__global__ void matrixMultiplication(matrixCol *A, matrixCol *B, matrixCol *C, int N)
{

    long long row = blockIdx.y*blockDim.y + threadIdx.y;
    long long col = blockIdx.x*blockDim.x + threadIdx.x;

    if (row < N && col < N){
        float sum = 0.f;
        for (long long n = 0; n < N; ++n)
            sum += A[row].col[n] * B[n].col[col];
        C[row].col[col] = sum;
    }
}