2020 03 18

2020-03-18 meeting minutes

Attendees:

• Jeff Squyres
• Martin Rüefenacht
• Dan Holmes
• Tony Skjellum

COVID-19

Wow. It's real.

Cache-refresh on pythonization.

Martin to work on fixing the conflicts in https://github.com/mpi-forum/mpi-standard/pull/132.

Embiggening

We looked for the errata that we want to file.

See https://github.com/mpiwg-large-count/large-count-issues/wiki/problems-possible-errata.

There's also a bunch of "to do" items -- logistics and mechanical stuff.

See https://github.com/mpiwg-large-count/large-count-issues/wiki/TODOs.

Tony suggests that we need to have someone look at the IO chapter to verify the embiggening. --> Tony sends email to Rajeev.

Re-iterating the rationale about all the confusion with datatype constructors:

We must keep the "int" signature (for backwards compat), and also embiggen datatype functions two ways:

1. MPI_Aint: for use case 2 (i.e., absolute addresses)
2. MPI_Count: for use case 3 (i.e., files)

An example of "odd" embiggening: MPI_TYPE_CONTIGUOUS. The "count" parameter is of type "int"

1. Have to keep the "int" signature (for backwards compat)

2. We also need an "MPI_Aint" signature ...

3. We also need the "MPI_Count" signature for file

4. The OUT "size" param can never be an absolute address, so we don't need to worry about an "MPI_Aint" signature

5. Embiggened MPI_Count "size" param to handle the file case