网站幕布拍照什么样子的,多语言网站建设方案,文昌网站建设 myvodo,做网站用那个浏览器此头文件是线程支持库的一部分。
类模板 std::barrier 提供一种线程协调机制#xff0c;阻塞已知大小的线程组直至该组中的所有线程到达该屏障。不同于 std::latch#xff0c;屏障是可重用的#xff1a;一旦到达的线程组被解除阻塞#xff0c;即可重用同一屏障。与 std::l…此头文件是线程支持库的一部分。
类模板 std::barrier 提供一种线程协调机制阻塞已知大小的线程组直至该组中的所有线程到达该屏障。不同于 std::latch屏障是可重用的一旦到达的线程组被解除阻塞即可重用同一屏障。与 std::latch 不同会在线程解除阻塞前执行一个可能为空的可调用对象。
屏障对象的生存期由一个或多个屏障阶段组成。每个阶段定义一个阻塞线程的阶段同步点。线程可以抵达屏障但通过调用 arrive 来推迟它在阶段同步点上的等待。这样的线程可以随后再通过调用 wait 在阶段同步点上阻塞。
屏障 阶段 由以下步骤组成
每次调用 arrive 或 arrive_and_drop 减少期待计数。期待计数抵达零时运行阶段完成步骤即调用 completion并解除所有在阶段同步点上阻塞的线程。完成步骤的结束强先发生于所有从完成步骤所除阻的调用的返回。 在期待计数抵达零后一个线程会在其调用 arrive、arrive_and_drop 或 wait 的过程中执行完成步骤恰好一次但如果没有线程调用 wait 则是否执行完成步骤为实现定义。完成步骤结束时重置期待计数为构造中指定的值它可能为 arrive_and_drop 调用所调整自此开始下一阶段。
并发调用barrier 除了析构函数外的成员函数不会引起数据竞争。
vs2022 类模板如下
class barrier {
public:static_assert(
#ifndef __cpp_noexcept_function_typeis_function_vremove_pointer_t_Completion_function ||
#endif // !defined(__cpp_noexcept_function_type)is_nothrow_invocable_v_Completion_function,N4950 [thread.barrier.class]/5: is_nothrow_invocable_vCompletionFunction shall be true);using arrival_token _Arrival_token_Completion_function;constexpr explicit barrier(const ptrdiff_t _Expected, _Completion_function _Fn _Completion_function()) noexcept /* strengthened */: _Val(_One_then_variadic_args_t{}, _STD move(_Fn), _Expected _Barrier_value_shift) {_STL_VERIFY(_Expected 0 _Expected (max) (),Precondition: expected 0 and expected max() (N4950 [thread.barrier.class]/9));}barrier(const barrier) delete;barrier operator(const barrier) delete;_NODISCARD static constexpr ptrdiff_t(max)() noexcept {return _Barrier_max;}_NODISCARD_BARRIER_TOKEN arrival_token arrive(ptrdiff_t _Update 1) noexcept /* strengthened */ {_STL_VERIFY(_Update 0 _Update (max) (), Precondition: update 0 (N4950 [thread.barrier.class]/12));_Update _Barrier_value_shift;// TRANSITION, GH-1133: should be memory_order_releaseptrdiff_t _Current _Val._Myval2._Current.fetch_sub(_Update) - _Update;_STL_VERIFY(_Current 0, Precondition: update is less than or equal to the expected count for the current barrier phase (N4950 [thread.barrier.class]/12));if ((_Current _Barrier_value_mask) 0) {// TRANSITION, GH-1133: should have this fence:// atomic_thread_fence(memory_order_acquire);_Completion(_Current);}// Embedding this into the token to provide an additional correctness check that the token is from the same// barrier and wasnt used. All bits of this fit, as barrier should be aligned to at least the size of an// atomic counter.return arrival_token{(_Current _Barrier_arrival_token_mask) | reinterpret_castintptr_t(this)};}void wait(arrival_token _Arrival) const noexcept /* strengthened */ {_STL_VERIFY((_Arrival._Value _Barrier_value_mask) reinterpret_castintptr_t(this),Preconditions: arrival is associated with the phase synchronization point for the current phase or the immediately preceding phase of the same barrier object (N4950 [thread.barrier.class]/19));const ptrdiff_t _Arrival_value _Arrival._Value _Barrier_arrival_token_mask;_Arrival._Value _Barrier_invalid_token;for (;;) {// TRANSITION, GH-1133: should be memory_order_acquireconst ptrdiff_t _Current _Val._Myval2._Current.load();_STL_VERIFY(_Current 0, Invariant counter 0, possibly caused by preconditions violation (N4950 [thread.barrier.class]/12));if ((_Current _Barrier_arrival_token_mask) ! _Arrival_value) {break;}_Val._Myval2._Current.wait(_Current, memory_order_relaxed);}}void arrive_and_wait() noexcept /* strengthened */ {// TRANSITION, GH-1133: should be memory_order_acq_relptrdiff_t _Current _Val._Myval2._Current.fetch_sub(_Barrier_value_step) - _Barrier_value_step;const ptrdiff_t _Arrival _Current _Barrier_arrival_token_mask;_STL_VERIFY(_Current 0, Precondition: update is less than or equal to the expected count for the current barrier phase (N4950 [thread.barrier.class]/12));if ((_Current _Barrier_value_mask) 0) {_Completion(_Current);return;}for (;;) {_Val._Myval2._Current.wait(_Current, memory_order_relaxed);// TRANSITION, GH-1133: should be memory_order_acquire_Current _Val._Myval2._Current.load();_STL_VERIFY(_Current 0, Invariant counter 0, possibly caused by preconditions violation (N4950 [thread.barrier.class]/12));if ((_Current _Barrier_arrival_token_mask) ! _Arrival) {break;}}}void arrive_and_drop() noexcept /* strengthened */ {const ptrdiff_t _Rem_count _Val._Myval2._Total.fetch_sub(_Barrier_value_step, memory_order_relaxed) - _Barrier_value_step;_STL_VERIFY(_Rem_count 0, Precondition: The expected count for the current barrier phase is greater than zero (N4950 [thread.barrier.class]/24) (checked initial expected count, which is not less than the current));(void) arrive(1);}private:void _Completion(const ptrdiff_t _Current) noexcept {const ptrdiff_t _Rem_count _Val._Myval2._Total.load(memory_order_relaxed);_STL_VERIFY(_Rem_count 0, Invariant: initial expected count less than zero, possibly caused by preconditions violation (N4950 [thread.barrier.class]/24));_Val._Get_first()();const ptrdiff_t _New_phase_count _Rem_count | ((_Current 1) _Barrier_arrival_token_mask);// TRANSITION, GH-1133: should be memory_order_release_Val._Myval2._Current.store(_New_phase_count);_Val._Myval2._Current.notify_all();}struct _Counter_t {constexpr explicit _Counter_t(ptrdiff_t _Initial) : _Current(_Initial), _Total(_Initial) {}// wait(arrival_token) accepts a token from the current phase or the immediately preceding phase; this means// we can track which phase is the current phase using 1 bit which alternates between each phase. For this// purpose we use the low order bit of _Current.atomicptrdiff_t _Current;atomicptrdiff_t _Total;};_Compressed_pair_Completion_function, _Counter_t _Val;
};
成员对象
名称定义completion (私有)CompletionFunction 类型的完成函数对象在每个阶段完成步骤调用。 (仅用于阐述的成员对象*)
成员类型
名称定义arrival_token未指定的对象类型满足可移动构造 (MoveConstructible) 、可移动赋值 (MoveAssignable) 及可析构 (Destructible)
成员函数 (构造函数) 构造 barrier (公开成员函数) (析构函数) 销毁 barrier (公开成员函数) operator [弃置] barrier 不可赋值 (公开成员函数) arrive 到达屏障并减少期待计数 (公开成员函数) wait 在阶段同步点阻塞直至运行其阶段完成步骤 (公开成员函数) arrive_and_wait 到达屏障并把期待计数减少一然后阻塞直至当前阶段完成 (公开成员函数) arrive_and_drop 将后继阶段的初始期待计数和当前阶段的期待计数均减少一 (公开成员函数) 常量 max [静态] 实现所支持的期待计数的最大值 (公开静态成员函数)
示例代码
#include barrier
#include iostream
#include string
#include syncstream
#include thread
#include vectorint main()
{const auto workers { Anil, Busara, Carl };auto on_completion []() noexcept{// 此处无需锁定static auto phase ... 完成\n清理...\n;std::cout phase;phase ... 完成\n;};std::barrier sync_point(std::ssize(workers), on_completion);auto work [](std::string name){std::string product name 已工作\n;std::osyncstream(std::cout) product; // OK, op 的调用是原子的sync_point.arrive_and_wait();product name 已清理\n;std::osyncstream(std::cout) product;sync_point.arrive_and_wait();};std::cout 启动...\n;std::vectorstd::jthread threads;threads.reserve(std::size(workers));for (auto const worker : workers)threads.emplace_back(work, worker);return 0;
}
运行结果 另一个示例
#include barrier
#include iostream
#include string
#include syncstream
#include thread
#include vectorstd::barrier bar(4); //创建一个barrier需要4个线程到达同步点void thread_func(int id) {// 线程执行一些任务//std::cout Thread ID: std::this_thread::get_id() is doing some work. std::endl;std::cout Thread IDstart is doing some work.\n;// 等待所有线程到达栅栏bar.arrive_and_wait();// 所有线程到达栅栏后继续执行后续任务//std::cout Thread ID: std::this_thread::get_id() continues working after barrier. std::endl;std::cout Thread IDend: continues working after barrier.\n;
}int main() {constexpr int num_threads 4;std::vectorstd::thread threads;// 创建线程并执行线程函数for (int i 0; i num_threads; i) {threads.emplace_back(thread_func, i);}// 等待所有线程执行完毕for (auto t : threads) {t.join();}return 0;
}示例代码
#include barrier
#include iostream
#include string
#include syncstream
#include thread
#include vectorstd::barrier bar(4); //创建一个barrier需要4个线程到达同步点void thread_func(int id) {// 线程执行一些任务//std::cout Thread ID: std::this_thread::get_id() is doing some work. std::endl;std::cout Thread IDstart is doing some work.\n;// 等待所有线程到达栅栏bar.arrive_and_wait();// 所有线程到达栅栏后继续执行后续任务//std::cout Thread ID: std::this_thread::get_id() continues working after barrier. std::endl;std::cout Thread IDend: continues working after barrier.\n;
}int main() {constexpr int num_threads 4;std::vectorstd::thread threads;// 创建线程并执行线程函数for (int i 0; i num_threads; i) {threads.emplace_back(thread_func, i);}// 等待所有线程执行完毕for (auto t : threads) {t.join();}return 0;
}
运行结果 参考
std::barrier - cppreference.com
【C 20 并发工具 std::barrier】掌握并发编程深入理解C的std::barrier_c barrier-CSDN博客
