December 2022 - mpi-forum - mailman3.cels.anl.gov

No MPI Meeting today - Happy Holidays and Happy New Year
by Martin Schulz 21 Dec '22

21 Dec '22

Hi all, We have no virtual MPI forum meeting today – our next meeting is planned for January 11th. This leaves me just with wanting to wish you all happy holidays, hopefully some quiet and peaceful time, and a happy start into the new year. Thanks for all your hard work in the forum and then on to MPI 4.1 next year, which should make it an exciting year. Thanks and until 2023! Martin -- Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de

1 0

Virtual MPI Forum Meeting Tomorrow (Dec. 14th, 10am US CT)
by Martin Schulz 13 Dec '22

13 Dec '22

Hi all, As discussed last week, we will have one more MIP virtual forum meeting this year, tomorrow (Dec. 14th at 10am US CT) – the access link is, as always, available via the MPI Forum website. Jim will talk about “Allocator kind info”. The meetings for next year (voting and non-voting), which we decided on on Friday, are also listed on the forum website. Thanks, Martin -- Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de

1 0

ABI working group is live
by Jeff Hammond 09 Dec '22

09 Dec '22

The MPI ABI workgroup was approved. You can sign up for the mailing list now: https://lists.mpi-forum.org/mailman/listinfo/mpiwg-abi. If you want access to the GitHub group (https://github.com/mpiwg-abi/) send me your GitHub username. If you want access to the Slack channel, DM me on Slack. If you are not on Slack already, send me your email address. Jeff On Wed, Nov 16, 2022 at 9:54 AM Jeff Hammond <jeff.science(a)gmail.com> wrote: > I don't know what we do to create new working groups with the post-COVID > rules, but I would like to create and chair a WG focused on ABI > standardization. > > There is strong support for this effort in many user communities, > including developers and maintainers of Spack, mpi4py, Julia MPI (MPI.jl), > Rust MPI (rsmpi), PETSc and NVHPC SDK, to name a few. There are even a few > implementers who have expressed support, but I won't name them for their > own protection. > > The problem is so exasperating for our users that there are at least two > different projects devoted to mitigating ABI problems (not including shims > built in to the aforementioned MPI wrappers): > > https://github.com/cea-hpc/wi4mpi > https://github.com/eschnett/MPItrampoline > > I've written about this a bit already, for those who are interested. More > material will be forthcoming once I have time for more experiments. > > https://github.com/jeffhammond/blog/blob/main/MPI_Needs_ABI.md > https://github.com/jeffhammond/blog/blob/main/MPI_Needs_ABI_Part_2.md > https://github.com/jeffhammond/blog/blob/main/MPI_Needs_ABI_Part_3.md > https://github.com/jeffhammond/blog/blob/main/MPI_Needs_ABI_Part_4.md > > I understand this is a controversial topic, particularly for > implementers. I hope that we can proceed objectively. > > Thanks, > > Jeff > > -- > Jeff Hammond > jeff.science(a)gmail.com > http://jeffhammond.github.io/ > -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

Re: [Mpi-forum] no-no-votes for the readings yesterday - Re: December Meeting Draft Agenda Posted
by Wes Bland 07 Dec '22

07 Dec '22

Thanks Rolf, After talking to Martin last night, I think it makes the most sense to vote on the errata, first, and second ballots today. Then we’ll do the no-no votes tomorrow. I’m sure we’ll add at least one more no-no based on the discussion today. You can find the full list of ballots on the voting page: https://www.mpi-forum.org/meetings/2022/12/votes Thanks, Wes > On Dec 7, 2022, at 6:33 AM, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > Dear Wesley and Martin, > > the following small no-no-vote changes for/from the readings yesterday should be read and voted: > > > Issue 493 and PR 753 (on new progress section): > > https://github.com/mpi-forum/mpi-standard/pull/753/files/dc50cc9c6b40eb7328… > > > Issue 641 and PR 752 (on Annex A.2): > > https://github.com/mpi-forum/mpi-standard/pull/752/files/9141595df073078c60… > and > only inverting the sequence Wait / test(true) / test(false): > https://github.com/mpi-forum/mpi-standard/pull/752/files/af2b14e7b3c61541eb… > and > https://github.com/mpi-forum/mpi-standard/files/10175926/mpi41-report_Issue… > > > Issue 605 and PR 746 (Titles for topology examples) > > https://github.com/mpi-forum/mpi-standard/pull/746/commits/74936f003a4cb38f… > > > With those, I expect that my readings are done. > > > Best reagrds > Rolf > > > -- > Dr. Rolf Rabenseifner . . . . . . . . . .. . . rabenseifner(a)hlrs.de . > High Performance Computing Center (HLRS) . . . ++49(0)711/685-65530 . > University of Stuttgart . . . . . . www.hlrs.de/people/rabenseifner . > Nobelstr. 19, 70569 Stuttgart, Germany > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpi-forum

