Open question to MPI developers: are you using the features added in MPI-2.2?
I ask because I took a little heat in the last MPI Forum meeting for not driving Open MPI to be MPI-2.2 compliant (Open MPI is MPI-2.1 compliant; there’s 4 open tickets that need to be completed for full MPI-2.2 compliance).
But I’m having a hard time finding users who want or need these specific functionalities (admittedly, they’re somewhat obscure). We’ll definitely get to these items someday — the question is whether that someday needs to be soon or whether it can be a while from now.
Here’s a quick summary of what we have left to do:
- #1368: We’re missing some C datatypes in the Fortran bindings, and vice versa. Although to be fair, this is a pretty common problem across MPI implementations. I guess no one has asked for this particular feature in a long, long time…
- #2219: Add some new MPI datatypes (e.g., MPI_UINT8_T, MPI_AINT, …etc.).
- #2221: Make MPI_EXSCAN support MPI_IN_PLACE.
- #2223: Add the MPI_DIST_GRAPH_* functions.
All of these are Good Things, of course, but it’s a question of priority. There’s a million other Good Things to do in Open MPI — which should we do first? So far, the answer has been to do these four… er… not first.
Good decision? Bad decision? Let me know if you need this stuff.