Parallel Matrix Multiply

/* Set the chunk size (amount of work for each process */
chunk = dim / sqrt(num_proc);

/* Get the process id */
rank = getpid() - base_pid;

/* Calculate the offsets for the result matrix */
offset_row = (rank % (dim / chunk)) * chunk;
offset_col = floor((rank * chunk) / dim) * chunk;

/* Multiply the matricies */
for (i = offset_row; i < (offset_row + chunk); i++) {
    for (j = offset_col; j < (offset_col + chunk); j++) {
        for (k = 0; k < dim; k++) {
            shmatrices->C[i][j] += shmatrices->A[i][k] * shmatrices->B[k][j];
        }
    }
}