main.cpp

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

#include <iostream>

#include "Vector.h"

using namespace std;

extern "C" void CALL_KERNEL(Vector& dev_a, Vector& dev_b, Vector& dev_result);

int main()
{
	Vector a(5, 4, 3);
	Vector b(1, 1, 1);
	Vector c;

	Vector* dev_a;
	Vector* dev_b;
	Vector* dev_c;

	cudaMalloc((void**)&dev_a, sizeof(Vector));
	cudaMalloc((void**)&dev_b, sizeof(Vector));
	cudaMalloc((void**)&dev_c, sizeof(Vector));

	cudaMemcpy(dev_a, (void*)&a, sizeof(Vector), cudaMemcpyHostToDevice);
	cudaMemcpy(dev_a, (void*)&b, sizeof(Vector), cudaMemcpyHostToDevice);
	cudaMemcpy(dev_a, (void*)&c, sizeof(Vector), cudaMemcpyHostToDevice);

	CALL_KERNEL(*dev_a, *dev_b, *dev_c);

	cudaMemcpy((void*)&c, dev_c, sizeof(Vector), cudaMemcpyDeviceToHost);

	cudaFree(dev_a);
	cudaFree(dev_b);
	cudaFree(dev_c);

	cout << c.x << endl;
}