MPI_Rget_accumulate function

Request-based RMA read-modify-write operation returns the data before the accumulate operation.

Syntax

int MPIAPI MPI_Rget_accumulate(
  _In_  void         *origin_addr,
        int          origin_count,
        MPI_Datatype origin_datatype,
  _Out_ void         *result_addr,
        int          result_count,
        MPI_Datatype result_datatype,
        int          target_rank,
        MPI_Aint     target_disp,
        int          target_count,
        MPI_Datatype datatype,
        MPI_Op       op,
        MPI_Win      win,
  _Out_ MPI_Request  *request
);

Parameters

• origin_addr [in]
  initial address of buffer

• origin_count
  number of entries in buffer

• origin_datatype
  datatype of each buffer entry

• result_addr [out]
  initial address of result buffer

• result_count
  number of entries in result buffer

• result_datatype
  datatype of each entry in result buffer

• target_rank
  rank of target

• target_disp
  displacement from start of window to beginning of target buffer

• target_count
  number of entries in target buffer

• datatype
  datatype of each entry in target buffer

• op
  predefined reduce operation

• win
  window object

• request [out]
  RMA request

Return value

Returns MPI_SUCCESS on success. Otherwise, the return value is an error code.

In Fortran, the return value is stored in the IERROR parameter.

Fortran

    MPI_GET_RACCUMULATE(ORIGIN_ADDR, ORIGIN_COUNT, ORIGIN_DATATYPE, RESULT_ADDR, RESULT_COUNT, RESULT_DATATYPE,
                TARGET_RANK, TARGET_DISP, TARGET_COUNT, TARGET_DATATYPE, OP, WIN, REQUEST, IERROR)
        <type> ORIGIN_ADDR(*), RESULT_ADDR(*)
        INTEGER(KIND=MPI_ADDRESS_KIND) TARGET_DISP
        INTEGER ORIGIN_COUNT, ORIGIN_DATATYPE, RESULT_COUNT, RESULT_DATATYPE, TARGET_RANK, TARGET_COUNT,
        TARGET_DATATYPE, OP, WIN, REQUEST, IERROR

Remarks

MPI_Rget_accumulate is similar to MPI_Get_accumulate, except that it allocates a communication request object and associates it with the request handle (the argument request) that can be used to wait or test for completion. The completion of an MPI_Rget_accumulate operation indicates that the data is available in the result buffer and the origin buffer is free to be updated. It does not indicate that the operation has been completed at the target window.

Requirements

See also

MPI One-Sided Communications Functions