#include <mpi.h>
int MPI_Type_create_darray(int size, int rank, int ndims,
int array_of_gsizes[], int array_of_distribs[],
int array_of_dargs[], int array_of_psizes[],
int order, MPI_Datatype oldtype, MPI_Datatype *newtype)
INCLUDE 'mpif.h'
MPI_TYPE_CREATE_DARRAY(SIZE, RANK, NDIMS, ARRAY_OF_GSIZES,
ARRAY_OF_DISTRIBS, ARRAY_OF_DARGS, ARRAY_OF_PSIZES, ORDER,
OLDTYPE, NEWTYPE, IERROR)
INTEGER SIZE, RANK, NDIMS, ARRAY_OF_GSIZES(*), ARRAY_OF_DISTRIBS(*),
ARRAY_OF_DARGS(*), ARRAY_OF_PSIZES(*), ORDER, OLDTYPE,
NEWTYPE, IERROR
#include <mpi.h>
MPI::Datatype MPI::Datatype::Create_darray(int size, int rank,
int ndims, const int array_of_gsizes[],
const int array_of_distribs[], const int array_of_dargs[],
const int array_of_psizes[], int order) const
MPI_Type_create_darray can be used to generate the data types corresponding to the distribution of an ndims-dimensional array of oldtype elements onto an ndims-dimensional grid of logical processes. Unused dimensions of array_of_psizes should be set to 1. For a call to MPI_Type_create_darray to be correct, the equation
ndims-1
pi array_of_psizes[i] = size
i=0
must be satisfied. The ordering of processes in the process grid is assumed to be row-major, as in the case of virtual Cartesian process topologies in MPI-1.
Each dimension of the array can be distributed in one of three ways:
- MPI_DISTRIBUTE_BLOCK - Block distribution - MPI_DISTRIBUTE_CYCLIC - Cyclic distribution - MPI_DISTRIBUTE_NONE - Dimension not distributed.
The constant MPI_DISTRIBUTE_DFLT_DARG specifies a default distribution argument. The distribution argument for a dimension that is not distributed is ignored. For any dimension i in which the distribution is MPI_DISTRIBUTE_BLOCK, it erroneous to specify array_of_dargs[i] * array_of_psizes[i] < array_of_gsizes[i].
For example, the HPF layout ARRAY(CYCLIC(15)) corresponds to MPI_DISTRIBUTE_CYCLIC with a distribution argument of 15, and the HPF layout ARRAY(BLOCK) corresponds to MPI_DISTRIBUTE_BLOCK with a distribution argument of MPI_DISTRIBUTE_DFLT_DARG.
The order argument is used as in MPI_TYPE_CREATE_SUBARRAY to specify the storage order. Therefore, arrays described by this type constructor may be stored in Fortran (column-major) or C (row-major) order. Valid values for order are MPI_ORDER_FORTRAN and MPI_ORDER_C.
This routine creates a new MPI data type with a typemap defined in terms of a function called "cyclic()" (see below).
Without loss of generality, it suffices to define the typemap for the MPI_DISTRIBUTE_CYCLIC case where MPI_DISTRIBUTE_DFLT_DARG is not used.
MPI_DISTRIBUTE_BLOCK and MPI_DISTRIBUTE_NONE can be reduced to the MPI_DISTRIBUTE_CYCLIC case for dimension i as follows.
MPI_DISTRIBUTE_BLOCK with array_of_dargs[i] equal to MPI_DISTRIBUTE_DFLT_DARG is equivalent to MPI_DISTRIBUTE_CYCLIC with array_of_dargs[i] set to
(array_of_gsizes[i] + array_of_psizes[i] - 1)/array_of_psizes[i]
If array_of_dargs[i] is not MPI_DISTRIBUTE_DFLT_DARG, then MPI_DISTRIBUTE_BLOCK and DISTRIBUTE_CYCLIC are equivalent.
MPI_DISTRIBUTE_NONE is equivalent to MPI_DISTRIBUTE_CYCLIC with array_of_dargs[i] set to array_of_gsizes[i].
Finally, MPI_DISTRIBUTE_CYCLIC with array_of_dargs[i] equal to MPI_DISTRIBUTE_DFLT_DARG is equivalent to MPI_DISTRIBUTE_CYCLIC with array_of_dargs[i] set to 1.
Before the error value is returned, the current MPI error handler is called. By default, this error handler aborts the MPI job, except for I/O function errors. The error handler may be changed with MPI_Comm_set_errhandler; the predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned. Note that MPI does not guarantee that an MPI program can continue past an error.