#include <stdio.h>
#define N 10

__global__ void add(int* a, int* b, int* c)
{
    int tid = blockIdx.x;  // 计算位于这个索引处的数据
    if (tid < N)
        c[tid] = a[tid] + b[tid];
}

int main(void)
{
    int a[N], b[N], c[N];
    int* dev_a, * dev_b, * dev_c;
    // 在GPU上分配内存
    cudaError_t(cudaMalloc((void**)& dev_a, N * sizeof(int)));
    cudaError_t(cudaMalloc((void**)& dev_b, N * sizeof(int)));
    cudaError_t(cudaMalloc((void**)& dev_c, N * sizeof(int)));
    // 在CPU上为数组'a'和'b'赋值
    for (int i = 0; i < N; i++)
    {
        a[i] = -i;
        b[i] = i * i;
    }
    // 将数组'a'和'b'复制到GPU
    cudaError_t(cudaMemcpy(dev_a, a, N * sizeof(int), cudaMemcpyHostToDevice));
    cudaError_t(cudaMemcpy(dev_b, b, N * sizeof(int), cudaMemcpyHostToDevice));
    add << <N, 1 >> > (dev_a, dev_b, dev_c);
    // 将数组'c'从GPU复制到CPU
    cudaError_t(cudaMemcpy(c, dev_c, N * sizeof(int), cudaMemcpyDeviceToHost));
    // 显示结果
    for (int i = 0; i < N; i++)
    {
        printf("%d + %d = %d\n", a[i], b[i], c[i]);
    }
    // 释放在GPU上分配的内存
    cudaFree(dev_a);
    cudaFree(dev_b);
    cudaFree(dev_c);
    return 0;
}