bosilca
bosilca
As described in my priuor comment, this PR does not provide a correct solution.
The choice is between a deterministic bad behavior or a non-deterministic behavior. In two months we will forget about this non-determinism and we will start complaining that our accelerator tests...
I don't think this analysis is correct. 1. Nobody can take that element. This entire function is done in the context of the thread handling the current device (where the...
It is not possible to identify neither unused nor incorrect MCA params. The first because we extract all registered MCA params in local variables, and after this parsec does not...
This will be addressed by [bitbucket:#552](https://bitbucket.org/icldistcomp/parsec/pull-requests/552).
please rebase and reassess the changes to the CI part (not clear they are still needed).
Discussion on 10/07/22: Add support for a simple API (based on the MPI Info) and we pass this as the last argument to `parsec_init`. We keep the pointers to argc...
As discussed on 03/31/23 we need to merge this with #443 and get them in.
Now that #570 has been merged is this PR still necessary ?
Few high level comments for now: - is there any performance impact with this change ? - what is this providing the engine with ? - what thread is triggering...