IIRC, the argument for RMA was that we had not wanted to require the MPI library to maintain a reference to the request. There could be many RMA operations pending and the MPI library maintaining a list of pending operations could be a significant source of overhead. I seem to recall this having been the general sentiment at the time and it may have also applied to NBC. ~Jim. On Sat, Aug 8, 2020 at 7:15 PM Jeff Hammond via mpi-forum < [email protected]> wrote:
We should fix the RMA chapter with an erratum. I care less about NBC but share your ignorance of why it was done that way.
Sent from my iPhone
On Aug 8, 2020, at 6:51 AM, Balaji, Pavan via mpi-forum < [email protected]> wrote:
Folks,
Does someone remember why we disallowed users from calling MPI_Request_free on nonblocking collective requests? I remember the reasoning for not allowing cancel (i.e., the operation might have completed on some processes, but not all), but not for Request_free. AFAICT, allowing the users to free the request doesn’t make any difference to the MPI library. The MPI library would simply maintain its own refcount to the request and continue forward till the operation completes. One of our users would like to free NBC requests so they don’t have to wait for the operation to complete in some situations.
Unfortunately, when I added the Rput/Rget operations in the RMA chapter, I copy-pasted that text into RMA as well without thinking too hard about it. My bad! Either the RMA committee missed it too, or they thought of a reason that I can’t think of now.
Can someone clarify or remind me what the reason was?
Regards,
— Pavan
MPI-3.1 standard, page 197, lines 26-27:
“It is erroneous to call MPI_REQUEST_FREE or MPI_CANCEL for a request associated with a nonblocking collective operation.”
_______________________________________________ mpi-forum mailing list [email protected] https://lists.mpi-forum.org/mailman/listinfo/mpi-forum
_______________________________________________ mpi-forum mailing list [email protected] https://lists.mpi-forum.org/mailman/listinfo/mpi-forum