MPI. Running mpi in function

I want to run some function using mpi from main but I don't know how it should be. It looks like:

#define MAXSIZE 100
int main (int argc, char **argv) {
  int i;
  float matrixA[MAXSIZE][MAXSIZE], matrixB[MAXSIZE][MAXSIZE], matrixC[MAXSIZE][MAXSIZE];
  for(i=0;i<10;i++){
    multiply(matrixA, matrixB, matrixC);
  }
}
void multiply(float matrixA[MAXSIZE][MAXSIZE], float matrixB[MAXSIZE][MAXSIZE], float matrixC[MAXSIZE][MAXSIZE]) {
  int rank; //process rank
  int size; //number of processes
  MPI_Init(&argc, &argv); //initialize MPI operations
  MPI_Comm_rank(MPI_COMM_WORLD, &rank); //get the rank
  MPI_Comm_size(MPI_COMM_WORLD, &size); //get number of processes

  ...someoperation...
  MPI_Finalize()开发者_开发百科;
}

I know how to run basic MPI without using other functions but I need this construction.

In an application instance, MPI can be initialized at most once. So the code structure you provided will not work.

the correct structure for your program is as follows:

#define MAXSIZE 100
int main (int argc, char **argv) {
  int i;
  float matrixA[MAXSIZE][MAXSIZE], matrixB[MAXSIZE][MAXSIZE], matrixC[MAXSIZE][MAXSIZE];
  int rank; //process rank
  int size; //number of processes
  MPI_Init(&argc, &argv); //initialize MPI operations
  MPI_Comm_rank(MPI_COMM_WORLD, &rank); //get the rank
  MPI_Comm_size(MPI_COMM_WORLD, &size); //get number of processes
  for(i=0;i<10;i++){
    multiply(matrixA, matrixB, matrixC);
  }
  MPI_Finalize();
}
void multiply(float matrixA[MAXSIZE][MAXSIZE], float matrixB[MAXSIZE][MAXSIZE], float matrixC[MAXSIZE][MAXSIZE]) {    
  ...someoperation...
}

This might be helpful to you.In your program, there is a for loop for(i=0;i<10;i++) { multiply(matrixA, matrixB, matrixC); } I feel that you are trying to execute multiplication 10 times.You can give each multiplication to one process.So you can use the command, mpirun -np 10 executable.