#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include <stdlib.h>
#include <stdio.h>
#include <string.h>

__global__ void suma(int* a, int* b, int* c)
{

	int idThread = blockIdx.x;
	c[idThread] = a[idThread] + b[idThread];

}

__host__ __device__ void printHolaMundo()
{


	printf("Hola mundo\n");
}

__global__ void holaMundo()
{
	printHolaMundo();

}

int main(int argc, char**  argv)
{

	int i;
	int *a, *b, *c;
	int *d_a, *d_b, *d_c;

	int vSize = 10; //vector Size

	a = (int*)malloc(sizeof(int) * vSize);
	b = (int*)malloc(sizeof(int) * vSize);
	c = (int*)malloc(sizeof(int) * vSize);


	cudaMalloc((void **)&d_a, sizeof(int) * vSize);
	cudaMalloc((void **)&d_b, sizeof(int) * vSize);
	cudaMalloc((void **)&d_c, sizeof(int) * vSize);

	//inicializacion

	for ( i = 0; i < vSize; i++)
	{
		a[i] = b[i] = i;
	}

	cudaMemcpy(d_a, a, vSize * sizeof(int), cudaMemcpyHostToDevice);
	cudaMemcpy(d_b, b, vSize * sizeof(int), cudaMemcpyHostToDevice);
	cudaMemset(d, 0xFF, vSize * sizeof(int));












	h_c = (int*)malloc(sizeof(int));
	a = 10;
	b = 10;
	*h_c = 10;

	//cudaMemcpy(	d_c, h_c, sizeof(int), cudaMemcpyHostToDevice);

	suma << <1, 1 >> > (a, b, d_c);

	cudaMemcpy(h_c, d_c, sizeof(int), cudaMemcpyDeviceToHost);
	printf("resultado: %d\n", *h_c);


	cudaFree(d_c);
	free(h_c);


	return 0;
}