CUDA, не получается вывести значение

#include <stdio.h>
#include <conio.h>
#include <iostream>
#include <math.h>
using namespace std;
__global__  void tablekernel(float * devPtr,float step)
{
 int index = blockIdx.x * blockDim.x + threadIdx.x;
 float x = step * index;
 devPtr [index] =  __sinf(x);
}
void buildTable(float * res, int n, float step)
{
  float * devPtr;
  cudaMalloc( &devPtr, n * sizeof(float));
  tablekernel<<<dim3(n/256), dim3(256)>>>(devPtr,step);
  cudaMemcpy(res,devPtr,n * sizeof(float),cudaMemcpyDeviceToHost);
  cudaFree(devPtr);
}
int main(int argc, char * argv [])
{
 cudaEvent_t start,stop;
 cudaEventCreate(&start);
 cudaEventCreate(&stop);
 float gputime =0.0f;
 float Table[10];
 cudaEventRecord(start,0);
 buildTable(Table,11,4);
  cudaEventRecord(stop,0);
  cudaEventSynchronize(stop);
 for(int i = 0;i < 10;i++)
 {
     cout<<Table[i]<<endl;
 }
 cudaEventElapsedTime(&gputime,start,stop);
 cout << "Elapsed time : "<<gputime<<"  ms";
 getch();
 return 0;
}

tablekernel<<<dim3((n+255)/256), dim3(256)>>>(devPtr,step,n);

__global__  void tablekernel(float * devPtr,float step, int N)
{
 int index = blockIdx.x * blockDim.x + threadIdx.x;
 if (index<N) {
 float x = step * index;
 devPtr [index] =  __sinf(x);}
}