cannam@95: Next: 2d MPI example, cannam@95: Previous: FFTW MPI Installation, cannam@95: Up: Distributed-memory FFTW with MPI cannam@95:
Programs using the MPI FFTW routines should be linked with
cannam@95: -lfftw3_mpi -lfftw3 -lm on Unix in double precision,
cannam@95: -lfftw3f_mpi -lfftw3f -lm in single precision, and so on
cannam@95: (see Precision). You will also need to link with whatever library
cannam@95: is responsible for MPI on your system; in most MPI implementations,
cannam@95: there is a special compiler alias named mpicc to compile and
cannam@95: link MPI code.
cannam@95:
cannam@95:
cannam@95:
Before calling any FFTW routines except possibly
cannam@95: fftw_init_threads (see Combining MPI and Threads), but after calling
cannam@95: MPI_Init, you should call the function:
cannam@95:
cannam@95:
void fftw_mpi_init(void); cannam@95:cannam@95:
cannam@95: If, at the end of your program, you want to get rid of all memory and cannam@95: other resources allocated internally by FFTW, for both the serial and cannam@95: MPI routines, you can call: cannam@95: cannam@95:
void fftw_mpi_cleanup(void); cannam@95:cannam@95:
cannam@95: which is much like the fftw_cleanup() function except that it
cannam@95: also gets rid of FFTW's MPI-related data. You must not execute
cannam@95: any previously created plans after calling this function.
cannam@95:
cannam@95:
cannam@95:
cannam@95: