 Sep '20

Dear all, the tools working group would like to read the following errata: Changes to MPI_T Events if info objects were available before MPI initialization: Issue: https://github.com/mpi-forum/mpi-issues/issues/148 PR: https://github.com/mpi-forum/mpi-standard/pull/274 Clarify when to return info objects in MPI_T events chapter Issue: https://github.com/mpi-forum/mpi-issues/issues/309 PR: https://github.com/mpi-forum/mpi-standard/pull/277 Cheers, Marc-Andre -- Dr. rer. nat. Marc-André Hermanns IT Center Group: High Performance Computing Division: Computational Science and Engineering RWTH Aachen University Seffenter Weg 23 52074 Aachen Phone: +49 241 80-24381 hermanns(a)itc.rwth-aachen.de www.itc.rwth-aachen.de

1 0

Announcing Erratum (Ticket #301) for Reading for the September Meeting
by Guillaume Mercier 13 Sep '20

13 Sep '20

Dear all, The Hardware Topologies Working Group would like to read the following erratum at the next meeting in September, dealing with MPI_Comm_split_type and MPI_COMM_TYPE_SHARED: Issue: https://github.com/mpi-forum/mpi-issues/issues/301 PR: https://github.com/mpi-forum/mpi-standard/pull/263 The pdf is available on the Issue page: https://github.com/mpi-forum/mpi-issues/files/5164743/mpi40-report-issue301… Thanks, Guillaume

1 0

Re: [Mpi-forum] Announcing Erratum (Ticket #308) for Reading for September meeting -- Add MPI_INFO argument to MPI_PSEND_INIT and MPI_PRECV_INIT
by Julien.JAEGER＠CEA.FR 13 Sep '20

13 Sep '20

Hi all, I made some suggestions in the PR source codes. In the added text for the info key argument, there are references to MPI_RECV_INIT and MPI_SEND_INIT, where it should be MPI_PRECV_INIT and MPI_PSEND_INIT. example: " Also unlike MPI_RECV_INIT, MPI_RECV_INIT takes an info argument. " should be "Also unlike MPI_RECV_INIT, MPI_PRECV_INIT takes an info argument." Also, I just spotted a minor typo: p90 l.9 in the send pdf: " for partitioned communication: They provide MPI_PSEND_INIT and MPI_PRECV_INIT". The "T" of "They" shouldn't be a capital letter. Cheers, Julien J. ________________________________ De : mpi-forum <mpi-forum-bounces(a)lists.mpi-forum.org> de la part de Skjellum, Anthony via mpi-forum <mpi-forum(a)lists.mpi-forum.org> Envoyé : samedi 12 septembre 2020 21:11 À : Main MPI Forum mailing list Cc : Skjellum, Anthony Objet : [Mpi-forum] Announcing Erratum (Ticket #308) for Reading for September meeting -- Add MPI_INFO argument to MPI_PSEND_INIT and MPI_PRECV_INIT Dear colleagues: Please see: Issue: https://github.com/mpi-forum/mpi-issues/issues/308 PR: https://github.com/mpi-forum/mpi-standard/pull/275 [When the Partitioned chapter is merged into the mpi-4 branch, this will be clean; till then, Dan made a PR local to our WG, as follows: https://github.com/mpiwg-persistence/mpi-standard/pull/3 (WG access required)] Latest PDF: https://github.com/mpi-forum/mpi-issues/files/5213130/mpi40-report-ticket30… Thank you, Tony Skjellum, Ryan Grant, Puri Bangalore Anthony Skjellum, PhD Professor of Computer Science and Chair of Excellence Director, SimCenter University of Tennessee at Chattanooga (UTC) tony-skjellum(a)utc.edu [or skjellum(a)gmail.com] cell: 205-807-4968

1 0

Announcing Erratum (Ticket #308) for Reading for September meeting -- Add MPI_INFO argument to MPI_PSEND_INIT and MPI_PRECV_INIT
by Skjellum, Anthony 12 Sep '20

12 Sep '20

Dear colleagues: Please see: Issue: https://github.com/mpi-forum/mpi-issues/issues/308 PR: https://github.com/mpi-forum/mpi-standard/pull/275 [When the Partitioned chapter is merged into the mpi-4 branch, this will be clean; till then, Dan made a PR local to our WG, as follows: https://github.com/mpiwg-persistence/mpi-standard/pull/3 (WG access required)] Latest PDF: https://github.com/mpi-forum/mpi-issues/files/5213130/mpi40-report-ticket30… Thank you, Tony Skjellum, Ryan Grant, Puri Bangalore Anthony Skjellum, PhD Professor of Computer Science and Chair of Excellence Director, SimCenter University of Tennessee at Chattanooga (UTC) tony-skjellum(a)utc.edu [or skjellum(a)gmail.com] cell: 205-807-4968

1 0

Errata reading/vote for MPI Sessions
by Pritchard Jr., Howard 10 Sep '20

10 Sep '20

HI Folks, The Sessions WG would like to read an errata for MPI Sessions and object isolation at the upcoming voting MPI Forum. Title: Clarify MPI object isolation description in Chapter 10.3 Issue: https://github.com/mpi-forum/mpi-issues/issues/310 PR: https://github.com/mpi-forum/mpi-standard/pull/269 Thanks, Howard -- Howard Pritchard HPC-ENV Los Alamos National Laboratory

1 0

Re: [Mpi-forum] Missed Pythonization routines
by Jeff Squyres (jsquyres) 10 Sep '20

10 Sep '20

On Sep 10, 2020, at 10:22 AM, William Gropp <wgropp(a)illinois.edu> wrote: > > I think this is an editorial fix, though with chapter committee review. Sounds good. I already review/approved -- I'll pass it on to the bindings committee, too. -- Jeff Squyres jsquyres(a)cisco.com

1 0

Re: [Mpi-forum] Missed Pythonization routines
by William Gropp 10 Sep '20

10 Sep '20

I think this is an editorial fix, though with chapter committee review. Bill William Gropp Director and Chief Scientist, NCSA Thomas M. Siebel Chair in Computer Science University of Illinois Urbana-Champaign > On Sep 10, 2020, at 9:17 AM, Jeff Squyres (jsquyres) via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > A handful of routines in the bindings chapter accidentally escaped Pythonization. > > Martin R. just Pythonized them: https://github.com/mpi-forum/mpi-standard/pull/276 > > Does this qualify as an errata? (i.e., does it need a reading+vote) > Or is it an editorial fix? > Or is is a chapter committee change? > Or ...? > > -- > Jeff Squyres > jsquyres(a)cisco.com > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpi-forum

1 0