Computes an exclusive scan (partial reduction)
#include <mpi.h> int MPI_Exscan(void *sendbuf, void *recvbuf, int count, MPI_Datatype datatype, MPI_Op op, MPI_Comm comm)
INCLUDE 'mpif.h' MPI_SCAN(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COMM, IERROR) <type> SENDBUF(*), RECVBUF(*) INTEGER COUNT, DATATYPE, OP, COMM, IERROR
#include <mpi.h> void MPI::Intracomm::Exscan(const void* sendbuf, void* recvbuf, int count, const MPI::Datatype& datatype, const MPI::Op& op) const
Send buffer (choice).
Number of elements in input buffer (integer).
Data type of elements of input buffer (handle).
Receive buffer (choice).
Fortran only: Error status (integer).
MPI_Exscan is used to perform an exclusive prefix reduction on data distributed across the calling processes. The operation returns, in the recvbuf of the process with rank i, the reduction (calculated according to the function op) of the values in the sendbufs of processes with ranks 0, ..., i-1. Compare this with the functionality of MPI_Scan, which calculates over the range 0, ..., i (inclusive). The type of operations supported, their semantics, and the constraints on send and receive buffers are as for MPI_Reduce.
The value in recvbuf on process 0 is undefined and unreliable as recvbuf is not significant for process 0. The value of recvbuf on process 1 is always the value in sendbuf on process 0.
No MPI_IN_PLACE operation is supported.
MPI does not specify which process computes which operation. In particular, both processes 0 and 1 may participate in the computation even though the results for both processes' recvbuf are degenerate. Therefore, all processes, including 0 and 1, must provide the same op.
It can be argued, from a mathematical perspective, that the definition of MPI_Exscan is unsatisfactory because the output at process 0 is undefined. The "mathematically correct" output for process 0 would be the unit element of the reduction operation. However, such a definition of an exclusive scan would not work with user-defined op functions as there is no way for MPI to "know" the unit value for these custom operations.
The reduction functions of type MPI_Op do not return an error value. As a result, if the functions detect an error, all they can do is either call MPI_Abort or silently skip the problem. Thus, if the error handler is changed from MPI_ERRORS_ARE_FATAL to something else (e.g., MPI_ERRORS_RETURN), then no error may be indicated.
The reason for this is the performance problems in ensuring that all collective routines return the same error value.
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.
See the MPI man page for a full list of MPI error codes.
MPI_Op_create MPI_Reduce MPI_Scan