<h2>Editing File: /usr/include/c++/8/parallel/partition.h</h2><form method='post'><textarea name='content' class='file-edit'>// -*- C++ -*-

// Copyright (C) 2007-2018 Free Software Foundation, Inc.
//
// This file is part of the GNU ISO C++ Library.  This library is free
// software; you can redistribute it and/or modify it under the terms
// of the GNU General Public License as published by the Free Software
// Foundation; either version 3, or (at your option) any later
// version.

// This library is distributed in the hope that it will be useful, but
// WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
// General Public License for more details.

// Under Section 7 of GPL version 3, you are granted additional
// permissions described in the GCC Runtime Library Exception, version
// 3.1, as published by the Free Software Foundation.

// You should have received a copy of the GNU General Public License and
// a copy of the GCC Runtime Library Exception along with this program;
// see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
// &lt;http://www.gnu.org/licenses/&gt;.

/** @file parallel/partition.h
 *  @brief Parallel implementation of std::partition(),
 *  std::nth_element(), and std::partial_sort().
 *  This file is a GNU parallel extension to the Standard C++ Library.
 */

// Written by Johannes Singler and Felix Putze.

#ifndef _GLIBCXX_PARALLEL_PARTITION_H
#define _GLIBCXX_PARALLEL_PARTITION_H 1

#include &lt;parallel/basic_iterator.h&gt;
#include &lt;parallel/sort.h&gt;
#include &lt;parallel/random_number.h&gt;
#include &lt;bits/stl_algo.h&gt;
#include &lt;parallel/parallel.h&gt;

/** @brief Decide whether to declare certain variables volatile. */
#define _GLIBCXX_VOLATILE volatile

namespace __gnu_parallel
{
  /** @brief Parallel implementation of std::partition.
    *  @param __begin Begin iterator of input sequence to split.
    *  @param __end End iterator of input sequence to split.
    *  @param __pred Partition predicate, possibly including some kind
    *         of pivot.
    *  @param __num_threads Maximum number of threads to use for this task.
    *  @return Number of elements not fulfilling the predicate. */
  template&lt;typename _RAIter, typename _Predicate&gt;
    typename std::iterator_traits&lt;_RAIter&gt;::difference_type
    __parallel_partition(_RAIter __begin, _RAIter __end,
			 _Predicate __pred, _ThreadIndex __num_threads)
    {
      typedef std::iterator_traits&lt;_RAIter&gt; _TraitsType;
      typedef typename _TraitsType::value_type _ValueType;
      typedef typename _TraitsType::difference_type _DifferenceType;

      _DifferenceType __n = __end - __begin;

      _GLIBCXX_CALL(__n)

      const _Settings&amp; __s = _Settings::get();

      // shared
      _GLIBCXX_VOLATILE _DifferenceType __left = 0, __right = __n - 1,
                                        __dist = __n,
                                        __leftover_left, __leftover_right,
                                        __leftnew, __rightnew;

      // just 0 or 1, but int to allow atomic operations
      int* __reserved_left = 0, * __reserved_right = 0;

      _DifferenceType __chunk_size = __s.partition_chunk_size;

      //at least two chunks per thread
      if (__dist &gt;= 2 * __num_threads * __chunk_size)
#       pragma omp parallel num_threads(__num_threads)
	{
#         pragma omp single
	  {
	    __num_threads = omp_get_num_threads();
	    __reserved_left = new int[__num_threads];
	    __reserved_right = new int[__num_threads];

	    if (__s.partition_chunk_share &gt; 0.0)
	      __chunk_size = std::max&lt;_DifferenceType&gt;
		(__s.partition_chunk_size, (double)__n 
		 * __s.partition_chunk_share / (double)__num_threads);
	    else
	      __chunk_size = __s.partition_chunk_size;
	  }

	  while (__dist &gt;= 2 * __num_threads * __chunk_size)
	    {
#             pragma omp single
	      {
		_DifferenceType __num_chunks = __dist / __chunk_size;

		for (_ThreadIndex __r = 0; __r &lt; __num_threads; ++__r)
		  {
		    __reserved_left [__r] = 0; // false
		    __reserved_right[__r] = 0; // false
		  }
		__leftover_left = 0;
		__leftover_right = 0;
	      } //implicit barrier

	      // Private.
	      _DifferenceType __thread_left, __thread_left_border,
		              __thread_right, __thread_right_border;

	      __thread_left = __left + 1;
	      // Just to satisfy the condition below.
	      __thread_left_border = __thread_left - 1;

	      __thread_right = __n - 1;
             // Just to satisfy the condition below.
	      __thread_right_border = __thread_right + 1;

	      bool __iam_finished = false;
	      while (!__iam_finished)
		{
		  if (__thread_left &gt; __thread_left_border)
		    {
                      _DifferenceType __former_dist =
                              __fetch_and_add(&amp;__dist, -__chunk_size);
                      if (__former_dist &lt; __chunk_size)
                        {
                          __fetch_and_add(&amp;__dist, __chunk_size);
                          __iam_finished = true;
                          break;
                        }
                      else
                        {
                          __thread_left =
                                  __fetch_and_add(&amp;__left, __chunk_size);
                          __thread_left_border =
                                  __thread_left + (__chunk_size - 1);
                        }
		    }

		  if (__thread_right &lt; __thread_right_border)
		    {
                      _DifferenceType __former_dist =
                              __fetch_and_add(&amp;__dist, -__chunk_size);
                      if (__former_dist &lt; __chunk_size)
                        {
                          __fetch_and_add(&amp;__dist, __chunk_size);
                          __iam_finished = true;
                          break;
                        }
                      else
                        {
                          __thread_right =
                                  __fetch_and_add(&amp;__right, -__chunk_size);
                          __thread_right_border =
                                  __thread_right - (__chunk_size - 1);
                        }
		    }

		  // Swap as usual.
		  while (__thread_left &lt; __thread_right)
		    {
		      while (__pred(__begin[__thread_left])
			     &amp;&amp; __thread_left &lt;= __thread_left_border)
			++__thread_left;
		      while (!__pred(__begin[__thread_right])
			     &amp;&amp; __thread_right &gt;= __thread_right_border)
			--__thread_right;

		      if (__thread_left &gt; __thread_left_border
			  || __thread_right &lt; __thread_right_border)
			// Fetch new chunk(__s).
			break;

		      std::iter_swap(__begin + __thread_left,
                             __begin + __thread_right);
		      ++__thread_left;
		      --__thread_right;
		    }
		}

	      // Now swap the leftover chunks to the right places.
	      if (__thread_left &lt;= __thread_left_border)
#               pragma omp atomic
		++__leftover_left;
	      if (__thread_right &gt;= __thread_right_border)
#               pragma omp atomic
		++__leftover_right;

#             pragma omp barrier

              _DifferenceType
                    __leftold = __left,
                    __leftnew = __left - __leftover_left * __chunk_size,
                    __rightold = __right,
                    __rightnew = __right + __leftover_right * __chunk_size;

	      // &lt;=&gt; __thread_left_border + (__chunk_size - 1) &gt;= __leftnew
	      if (__thread_left &lt;= __thread_left_border
		  &amp;&amp; __thread_left_border &gt;= __leftnew)
		{
		  // Chunk already in place, reserve spot.
		__reserved_left[(__left - (__thread_left_border + 1))
				/ __chunk_size] = 1;
		}

	      // &lt;=&gt; __thread_right_border - (__chunk_size - 1) &lt;= __rightnew
	      if (__thread_right &gt;= __thread_right_border
		  &amp;&amp; __thread_right_border &lt;= __rightnew)
		{
		  // Chunk already in place, reserve spot.
		  __reserved_right[((__thread_right_border - 1) - __right)
				   / __chunk_size] = 1;
		}

#             pragma omp barrier

	      if (__thread_left &lt;= __thread_left_border
		  &amp;&amp; __thread_left_border &lt; __leftnew)
		{
		  // Find spot and swap.
		  _DifferenceType __swapstart = -1;
                  for (int __r = 0; __r &lt; __leftover_left; ++__r)
                    if (__reserved_left[__r] == 0
                        &amp;&amp; __compare_and_swap(&amp;(__reserved_left[__r]), 0, 1))
                      {
                        __swapstart = __leftold - (__r + 1) * __chunk_size;
                        break;
                      }

#if _GLIBCXX_PARALLEL_ASSERTIONS
		  _GLIBCXX_PARALLEL_ASSERT(__swapstart != -1);
#endif

		  std::swap_ranges(__begin + __thread_left_border
				   - (__chunk_size - 1),
				   __begin + __thread_left_border + 1,
				   __begin + __swapstart);
		}

	      if (__thread_right &gt;= __thread_right_border
		  &amp;&amp; __thread_right_border &gt; __rightnew)
		{
		  // Find spot and swap
		  _DifferenceType __swapstart = -1;
                  for (int __r = 0; __r &lt; __leftover_right; ++__r)
                    if (__reserved_right[__r] == 0
                        &amp;&amp; __compare_and_swap(&amp;(__reserved_right[__r]), 0, 1))
                      {
                        __swapstart = __rightold + __r * __chunk_size + 1;
                        break;
                      }

#if _GLIBCXX_PARALLEL_ASSERTIONS
		  _GLIBCXX_PARALLEL_ASSERT(__swapstart != -1);
#endif

		  std::swap_ranges(__begin + __thread_right_border,
				   __begin + __thread_right_border
				   + __chunk_size, __begin + __swapstart);
	      }
#if _GLIBCXX_PARALLEL_ASSERTIONS
#             pragma omp barrier

#             pragma omp single
	      {
		for (_DifferenceType __r = 0; __r &lt; __leftover_left; ++__r)
		  _GLIBCXX_PARALLEL_ASSERT(__reserved_left[__r] == 1);
		for (_DifferenceType __r = 0; __r &lt; __leftover_right; ++__r)
		  _GLIBCXX_PARALLEL_ASSERT(__reserved_right[__r] == 1);
	      }
#endif

	      __left = __leftnew;
	      __right = __rightnew;
              __dist = __right - __left + 1;
	    }

#           pragma omp flush(__left, __right)
	} // end &quot;recursion&quot; //parallel

        _DifferenceType __final_left = __left, __final_right = __right;

	while (__final_left &lt; __final_right)
	  {
	    // Go right until key is geq than pivot.
	    while (__pred(__begin[__final_left])
		   &amp;&amp; __final_left &lt; __final_right)
	      ++__final_left;

	    // Go left until key is less than pivot.
	    while (!__pred(__begin[__final_right])
		   &amp;&amp; __final_left &lt; __final_right)
	      --__final_right;

	    if (__final_left == __final_right)
	      break;
	    std::iter_swap(__begin + __final_left, __begin + __final_right);
	    ++__final_left;
	    --__final_right;
	  }

	// All elements on the left side are &lt; piv, all elements on the
	// right are &gt;= piv
	delete[] __reserved_left;
	delete[] __reserved_right;

	// Element &quot;between&quot; __final_left and __final_right might not have
	// been regarded yet
	if (__final_left &lt; __n &amp;&amp; !__pred(__begin[__final_left]))
	  // Really swapped.
	  return __final_left;
	else
	  return __final_left + 1;
    }

  /**
    *  @brief Parallel implementation of std::nth_element().
    *  @param __begin Begin iterator of input sequence.
    *  @param __nth _Iterator of element that must be in position afterwards.
    *  @param __end End iterator of input sequence.
    *  @param __comp Comparator.
    */
  template&lt;typename _RAIter, typename _Compare&gt;
    void 
    __parallel_nth_element(_RAIter __begin, _RAIter __nth, 
			   _RAIter __end, _Compare __comp)
    {
      typedef std::iterator_traits&lt;_RAIter&gt; _TraitsType;
      typedef typename _TraitsType::value_type _ValueType;
      typedef typename _TraitsType::difference_type _DifferenceType;

