August 2023 - mpi-forum - mailman3.cels.anl.gov

RMA WG Reading Announcement September'23 Meeting
by Joseph Schuchart 31 Aug '23

31 Aug '23

The RMA WG would like to read the following errata: RMA chapter committee changes Issue: https://github.com/mpi-forum/mpi-issues/issues/745 PR: https://github.com/mpi-forum/mpi-standard/pull/857 These changes were somehow missed for the 4.1rc and will likely address many of the issues reviewers will have found. Thanks Joseph

1 0

Reading announcement for ULFM slices 2, 3: AGREE and SHRINK
by Aurelien Bouteiller 31 Aug '23

31 Aug '23

The Error Handling and Fault TOlerance Working Group would like to read the following slices of the ULFM fine-grain fault tolerance ULFM Fault Tolerance (slice 2: agree) #582 https://github.com/mpi-forum/mpi-issues/issues/582 ULFM Fault Tolerance (slice 3: shrink) #583 https://github.com/mpi-forum/mpi-issues/issues/583 This work is slated for MPI-5 and follows upon the (already voted-in) ULFM slice1 Please refer to the diff of the latest commit when reviewing: only the last commit is relevant for each PR (the prior commits pertain to the prior slices). I will try to produce a latexdiff as soon as possible. Best Regards, Aurelien -- Aurelien Bouteiller, Ph.D. Innovative Computing Laboratory; The University of Tennessee; 1122 Volunteer Blvd.; Claxton suite 203; 37996, Knoxville, TN, USA +1 865 974 9308 (p); +1 865 974 8296 (f); Claxton 316 (f2f)

1 0

September MPI Forum Registration and Readings
by Wes Bland 30 Aug '23

30 Aug '23

Hi all, Quick reminder to register for the upcoming meeting. Remember that you need to register twice (I know it’s annoying…). First, register for the conference as an in-person or remote attendee: https://www.eventbrite.co.uk/e/iwomp-2023-eurompi-and-mpi-forum-tickets-666… Second, register for the MPI Forum meeting so I can count you for voting eligibility: https://forms.gle/EMHZgSq7nSJ7gFwk6 It seems like there’s a lot of folks who have registered for the conference, but not for the meeting. So please make sure you do both! Lastly, if you haven’t already looked at it, remember that if you’ve been participating in the forum over the last year or so, you probably have a review assignment for MPI 4.1. I sent those out a little bit ago with instructions so go find that email and make sure you’re completing those reviews ASAP so the chapter chairs can make any edits, if necessary. Thanks, Wes

1 0

reading announcement for pull 875: standard ABI
by Jeff Hammond 29 Aug '23

29 Aug '23

See https://github.com/mpi-forum/mpi-standard/pull/875. I expect to read this more than once but we have to start somewhere. This PR defines the overall structure and specifies many of the ABI features. Many of the choices have strong consensus in the working group. The constants are not listed in tables yet because that's a bunch of LaTeX tedium for which I have not yet had time. Nothing in this PR depends on the values of the constants, and thus they can be added in a second PR. There are open issues regarding Fortran that will be added later, possibly as a different PR, depending on the working group discussion later today. The current proposal assumes one choice of MPI_Aint, which is only relevant in the context of exotic platforms like CHERI that have wide pointers (pointers are bigger than addresses). Jeff -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

has anyone ever implemented MPI for a 16-bit OS or x86 real mode or DOS?
by Jeff Hammond 25 Aug '23

25 Aug '23

I am trying to figure if there is any justification for the continued mention of segmented addressing in the standard. Also, it is not clear to me that we have ever had any implementation experience that gives us the expertise to standardize features for it. x86 real mode support was dropped in Windows 3.1. Linux never supported it because it was created for 386. I cannot find any evidence that anyone has ever supported MPI on a 286, or any operating system that supported 16-bit real mode. https://en.wikipedia.org/wiki/Coherent_(operating_system) ran on 286 but didn't support TCP/IP so it's not clear how MPI would have achieved parallelism. https://en.wikipedia.org/wiki/Xenix was discontinued in 1987. What operating system are we using as the reference for reasoning about how MPI works on such systems? https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=06e491ccb3ce… has some relevant history. Jeff -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

Re: [Mpi-forum] reading announcement for issue 738 / pull 872 (MPI_GET_ADDRESS and C &)
by Jeff Hammond 25 Aug '23

25 Aug '23

I rewrote the text in response to feedback. I just removed the part about segmented addressing, which is no longer relevant. If you think we need to say something about not supporting 16-bit x86 real mode, feel free to propose a change. Jeff On Thu, Aug 24, 2023 at 12:36 PM Jeff Hammond <jeff.science(a)gmail.com> wrote: > https://github.com/mpi-forum/mpi-standard/pull/872, which addresses > https://github.com/mpiwg-abi/mpi-standard/tree/issue-738 > >> > If you want to fight for continued support for segmented addressing or > deny the behavior of & in C99, now is your chance. > > 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

reading announcement for issue 738 / pull 872 (MPI_GET_ADDRESS and C &)
by Jeff Hammond 24 Aug '23

24 Aug '23

https://github.com/mpi-forum/mpi-standard/pull/872, which addresses https://github.com/mpiwg-abi/mpi-standard/tree/issue-738 > If you want to fight for continued support for segmented addressing or deny the behavior of & in C99, now is your chance. Jeff -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

MPI 4.1 Git Repository
by Wes Bland 23 Aug '23

23 Aug '23

Hi all, I want to update folks on the status of the MPI standard document repository. As discussed in the July 2023 meeting, we’re now in “release candidate mode”, which means that between now and the release of the MPI 4.1 document, we won’t be merging any more changes into the mpi-4.x branch until the document has been ratified (or we start the process over). This means that all changes from this point on need to be based on the mpi-41-rc branch. I created the branch this morning and it’s now available in the GitHub repository. From a practical perspective, this means that if you’re making any further edits to address minor issues, you should start from the mpi-41-rc branch on your laptop. When you push your changes to GitHub as a pull request, it will automatically use the mpi-41-rc branch as I’ve made it the default. If there are any outstanding changes that are already in pull requests, I haven’t changed any existing PRs. You can do that yourself, if you’d like, by opening your pull requests’s page, clicking the “edit” button in the top right corner, then clicking the “base” dropdown underneath the PR title. Choose mpi-41-rc and GitHub should do the rest for you and let you know if there’s a problem. If you have any issues, please let me know. If you’d like to see the compiled MPI 4.1 RC1 document, it’s available (or will be in a few minutes as the site was just published) on the docs page on the website: https://www.mpi-forum.org/docs/ Thanks, Wes

1 0

reading announcement for issue 693 / pull 816 (normative language references)
by Jeff Hammond 21 Aug '23

21 Aug '23

I am not sure what the exact deadline is, but I will read this one virtually in Bristol. https://github.com/mpi-forum/mpi-standard/pull/816, which addresses https://github.com/mpi-forum/mpi-issues/issues/693. Jeff -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

September 2023 Meeting Registration
by Wes Bland 14 Aug '23

14 Aug '23

Hi all, Registration for the EuroMPI 2023 conference and MPI Forum meeting is now open. You can find that link here: https://www.eventbrite.co.uk/e/iwomp-2023-eurompi-and-mpi-forum-tickets-666… If you plan to attend only the MPI Forum portion of the meeting (even remotely), you need to register at the link above. If you plan to attend the MPI Forum meeting at all, please also register at the link below so we can gather the appropriate information for our voting rules and demographics that we provide to Women in HPC. https://forms.gle/EMHZgSq7nSJ7gFwk6 Thanks, Wes

1 0