-
Notifications
You must be signed in to change notification settings - Fork 859
WeeklyTelcon_20210914
Geoffrey Paulsen edited this page Sep 14, 2021
·
1 revision
- Geoffrey Paulsen (IBM)
- Raghu Raja
- Austen Lauria (IBM)
- Jeff Squyres (Cisco)
- Hessam Mirsadeghi (NVIDIA))
- Josh Hursey (IBM)
- Siripaul (Intel)
- Todd Kordenbrock (Sandia)
- Howard Pritchard (LANL)
- Thomas Naughton (ORNL)
- William Zhang (AWS)
- Michael Heinz (Cornelis Networks)
- Brendan Cunningham (Cornelis Networks)
- Joseph Schuchart (HLRS)
- Tomislav Janjusic (NVIDIA)
- Matthew Dosanjh (Sandia)
- George Bosilca (UTK)
- Brian Barrett (AWS) - Welcome Back!
- David Bernholdt (ORNL)
- Harumi Kuno (HPE)
- Marisa Roman (Cornelius)
- Ralph Castain (Intel)
- Sam Gutierrez (LANL)
- Akshay Venkatesh (NVIDIA)
- Artem Polyakov (NVIDIA)
- Aurelien Bouteiller (UTK)
- Brandon Yates (Intel)
- Charles Shereda (LLNL)
- Christoph Niethammer (HLRS)
- Edgar Gabriel (UH)
- Erik Zeiske (HPE)
- Geoffroy Vallee (ARM)
- Joshua Ladd (NVIDIA)
- Mark Allen (IBM)
- Matias Cabral (Intel)
- Nathan Hjelm (Google)
- Noah Evans (Sandia)
- Scott Breyer (Sandia?)
- Shintaro iwasaki
- Xin Zhao (NVIDIA)
- Does Fortran Fixes affect API? (i.e. needed for v5.0.0?)
- PR https://github.com/open-mpi/ompi/pull/9259
- Jeff reviewed 16 days ago, looks incomplete.
- Think that 9367 addresses the issue with 9259.
- and PR https://github.com/open-mpi/ompi/pull/9367
- PR https://github.com/open-mpi/ompi/pull/9259
- Schedule: milestone is set for September for 4.0.7
- MPI_Thread_Init - https://github.com/open-mpi/ompi/pull/9312
-
https://github.com/open-mpi/ompi/issues/9332
- In ompi_init. Thread level is an INFO parameter.
- George will dig into how MCAs affect Info parameters.
- Need to revert on v4.0.x. Already reverted on v4.1.x
-
https://github.com/open-mpi/ompi/issues/9332
- Josh Investigated. Issue 9369
- non MPI procs get orphaned in timeout based shutdown.
- Schedule: Behind schedule, approximate?
- Possibly make an RC this week.
- One more pending on v4.1.x Jenkins had some issues that Brian is looking at.
- ROMIO 3.2.1 based PR 8371 do we want to take this?
- v4.1.x does this need to go back to v4.0.x?
- Schedule: aiming for rc1 on Sept 23rd.
- PMIx and/or PRRTE are releasing a new minor rev that we'll pickup for v5.0.x
- Github Project of [critical v5.0.x issues|https://github.com/open-mpi/ompi/projects/3]
- Issue 8983 - Nathan volunteered to put out a fix.
- If we partially disable OSC/TCP BTL - Not breaking MPI compliance, just breaking One-sided performance badly.
- https://github.com/open-mpi/ompi/pull/8984
- https://github.com/open-mpi/ompi/issues/7830
- users could fall back to using UCX or OFI, and not the BTLs.
- But that's a different can-of-worms
- Brian will take a look at issue.
- Described approach of rc1 on Sept 23, disabling any functionality that are blockers to allow for the rc.
- Worried that blockers might not be fixed in time, so will put in code to issue an error at runtime to prevent getting into those paths, and document it heavily.
- MPIAlltoallw needs to go in. Is a PR from Giles George
- https://github.com/open-mpi/ompi/issue/9329
- Test has been merged not a fix.
-
https://github.com/open-mpi/ompi/pull/9330
- George thinks it's ready to go.
- Jeff will review.
- Janjust - has a long outstanding one incomming.
- Still working on
- Portals bugfixes incomming.
- Todd's working on this. Hasn't posted yet. Will post this week.
-
https://github.com/open-mpi/ompi/pull/9326 should get into 5.0 too
- This fixes a correctness issues, and George is concerned about performance.
- Is argobots now unsupported?
- no. Our integration allow users to call MPI withing a blocking argobot function and this still works.
- What we think is a thread that will block in libevent, because libevent isn't aware of argobots, so libeven will block entire thread.
- George joined about this time. I think he said this was ready or that he'd re-read.
- No discussion
- MTT results look pretty good
- No update
- Don't do the old system, use this new system for v5.0.0
- No discussion [Open MPI 4.0 API Compliance Github Project|https://github.com/open-mpi/ompi/projects/2]
- Looking okay.
- Ciscos results are still hidden by default.
- No discussion.