Subversion Repositories Kolibri OS

Rev

Blame | Last modification | View Log | RSS feed

  1. // -*- C++ -*-
  2.  
  3. // Copyright (C) 2007-2015 Free Software Foundation, Inc.
  4. //
  5. // This file is part of the GNU ISO C++ Library.  This library is free
  6. // software; you can redistribute it and/or modify it under the terms
  7. // of the GNU General Public License as published by the Free Software
  8. // Foundation; either version 3, or (at your option) any later
  9. // version.
  10.  
  11. // This library is distributed in the hope that it will be useful, but
  12. // WITHOUT ANY WARRANTY; without even the implied warranty of
  13. // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14. // General Public License for more details.
  15.  
  16. // Under Section 7 of GPL version 3, you are granted additional
  17. // permissions described in the GCC Runtime Library Exception, version
  18. // 3.1, as published by the Free Software Foundation.
  19.  
  20. // You should have received a copy of the GNU General Public License and
  21. // a copy of the GCC Runtime Library Exception along with this program;
  22. // see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
  23. // <http://www.gnu.org/licenses/>.
  24.  
  25. /** @file parallel/par_loop.h
  26.  *  @brief Parallelization of embarrassingly parallel execution by
  27.  *  means of equal splitting.
  28.  *  This file is a GNU parallel extension to the Standard C++ Library.
  29.  */
  30.  
  31. // Written by Felix Putze.
  32.  
  33. #ifndef _GLIBCXX_PARALLEL_PAR_LOOP_H
  34. #define _GLIBCXX_PARALLEL_PAR_LOOP_H 1
  35.  
  36. #include <omp.h>
  37. #include <parallel/settings.h>
  38. #include <parallel/base.h>
  39. #include <parallel/equally_split.h>
  40.  
  41. namespace __gnu_parallel
  42. {
  43.   /** @brief Embarrassingly parallel algorithm for random access
  44.    * iterators, using hand-crafted parallelization by equal splitting
  45.    * the work.
  46.    *
  47.    *  @param __begin Begin iterator of element sequence.
  48.    *  @param __end End iterator of element sequence.
  49.    *  @param __o User-supplied functor (comparator, predicate, adding
  50.    *  functor, ...)
  51.    *  @param __f Functor to "process" an element with __op (depends on
  52.    *  desired functionality, e. g. for std::for_each(), ...).
  53.    *  @param __r Functor to "add" a single __result to the already
  54.    *  processed elements (depends on functionality).
  55.    *  @param __base Base value for reduction.
  56.    *  @param __output Pointer to position where final result is written to
  57.    *  @param __bound Maximum number of elements processed (e. g. for
  58.    *  std::count_n()).
  59.    *  @return User-supplied functor (that may contain a part of the result).
  60.    */
  61.   template<typename _RAIter,
  62.            typename _Op,
  63.            typename _Fu,
  64.            typename _Red,
  65.            typename _Result>
  66.     _Op
  67.     __for_each_template_random_access_ed(_RAIter __begin, _RAIter __end,
  68.                                          _Op __o, _Fu& __f, _Red __r,
  69.                                          _Result __base, _Result& __output,
  70.       typename std::iterator_traits<_RAIter>::difference_type __bound)
  71.     {
  72.       typedef std::iterator_traits<_RAIter> _TraitsType;
  73.       typedef typename _TraitsType::difference_type _DifferenceType;
  74.       const _DifferenceType __length = __end - __begin;
  75.       _Result *__thread_results;
  76.       bool* __constructed;
  77.  
  78.       _ThreadIndex __num_threads = __gnu_parallel::min<_DifferenceType>
  79.         (__get_max_threads(), __length);
  80.  
  81. #     pragma omp parallel num_threads(__num_threads)
  82.       {
  83. #       pragma omp single
  84.         {
  85.           __num_threads = omp_get_num_threads();
  86.           __thread_results = static_cast<_Result*>
  87.             (::operator new(__num_threads * sizeof(_Result)));
  88.           __constructed = new bool[__num_threads];
  89.         }
  90.  
  91.         _ThreadIndex __iam = omp_get_thread_num();
  92.  
  93.         // Neutral element.
  94.         _Result* __reduct;
  95.  
  96.         _DifferenceType
  97.           __start = __equally_split_point(__length, __num_threads, __iam),
  98.           __stop = __equally_split_point(__length, __num_threads, __iam + 1);
  99.  
  100.         if (__start < __stop)
  101.           {
  102.             __reduct = new _Result(__f(__o, __begin + __start));
  103.             ++__start;
  104.             __constructed[__iam] = true;
  105.           }
  106.         else
  107.           __constructed[__iam] = false;
  108.  
  109.         for (; __start < __stop; ++__start)
  110.           *__reduct = __r(*__reduct, __f(__o, __begin + __start));
  111.  
  112.         if (__constructed[__iam])
  113.           {
  114.             ::new(&__thread_results[__iam]) _Result(*__reduct);
  115.             delete __reduct;
  116.           }
  117.       } //parallel
  118.  
  119.       for (_ThreadIndex __i = 0; __i < __num_threads; ++__i)
  120.         if (__constructed[__i])
  121.           {
  122.             __output = __r(__output, __thread_results[__i]);
  123.             __thread_results[__i].~_Result();
  124.           }
  125.  
  126.       // Points to last element processed (needed as return value for
  127.       // some algorithms like transform).
  128.       __f._M_finish_iterator = __begin + __length;
  129.  
  130.       ::operator delete(__thread_results);
  131.  
  132.       delete[] __constructed;
  133.  
  134.       return __o;
  135.     }
  136.  
  137. } // end namespace
  138.  
  139. #endif /* _GLIBCXX_PARALLEL_PAR_LOOP_H */
  140.