NAME

MPI_Finalize - Finish using the MPI library

SYNOPSIS

#include <mpi.h> 
int MPI_Finalize(void); 
#include <pmpi.h> 
int PMPI_Finalize(void); 

DESCRIPTION

The MPI_Finalize() function is used to shut down the MPI implementation. No MPI function other than MPI_Finalized() may be called after this is done.

This function will terminate all ranks except for the initial process used to start MPI. All buffers will be freed, and all MPI network communication will end. This function will free all attributes attached to MPI_COMM_SELF and MPI_COMM_WORLD. By having an attribute delete function for these communicators, code may be notified of MPI's shutdown.

This function will detach any buffers attached by the MPI_Buffer_attach() function.

PMPI_Finalize() is the profiling version of this function.

ERRORS

All MPI routines except for MPI_Wtime and MPI_Wtick return an error value. The the current MPI error handler is invoked if this return value is not MPI_SUCCESS. The default error handler aborts, but this may be changed with by using the MPI_Errhandler_set() function. The predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned instead. Note that MPI does not guarentee that an MPI program can continue past an error. In this implementation, all errors except MPI_ERR_INTERN or MPI_ERR_OTHER should always be recoverable.

MPI_SUCCESS - No error;

MPI_ERR_OTHER - MPI has already shut down;

MPI_ERR_INTERN - Out of Memory. This may be fatal.

SEE ALSO

MPI_Init (3) MPI_Initialized (3) MPI_Finalized (3) MPI_Abort (3) MPI_Buffer_detach (3) MPI_Keyval_create (3)

• Lockless
• MPI Documentation
• MPI_Finalize