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

__global__ void incr(float *a, int N){
    int idx = blockIdx.x * blockDim.x + threadIdx.x;
    if(idx < N){
        a[idx] *= 10;
    }
}

int main(void)
{
float *a_h, *a_d;
const int N = 100;
size_t size = N*sizeof(float);

a_h = (float *)malloc(size);
cudaMalloc( (void**) &a_d, size);

for(int i=0; i<N; ++i){
    a_h[i] = (float)i;
}

cudaMemcpy(a_d, a_h, size, cudaMemcpyHostToDevice);

int block_size = 4;
int n_blocks = N/block_size + (N%block_size == 0 ? 0:1);

incr <<<n_blocks, block_size>>> (a_d, N);

cudaMemcpy(a_h, a_d, size, cudaMemcpyDeviceToHost);

for(int i=0; i<N; ++i){
    printf("%d: %3.0f\n", i, a_h[i]);
}

free(a_h); cudaFree(a_d);

system("Pause");
}