CUDA vector addition

vector_addition.cu

/* vector_addition.cu
 * Available as a Gist at https://tinyurl.com/y34nfxgc
 */

#include <stdio.h>
#include <cuda.h>

#define N 16

__global__ void vector_addition(int* d_u, int* d_v, int* d_w)
{
    int i = blockDim.x * blockIdx.x + threadIdx.x;
    d_w[i] = d_u[i] + d_v[i];
}

int main()
{
    // Allocate host memory
    int a[N], b[N], c[N];
    for (int i=0; i<N; i++) {
        a[i] = i;
        b[i] = N - i;
    }

    // Allocate device memory
    int *d_a, *d_b, *d_c;
    size_t N_B = N * sizeof(int);
    cudaMalloc(&d_a, N_B);
    cudaMalloc(&d_b, N_B);
    cudaMalloc(&d_c, N_B);

    // Copy to device
    cudaMemcpy(d_a, a, N_B, cudaMemcpyHostToDevice);
    cudaMemcpy(d_b, b, N_B, cudaMemcpyHostToDevice);

    // Add the Vectors
    dim3 blocks_per_grid(N/16, 1, 1);
    dim3 threads_per_block(16, 1, 1);
    vector_addition<<<blocks_per_grid, threads_per_block>>>(d_a, d_b, d_c);

    // Copy from the device
    cudaMemcpy(c, d_c, N_B, cudaMemcpyDeviceToHost);

    // Print stuff
    for (int i=0; i<N; i++)
        printf("%i + %i = %i\n", a[i], b[i], c[i]);

    // Free memory
    cudaFree(d_a);
    cudaFree(d_b);
    cudaFree(d_c);

    return 0;
}