Joseph Schuchart

Installing two binaries with the same name (putty `pterm` and prrte `pterm`) in different directories and then using `PATH` is not a solution. One of them will shadow the other,...

CI currently fails for NVIDIA because hcoll was not adapted but I figure I'll just wait for #13467 to land. Also, I plan to port the datatype tests for bigcount,...

>> why did you remove this? I'll just need to put it back in for the ABI support. > type_get_envelop.c.in? It didn't seem like it was being compiled (it had...

There are still some issues in the mpi4ai test with 4.1 enabled that I need to resolve (for some of the combiner types). I'm not sure when I'll get to...

I should say though: the design won't change anymore, so I'd like to get @bosilca's input on this approach.

All tests are passing now. I squashed some of the fixups. @hppritcha @bosilca please review

@amckinstry I suspect that it warns about the way Open MPI implements MPI handles: pointers to opaque structs. MPICH has historically used integers as MPI handles, requiring a lookup in...

Not sure there is any such guarantee. Maybe if you don't use non-world communicators. Or it might just be the way they do lookups or handle predefined communicators/windows. This should...

@bosilca and I are not sure that we really want this. I'll mark it as draft for now.

In my mind, this requires user-created datatypes to take a reference on the internal session so that the datatype survives the end of all sessions and the WPM and can...