#include <stdlib.h>
#include <stdio.h>
#include <mkl_cblas.h>

int main()
{
    float *a, *b, *c;
    int n = 1000;

    a = calloc(n*n, sizeof(float));
    b = calloc(n*n, sizeof(float));
    c = calloc(n*n, sizeof(float));

    cblas_sgemm(CblasColMajor, CblasNoTrans, CblasNoTrans, n, n, 
                n, 0.5, a,
                n, b, n,
                0.5, c, n);

    printf("%f\n", c[n*n/2]);
    fflush(stdout);

    system("/bin/ls");

    free(a);
    free(b);
    free(c);
}