void multiply_element_threaded(struct m_data *data)
{
    mat_data = data;
    unsigned int threads_number = data->arow_size * data->bcol_size;
    pthread_t element_threads[threads_number];
    unsigned int counter = 0;
    for (int i = 0; i < data->arow_size; i++)
    {
        for (int j = 0; j < data->bcol_size; j++)
        {
            struct state *st = create_state(i, j);
            int ret = pthread_create(&element_threads[counter++], NULL, calculate_element, (void *)st);
            if (ret != 0)
            {
                fprintf(stderr, "Error occured while creating thread, error code = %d Thread number: %d\n", ret, counter);
                return;
            }
           // printf("Thread number %d is created\n", counter);
        }
    }
    for (int i = 0; i < threads_number; i++)
    {
        pthread_join(element_threads[i], NULL);
    }
    return;
}

void *calculate_element(void *ptr)
{
    struct state *st = (struct state *) ptr;
    mat_data->matrix_o[st->curr_row][st->curr_col] = 0;
    for (int i = 0; i < mat_data->acol_size; i++)
    {
        mat_data->matrix_o[st->curr_row][st->curr_col] += (mat_data->matrix_a[st->curr_row][i]
                                                 * mat_data->matrix_b[i][st->curr_col]);
    }
    free(st);
    //printf("Thread is finished : %d\n", cnt);

    pthread_exit(NULL);
}