Generates a new attribute key for caching on data types.
#include <mpi.h> int MPI_Type_create_keyval(MPI_Type_copy_attr_function *type_copy_attr_fn, MPI_Type_delete_attr_function *type_delete_attr_fn, int *type_keyval, void *extra_state)
INCLUDE 'mpif.h' MPI_TYPE_CREATE_KEYVAL(TYPE_COPY_ATTR_FN, TYPE_DELETE_ATTR_FN, TYPE_KEYVAL, EXTRA_STATE, IERROR) EXTERNAL TYPE_COPY_ATTR_FN, TYPE_DELETE_ATTR_FN INTEGER TYPE_KEYVAL, IERROR INTEGER(KIND=MPI_ADDRESS_KIND) EXTRA_STATE
#include <mpi.h> static int MPI::Datatype::Create_keyval(MPI::Datatype::Copy_attr_function* type_copy_attr_fn, MPI::Datatype::Delete_attr_function* type_delete_attr_fn, void* extra_state)
type_copy_attr_fn
Copy callback function for type_keyval (function).
type_delete_attr_fn
Delete callback function for type_keyval (function).
extra_state
Extra state for callback functions.
type_keyval
Key value for future access (integer).
IERROR
Fortran only: Error status (integer).
MPI_Type_create_keyval generates a new attribute key for caching on data types. This routine partially replaces MPI_Keyval_create.
The argument type_copy_attr_fn may be specified as MPI_TYPE_NULL_COPY_FN or MPI_TYPE_DUP_FN from C, C++, or Fortran. MPI_TYPE_NULL_COPY_FN is a function that does nothing other than returning flag = 0 and MPI_SUCCESS. MPI_TYPE_DUP_FN is a simple-minded copy function that sets flag = 1, returns the value of attribute_val_in in attribute_val_out, and returns MPI_SUCCESS.
The argument type_delete_attr_fn may be specified as MPI_TYPE_NULL_DELETE_FN from C, C++, or Fortran. MPI_TYPE_NULL_DELETE_FN is a function that does nothing beyond returning MPI_SUCCESS. The C callback functions are:
typedef int MPI_Type_copy_attr_function(MPI_Datatype oldtype, int type_keyval, void *extra_state, void *attribute_val_in, void *attribute_val_out, int *flag);
and
typedef int MPI_Type_delete_attr_function(MPI_Datatype type, int type_keyval, void *attribute_val, void *extra_state);
The Fortran callback functions are:
SUBROUTINE TYPE_COPY_ATTR_FN(OLDTYPE, TYPE_KEYVAL, EXTRA_STATE, ATTRIBUTE_VAL_IN, ATTRIBUTE_VAL_OUT, FLAG, IERROR) INTEGER OLDTYPE, TYPE KEYVAL, IERROR INTEGER(KIND=MPI_ADDRESS_KIND) EXTRA_STATE, ATTRIBUTE_VAL_IN, ATTRIBUTE_VAL_OUT LOGICAL FLAG
and
SUBROUTINE TYPE_DELETE_ATTR_FN(TYPE, TYPE_KEYVAL, ATTRIBUTE_VAL, EXTRA_STATE, IERROR) INTEGER TYPE, TYPE_KEYVAL, IERROR INTEGER(KIND=MPI_ADDRESS_KIND) ATTRIBUTE VAL, EXTRA_STATE
The C++ callbacks are:
typedef int MPI::Datatype::Copy_attr_function(const MPI::Datatype& oldtype, int type_keyval, void* extra_state, const void* attribute_val_in, void* attribute_val_out, bool& flag);
and
typedef int MPI::Datatype::Delete_attr_function(MPI::Datatype& type, int type_keyval, void* attribute_val, void* extra_state);
The MPI standard prescribes portable Fortran syntax for the EXTRA_STATE argument only for Fortran 90. FORTRAN 77 users may use the non-portable syntax
INTEGER*MPI_ADDRESS_KIND EXTRA_STATE
where MPI_ADDRESS_KIND is a constant defined in mpif.h and gives the length of the declared integer in bytes.
Almost all MPI routines return an error value; C routines as the value of the function and Fortran routines in the last argument. C++ functions do not return errors. If the default error handler is set to MPI::ERRORS_THROW_EXCEPTIONS, then on error the C++ exception mechanism will be used to throw an MPI:Exception object.
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.
MPI_Type_free_keyval