simple SAXPY operation

saxpy.cu

__global__ void saxpy(int n, float a, float *__restrict__ x, float *__restrict__ y)
{
  int i = blockIdx.x * blockDim.x + threadIdx.x;
  if( i < n )
    y[i] = a * x[i] + y[i];
}

int main()
{
  int N = 1 << 16;
  
  int size = N * sizeof(float);
  float *h_x = (float*)malloc(size);
  float *h_y = (float*)malloc(size);
  
  float *d_x;
  float *d_y;
  
  cudaMalloc((void**) &d_x, size);
  cudaMalloc((void**) &d_y, size);
  
  for(int i=0; i < N; i++)
  {
    h_x[i] = 2.0;
    h_y[i] = 2.0;
  }
  
  cudaMemcpy(d_x, h_x, size, cudaMemcpyHostToDevice);
  cudaMemcpy(d_y, h_y, size, cudaMemcpyHostToDevice);
  
  saxpy<<<256, 256>>>(N, 2.0, d_x, d_y);
  
  cudaMemcpy(h_y, d_y, size, cudaMemcpyDeviceToHost);
  
  cudaFree(d_x);
  cudaFree(d_y);
  
  free(h_x);
  free(h_y);
  
  return 0;
}