jmmcg::ppd::thread_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P > Class Template Reference
This compatibility-pool has an unlimited size, and uses a master to execute the work. More...
#include <thread_pool_sequential.hpp>
Inheritance diagram for jmmcg::ppd::thread_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P >:
Collaboration diagram for jmmcg::ppd::thread_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P >:
Public Types | |
| typedef private_::sequential_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P > | base_t |
| typedef base_t::api_params_type | api_params_type |
 Public Types inherited from jmmcg::ppd::private_::sequential_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P > | |
| using | pool_traits_type = P |
| using | pool_type = std::vector< fake_thread * > |
| A hack to give us a size_type member-typedef. More... | |
| using | work_distribution_mode = pool_traits::work_distribution_mode_t::one_thread_distributes<> |
| using | signalled_work_queue_type = typename pool_traits_type::template signalled_work_queue_type< typename work_distribution_mode::queue_model > |
| using | queue_size_type = typename signalled_work_queue_type::size_type |
| typedef pool_type::size_type | pool_size_type |
| using | statistics_type = typename pool_traits_type::template statistics_type< typename work_distribution_mode::queue_model > |
| Just stub statistics - to support the interface. No actual statistics are collected. More... | |
| typedef pool_traits_type::cfg_type | cfg_type |
| The type of the control-flow graph that will be generated at run-time, if supported. More... | |
| typedef pool_traits_type::os_traits | os_traits |
| A useful typedef to easily get to the various OS traits. More... | |
| typedef os_traits::exception_type | exception_type |
| typedef os_traits::thread_traits | thread_traits |
| A useful typedef to easily get to the various OS specific thread-traits. More... | |
| typedef thread_traits::api_params_type | api_params_type |
| A useful typedef to easily get to the various API details. More... | |
| typedef api_params_type::priority_type | priority_type |
| A useful typedef to easily get to the various priorities. More... | |
| typedef private_::nonjoinable_t< sequential_pool > | nonjoinable |
| A useful typedef to easily get to the nonjoinable grammar element. More... | |
| typedef private_::joinable_t< sequential_pool > | joinable |
| A useful typedef to easily get to the joinable grammar element. More... | |
| typedef private_::nonjoinable_buff_t< sequential_pool > | nonjoinable_buff |
| A useful typedef to easily get to the nonjoinable_buff grammar element. More... | |
| typedef execution_context_stack< void_work > | void_exec_cxt |
| typedef private_::stl_functor_result_type< bool > | boolean_result_type |
Additional Inherited Members | |
| Public Member Functions inherited from jmmcg::ppd::private_::sequential_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P > | |
| constexpr __stdcall | sequential_pool () noexcept(true) FORCE_INLINE |
| __stdcall | sequential_pool (pool_size_type sz) noexcept(true) FORCE_INLINE |
| sequential_pool (sequential_pool const &)=delete | |
| __stdcall | ~sequential_pool () FORCE_INLINE |
| constexpr pool_size_type __fastcall | pool_size () const noexcept(true) FORCE_INLINE |
| A stub just for compatibility. Always returns "0". More... | |
| const statistics_type &__fastcall | statistics () const noexcept(true) FORCE_INLINE |
| A stub just for compatibility. More... | |
| parallel_algorithm< algo_hack_stack_t > __fastcall FORCE_INLINE | for_each (Colln const &c, Fn const &fn) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< count_if_t< Colln, Pred > > __fastcall FORCE_INLINE | count_if (Colln &c, Pred const &p) const |
| parallel_algorithm< count_t< Colln > > __fastcall FORCE_INLINE | count (Colln &c, typename Colln::value_type const &v) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< find_if_t< Colln, Pred > > __fastcall FORCE_INLINE | find_if (Colln const &c, Pred const &p) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< find_t< Colln > > __fastcall FORCE_INLINE | find (Colln const &c, typename Colln::value_type const &v) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | transform (CollnIn const &in, CollnOut &out, UniOp const &op) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | transform (CollnIn1 const &in1, CollnIn2 const &in2, CollnOut &out, BinOp const &op) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | copy (CollnIn const &in, CollnOut &out) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< accumulate_op_processor< Colln, BinOp > > __fastcall FORCE_INLINE | accumulate (Colln const &c, typename BinOp::result_type const &v, BinOp const &binop) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< accumulate_processor< Colln, V > > __fastcall FORCE_INLINE | accumulate (Colln const &c, V const &v) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | fill_n (Colln &c, typename Colln::size_type sz, typename Colln::value_type const &v) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | fill (Colln &c, typename Colln::value_type const &v) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | reverse (Colln &c) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< max_element_t< Colln, Comp > > __fastcall FORCE_INLINE | max_element (Colln const &c, Comp const &comp) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< max_element_t< Colln, std::less< typename Colln::value_type > > > __fastcall FORCE_INLINE | max_element (Colln const &c) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< min_element_t< Colln, Comp > > __fastcall FORCE_INLINE | min_element (Colln const &c, Comp const &comp) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< min_element_t< Colln, std::less< typename Colln::value_type > > > __fastcall FORCE_INLINE | min_element (Colln const &c) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | sort (Colln &c, Compare const &comp) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | sort (Colln &c) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | merge (CollnIn1 const &in1, CollnIn2 const &in2, CollnOut &out, Compare const &comp) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| parallel_algorithm< algo_hack_stack_t > FORCE_INLINE | merge (CollnIn1 const &in1, CollnIn2 const &in2, CollnOut &out) const |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| execution_context_stack< typename private_::unary_fun_work_type< ArgT, UniFn, sequential_pool > > FORCE_INLINE | unary_fun (ArgT &&a, UniFn const &op=UniFn()) |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| execution_context_stack< typename private_::binary_fun_work_type< LHSArg, RHSArg, BinFn, sequential_pool > > FORCE_INLINE | binary_fun (LHSArg &&lhs, RHSArg &&rhs, BinFn const &op=BinFn()) |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| execution_context_stack< typename private_::binary_fun_work_type< T, T, std::logical_and< bool >, sequential_pool > > FORCE_INLINE | logical_and (T &&lhs, T &&rhs) |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| execution_context_stack< typename private_::binary_fun_work_type< T, T, std::logical_or< bool >, sequential_pool > > FORCE_INLINE | logical_or (T &&lhs, T &&rhs) |
| This just forwards to the STL algorithm of the same name, for compatibility with the parallel version. More... | |
| nonjoinable __fastcall | operator<< (nonjoinable &&nj) noexcept(true) FORCE_INLINE |
| Non-joinably transfer the closure_base-derived closure into the thread_pool. More... | |
| nonjoinable_buff __fastcall | operator<< (nonjoinable_buff &&njb) noexcept(true) FORCE_INLINE |
| Non-joinably transfer the closure_base-derived closure into the thread_pool. More... | |
| joinable_t< sequential_pool > __fastcall | operator<< (joinable_t< sequential_pool > &&j) noexcept(true) FORCE_INLINE |
| Joinably transfer the closure_base-derived closure into the thread_pool. More... | |
| cfg_type & | cfg () noexcept(true) FORCE_INLINE |
| Access the control-flow graph, if supported. More... | |
| const cfg_type & | cfg () const noexcept(true) FORCE_INLINE |
| Access the control-flow graph, if supported. More... | |
| Static Public Member Functions inherited from jmmcg::ppd::private_::sequential_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P > | |
| static constexpr bool __fastcall | pool_empty () noexcept(true) FORCE_INLINE |
| A stub just for compatibility. Always returns "true". More... | |
| static constexpr bool __fastcall | queue_empty () noexcept(true) FORCE_INLINE |
| A stub just for compatibility. Always returns "true". More... | |
| static constexpr pool_size_type __fastcall | queue_size () noexcept(true) FORCE_INLINE |
| A stub just for compatibility. Always returns "0". More... | |
| static void __fastcall | queue_clear () noexcept(true) FORCE_INLINE |
| A stub just for compatibility. Always returns "0". More... | |
| static constexpr unsigned long __fastcall | min_time (generic_traits::memory_access_modes) noexcept(true) FORCE_INLINE |
| Return the theoretical minimum time in computations required to complete the current work. More... | |
| static constexpr unsigned long __fastcall FORCE_INLINE | min_time (T) noexcept(true) |
| static constexpr unsigned long __fastcall | min_processors (generic_traits::memory_access_modes) noexcept(true) FORCE_INLINE |
| Return the theoretical minimum number of processors required to achieve the minimum computation time required to complete the current work. More... | |
| static constexpr unsigned long __fastcall FORCE_INLINE | min_processors (T) noexcept(true) |
| Static Public Attributes inherited from jmmcg::ppd::private_::sequential_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P > | |
| static const pool_traits::size_mode_t | size_mode |
Detailed Description
template<class P>
class jmmcg::ppd::thread_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P >
This compatibility-pool has an unlimited size, and uses a master to execute the work.
There is no threading at all in this pool. It tries to be as fast as directly executing the methods on the closure_base-derived closure class. The idea of this class is that you can use it to provide both single & multi-threaded implementations within your objects and code, by just changing a typedef or a trait.
Definition at line 75 of file thread_pool_sequential.hpp.
Member Typedef Documentation
◆ api_params_type
template<class P >
| typedef base_t::api_params_type jmmcg::ppd::thread_pool< pool_traits::work_distribution_mode_t::one_thread_distributes<>, pool_traits::size_mode_t::sequential, P >::api_params_type |
Definition at line 96 of file thread_pool_sequential.hpp.
◆ base_t
template<class P >
Definition at line 95 of file thread_pool_sequential.hpp.
The documentation for this class was generated from the following file:
