使用Mpi_Scatter和Mpi_Gather实现矩阵乘法

#include <mpi.h>
#include <stdio.h>

void matrix_multiply(double *A, double *B, double *C, int n) {
    int rank, size;
    MPI_Comm_rank(MPI_COMM_WORLD, &rank);
    MPI_Comm_size(MPI_COMM_WORLD, &size);

    int block_size = n / size;
    double local_A[block_size][n], local_B[n][block_size], local_C[block_size][block_size];

    // Scatter A and B to all processes
    MPI_Scatter(A, block_size * n, MPI_DOUBLE, local_A, block_size * n, MPI_DOUBLE, 0, MPI_COMM_WORLD);
    MPI_Scatter(B, n * block_size, MPI_DOUBLE, local_B, n * block_size, MPI_DOUBLE, 0, MPI_COMM_WORLD);

    // Local multiplication
    for (int i = 0; i < block_size; i++) {
        for (int j = 0; j < block_size; j++) {
            local_C[i][j] = 0;
            for (int k = 0; k < n; k++) {
                local_C[i][j] += local_A[i][k] * local_B[k][j];
            }
        }
    }

    // Gather results back to root process
    MPI_Gather(local_C, block_size * block_size, MPI_DOUBLE, C, block_size * block_size, MPI_DOUBLE, 0, MPI_COMM_WORLD);
}

int main(int argc, char **argv) {
    MPI_Init(&argc, &argv);

    int n = 1000; // Matrix size
    double *A = malloc(n * n * sizeof(double));
    double *B = malloc(n * n * sizeof(double));
    double *C = malloc(n * n * sizeof(double));

    // Initialize matrices A and B
    // ...

    matrix_multiply(A, B, C, n);

    // Print or process the result matrix C
    // ...

    free(A);
    free(B);
    free(C);
    MPI_Finalize();
    return 0;
}