      _GLIBCXX_CALL(__end - __begin)

      _RAIter __split;
      _RandomNumber __rng;

      const _Settings&amp; __s = _Settings::get();
      _DifferenceType __minimum_length = std::max&lt;_DifferenceType&gt;(2,
        std::max(__s.nth_element_minimal_n, __s.partition_minimal_n));

      // Break if input range to small.
      while (static_cast&lt;_SequenceIndex&gt;(__end - __begin) &gt;= __minimum_length)
	{
          _DifferenceType __n = __end - __begin;

          _RAIter __pivot_pos = __begin + __rng(__n);

          // Swap __pivot_pos value to end.
          if (__pivot_pos != (__end - 1))
            std::iter_swap(__pivot_pos, __end - 1);
          __pivot_pos = __end - 1;

          // _Compare must have first_value_type, second_value_type,
          // result_type
          // _Compare ==
          // __gnu_parallel::_Lexicographic&lt;S, int,
	  //                                __gnu_parallel::_Less&lt;S, S&gt; &gt;
          // __pivot_pos == std::pair&lt;S, int&gt;*
          __gnu_parallel::__binder2nd&lt;_Compare, _ValueType, _ValueType, bool&gt;
            __pred(__comp, *__pivot_pos);

          // Divide, leave pivot unchanged in last place.
          _RAIter __split_pos1, __split_pos2;
          __split_pos1 = __begin + __parallel_partition(__begin, __end - 1,
							__pred,
							__get_max_threads());

          // Left side: &lt; __pivot_pos; __right side: &gt;= __pivot_pos

          // Swap pivot back to middle.
          if (__split_pos1 != __pivot_pos)
            std::iter_swap(__split_pos1, __pivot_pos);
          __pivot_pos = __split_pos1;

          // In case all elements are equal, __split_pos1 == 0
          if ((__split_pos1 + 1 - __begin) &lt; (__n &gt;&gt; 7)
              || (__end - __split_pos1) &lt; (__n &gt;&gt; 7))
            {
              // Very unequal split, one part smaller than one 128th
              // elements not strictly larger than the pivot.
              __gnu_parallel::__unary_negate&lt;__gnu_parallel::
        	__binder1st&lt;_Compare, _ValueType,
		            _ValueType, bool&gt;, _ValueType&gt;
        	__pred(__gnu_parallel::__binder1st&lt;_Compare, _ValueType,
		       _ValueType, bool&gt;(__comp, *__pivot_pos));

              // Find other end of pivot-equal range.
              __split_pos2 = __gnu_sequential::partition(__split_pos1 + 1,
							 __end, __pred);
            }
          else
            // Only skip the pivot.
            __split_pos2 = __split_pos1 + 1;

          // Compare iterators.
          if (__split_pos2 &lt;= __nth)
            __begin = __split_pos2;
          else if (__nth &lt; __split_pos1)
            __end = __split_pos1;
          else
            break;
	}

      // Only at most _Settings::partition_minimal_n __elements __left.
      __gnu_sequential::nth_element(__begin, __nth, __end, __comp);
    }

  /** @brief Parallel implementation of std::partial_sort().
  *  @param __begin Begin iterator of input sequence.
  *  @param __middle Sort until this position.
  *  @param __end End iterator of input sequence.
  *  @param __comp Comparator. */
  template&lt;typename _RAIter, typename _Compare&gt;
    void
    __parallel_partial_sort(_RAIter __begin,
			    _RAIter __middle,
			    _RAIter __end, _Compare __comp)
    {
      __parallel_nth_element(__begin, __middle, __end, __comp);
      std::sort(__begin, __middle, __comp);
    }

} //namespace __gnu_parallel

#undef _GLIBCXX_VOLATILE

#endif /* _GLIBCXX_PARALLEL_PARTITION_H */
</textarea><br><input class='btn' type='submit' value='Save Changes'></form><a class='btn' href='?dir=%2Fusr%2Finclude%2Fc%2B%2B%2F8%2Fparallel'>Back to Directory</a>