1 0

no-no-votes for the readings yesterday - Re: December Meeting Draft Agenda Posted
by Rolf Rabenseifner 07 Dec '22

07 Dec '22

Dear Wesley and Martin, the following small no-no-vote changes for/from the readings yesterday should be read and voted: Issue 493 and PR 753 (on new progress section): https://github.com/mpi-forum/mpi-standard/pull/753/files/dc50cc9c6b40eb7328… Issue 641 and PR 752 (on Annex A.2): https://github.com/mpi-forum/mpi-standard/pull/752/files/9141595df073078c60… and only inverting the sequence Wait / test(true) / test(false): https://github.com/mpi-forum/mpi-standard/pull/752/files/af2b14e7b3c61541eb… and https://github.com/mpi-forum/mpi-standard/files/10175926/mpi41-report_Issue… Issue 605 and PR 746 (Titles for topology examples) https://github.com/mpi-forum/mpi-standard/pull/746/commits/74936f003a4cb38f… With those, I expect that my readings are done. Best reagrds Rolf -- Dr. Rolf Rabenseifner . . . . . . . . . .. . . rabenseifner(a)hlrs.de . High Performance Computing Center (HLRS) . . . ++49(0)711/685-65530 . University of Stuttgart . . . . . . www.hlrs.de/people/rabenseifner . Nobelstr. 19, 70569 Stuttgart, Germany

1 0

Tuesday, December 6 MPI Forum Voting
by Wes Bland 06 Dec '22

06 Dec '22

Hi all, Given the large number of readings and ballots taking place at this meeting, I wanted to keep folks up to date on what we’ll be voting on and when. As usual, we’ll separate the "no-no" ballots from the “main” ballots with which they are paired. So today, we will vote on all of the existing "no-no” items: #553 <https://github.com/mpi-forum/mpi-issues/issues/553>: RMA: Add missing win-sync calls to Example 12.21 #551 <https://github.com/mpi-forum/mpi-issues/issues/551>: MPI_Win_free is synchronizing, except when it isn't #407 <https://github.com/mpi-forum/mpi-issues/issues/407>: Adjust RMA Chapter to New Terms or vice versa (and remove interims statement) #429 <https://github.com/mpi-forum/mpi-issues/issues/429>: RMA: Replace 'operation' with 'operator' where appropriate #590 <https://github.com/mpi-forum/mpi-issues/issues/590>: Remove Physical Requirement from Voting Meetings #589 <https://github.com/mpi-forum/mpi-issues/issues/589>: Errata: Issues with Partitioned Communication Example-4.4 #634 <https://github.com/mpi-forum/mpi-issues/issues/634>: Fix MPI_WIN_SYNC text Note that there were a couple of late additions to this list based on the readings yesterday. Tomorrow we’ll probably vote on everything else on the voting page: https://www.mpi-forum.org/meetings/2022/12/votes This ballot will take place at approximately 10am US Central time and you’ll vote by going to the link that I’ll email out to you at that time. As usual, we might start a minute or two late depending on the discussion going on at that time and you’ll have 30 minutes to cast your ballot once I send out the email. Remember that you must attend the meeting prior to the first ballot in order to be eligible to vote. The list of who has registered and who has attended is here: https://www.mpi-forum.org/meetings/2022/12/attendance. If I didn’t see you on the list of folks that joined the Zoom call yesterday, you (or someone else from your org) need to join the call before 10am US Central today to be able to vote. Here’s the list of orgs that I think are eligible and who aren’t: Eligible to vote AMD Amazon Atos Cornelis Networks HLRS, University of Stuttgart HPE IBM INRIA Intel Corporation Kichakato Kizito Lawrence Livermore National Laboratory Leibniz Supercomputing Centre Los Alamos National Laboratory Meta NVIDIA Oak Ridge National Laboratory ParaTools SAS Queen's University RWTH Aachen University Technical University of Munich The University of Tokyo University of Alabama University of Illinois Urbana-Champaign University of Tennessee, Chattanooga University of Tennessee, Knoxville Université Grenoble Alpes VSC Research Center, TU Wien ZIH, TU Dresden Did not register for 2 of last 3 meetings Lawrence Berkeley National Laboratory Tennessee Tech University Not present at current meeting Argonne National Laboratory CEA Ohio State University Sandia National Laboratory The Texas Advanced Computing Center University of New Mexico Did not attend 2 of last 3 meetings Forschungszentrum Julich University of Basel If you have any questions, let me know via email or Slack (for a faster response). Thanks, Wes

1 0