What is the correct way to use the parallel_for_each loop in the new C11 standard?

Question

C++ / C#

Good afternoon, in my program I wanted to use the new C11 standard to calculate pi, but I can’t cope with the parallel_for_each loop, because an error appears when using it.
Program code:

#include <amp.h>
#include <iostream>
#include "amp_tinymt_rng.h"
#include "amp_math.h"
#include <sstream>
#include <ctime>
#include <fstream>
#include <string>
#include <iomanip>
using namespace concurrency;
#define licznik1_PFEO 2 //(200/(2*2*2*2*2*2))

#define licznik2_GPU  (5000000/(2*2*2*2*2*2*2*2*2))

#define n 2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2
// n CUDA Cores - Shader Processing Units


////////////////////////////////////////////////////////////////////

int main()
{
  //////-accelerator-//////
  accelerator default_device;
  std::wcout << L"Using device : " << default_device.get_description() << std::endl;
  if (default_device == accelerator(accelerator::direct3d_ref))
    std::cout << "WARNING!! Running on very slow emulator! Only use this accelerator for debugging." << std::endl;
  //////-FILE-//////
  accelerator_view view = default_device.create_view(queuing_mode::queuing_mode_automatic);
  FILE *ofile = NULL;
  fopen_s(&ofile, "PiMonteCarlo.txt", "wt");

  //////-Pi-//////
  int seed = 5489;
  double PI25DT = 3.141592653589793238462643;

  double piMC;
  double t1, t2; // time’s momemts
  int k;
  int licznik1;
  double trafenie_sum, N;
  N = n*licznik1_PFEO*licznik2_GPU;
  trafenie_sum = 0;
  

  extent<1> e_trafenie_idx_sum_h(n);

  array<double, 1> trafenie_idx_sum_d(e_trafenie_idx_sum_h);
  std::vector<double> trafenie_idx_sum_h(e_trafenie_idx_sum_h.size());


  t1 = clock(); 
  for (licznik1 = 1; licznik1 <= licznik1_PFEO; licznik1++)
  {
    tinymt_collection<1> myrand(e_trafenie_idx_sum_h, seed + licznik1);
    //////////////////////////parallel_for_each////////////////////////////
    parallel_for_each(e_trafenie_idx_sum_h, [=, &trafenie_idx_sum_d](index<1> idx) restrict(amp)
    {
      auto t = myrand[idx];
      float x, y;
      int licznik2;
      int trafenie_idx = 0;

      for (licznik2 = 0; licznik2 < licznik2_GPU; licznik2++)
      {
        x = t.next_single();
        y = t.next_single();
        if (x*x + y*y <= 1) trafenie_idx++;
      } //end for licznik2

      trafenie_idx_sum_d[idx] = trafenie_idx;
    });
    //////////////////////////parallel_for_each////////////////////////////

    copy(trafenie_idx_sum_d, trafenie_idx_sum_h.begin());

    for (k = 0; k < n; k++) trafenie_sum = trafenie_sum + trafenie_idx_sum_h[k];

  } // end for licznik1

  piMC = (4 * trafenie_sum) / N;
  t2 = clock();

  fprintf_s(ofile, "%.0f wielkosc_statystiki; czas %.5lf [sec];\n", N, ((t2 - t1) / CLOCKS_PER_SEC));
  fprintf_s(ofile, "seed=%.d generator licb losowych 'TinyMT', a variant of Mersenne twister;\n", seed);
  fprintf_s(ofile, "pi=%.10f; Epi=%.10f; \n", piMC, ((100 * (PI25DT - piMC)) / PI25DT));
  fprintf_s(ofile, "liczba_serii * liczba_cores * liczba_losowan_w_serii = wielkosc_statystiki \n");
  fprintf_s(ofile, "%d * %d * %d = %.0f; \n", licznik1_PFEO, n, licznik2_GPU, N);

  fclose(ofile);

  return 0;
}

As I understand it, the problem appears through exceeding the TDR limit of 2 seconds. But even if I decrease the maximum number of calculation cycles, I still get this error:

Perhaps the problem is that the program is executing outside the memory defined for it, but I am new to programming and do not know how to deal with such problems. Help me please.