test.cu

#include "cuda_runtime.h"
#include "device_launch_parameters.h"

#include "Vector.h"

__global__ void vectorOperations(Vector* a, Vector* b, Vector* result)
{
    int index = blockIdx.x;
    result[index] = a[index] + b[index];
}

extern "C" void CALL_KERNEL(Vector* dev_a, Vector* dev_b, Vector* dev_result)
{
    vectorOperations<<<3, 1>>>(dev_a, dev_b, dev_result);
}