[mpiwg-rma] [EXTERNAL] ambiguity in MPI_WIN_FLUSH_LOCAL and completion of bidirectional operations

[Barrett, Brian W]

On 10/21/13 11:30 AM, "Jeff Hammond" <jeff.science at gmail.com> wrote:

>My second suggestion is not backwards-compatible, but only for people
>using MPI-3 and using MPI_WIN_FLUSH_LOCAL to end-to-end-complete
>MPI_GET_ACCUMULATE, MPI_FETCH_AND_OP and MPI_COMPARE_AND_SWAP.  How
>much such people are there that are not in the RMA working group or
>working directly for/with those individuals?

Assuming we made the change, it would have to occur in a 3.1-type release
(not an errata), which means that it will be a couple of years before the
change is implemented.  So I don't think breaking backwards compatibility
is a good idea at this point.

Brian

--
  Brian W. Barrett
  Scalable System Software Group
  Sandia National Laboratories