[Mpi3-rma] RMA WG discussion

[Jim Dinan]

On 12/7/12 9:56 AM, Jeff Hammond wrote:
>> I have a further question to MPI_NO_OP and MPI_{R}Get_accumulate.
>> I assume that origin_count is not significant for MPI_NO_OP
>> (such as origin_datatype)
>
> Like origin_addr, origin_count and origin_datatype should be ignored
> when MPI_NO_OP is used.  I'm not sure that the text needs to be
> changed for this though.  Pavan said that there is some text that
> asserts that buffers (i.e. pointers) be valid (i.e. not NULL) unless
> otherwise stipulated.  Is there text to the same effect for count and
> datatype?  If one uses count=0 in MPI_SEND, does the datatype have to
> be valid?  If not, then we can ignore that argument at least.  Since
> we don't have precedent for MPI_NO_OP, it seems that the notion that
> count can be ignored is implicitly obvious but nonetheless should be
> stated (or we could require count=0 for MPI_NO_OP...).

This is covered by ticket #350:

https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/350

Please take a look and attach any feedback to the ticket.  The current 
proposed text, that would be added to cover all accumulate ops where 
NO_OP is allowed, states:

"When MPI_NO_OP is specified as the operation, the origin buffer 
arguments are ignored."

This includes the pointer, count, and datatype.

>> Which count variable defines the number of basic elements to be returned ?
>>    (result_count, result_type) or (target_count, target_type) ?
>
> result_addr is the buffer that is written at the origin so I believe
> the answer is (result_count, result_type).

The number of basic elements needs to be the same across all <count, 
dtype> pairs.  The count and datatype for the result buffer determine 
where these elements will be in the origin process' memory.

>> Should an error code be returned if the other number is too small ?

It is erroneous if the number of basic elements does not agree across 
origin, result, and target buffers.  Although, I would expect most MPI 
implementations to not report this error, since there is a significant 
cost to detect it.

  ~Jim.