bosilca
bosilca
**First**, the weird output of the test provided here is due to a bug in the test. `MPI_T_pvar_get_info` has the `name_len` and `desc_len` parameters as INOUT. This means before the...
Your barrier graphs are missing legends or titles to be able to understand what they are showing. Different number of processes I assume ?
finally done with the review and :+1:. However, I would like to revisit the discussion above about the `rbuf` on reduce for non-root. I understand the interest, but I do...
I see, you need more than local knowledge about the buffers. My patch will definitely not provide that, it just give us all (aka implementers of collective algorithms) a better...
@gkatev I sent you an email ealier this week about the topics mentioned above but did not have received an answer yet.
No problem, they either bounced back (the email from the XHC paper), and acted as a black hole (his github email). Let's hope direct poke @gkatev works better ;)
Please revise this PR to account for the merge of the init/fini PR (https://github.com/open-mpi/ompi/pull/12429)
It took me a while to understand what this change is doing on the tests. We should document this change better. Btw, how are the tests using the components in...
I am not completely convinced by this. Preventing the use of components from an old installation (especially one that is not binary compatible internally), is a desired features, but this...