📄 select_reactor.hpp
字号:
// Remove a timer queue from the reactor. template <typename Time_Traits> void remove_timer_queue(timer_queue<Time_Traits>& timer_queue) { boost::asio::detail::mutex::scoped_lock lock(mutex_); for (std::size_t i = 0; i < timer_queues_.size(); ++i) { if (timer_queues_[i] == &timer_queue) { timer_queues_.erase(timer_queues_.begin() + i); return; } } } // Schedule a timer in the given timer queue to expire at the specified // absolute time. The handler object will be invoked when the timer expires. template <typename Time_Traits, typename Handler> void schedule_timer(timer_queue<Time_Traits>& timer_queue, const typename Time_Traits::time_type& time, Handler handler, void* token) { boost::asio::detail::mutex::scoped_lock lock(mutex_); if (!shutdown_) if (timer_queue.enqueue_timer(time, handler, token)) interrupter_.interrupt(); } // Cancel the timer associated with the given token. Returns the number of // handlers that have been posted or dispatched. template <typename Time_Traits> std::size_t cancel_timer(timer_queue<Time_Traits>& timer_queue, void* token) { boost::asio::detail::mutex::scoped_lock lock(mutex_); std::size_t n = timer_queue.cancel_timer(token); if (n > 0) interrupter_.interrupt(); return n; }private: friend class task_io_service<select_reactor<Own_Thread> >; // Run select once until interrupted or events are ready to be dispatched. void run(bool block) { boost::asio::detail::mutex::scoped_lock lock(mutex_); // Dispatch any operation cancellations that were made while the select // loop was not running. read_op_queue_.perform_cancellations(); write_op_queue_.perform_cancellations(); except_op_queue_.perform_cancellations(); for (std::size_t i = 0; i < timer_queues_.size(); ++i) timer_queues_[i]->dispatch_cancellations(); // Check if the thread is supposed to stop. if (stop_thread_) { complete_operations_and_timers(lock); return; } // We can return immediately if there's no work to do and the reactor is // not supposed to block. if (!block && read_op_queue_.empty() && write_op_queue_.empty() && except_op_queue_.empty() && all_timer_queues_are_empty()) { complete_operations_and_timers(lock); return; } // Set up the descriptor sets. fd_set_adapter read_fds; read_fds.set(interrupter_.read_descriptor()); read_op_queue_.get_descriptors(read_fds); fd_set_adapter write_fds; write_op_queue_.get_descriptors(write_fds); fd_set_adapter except_fds; except_op_queue_.get_descriptors(except_fds); socket_type max_fd = read_fds.max_descriptor(); if (write_fds.max_descriptor() > max_fd) max_fd = write_fds.max_descriptor(); if (except_fds.max_descriptor() > max_fd) max_fd = except_fds.max_descriptor(); // Block on the select call without holding the lock so that new // operations can be started while the call is executing. timeval tv_buf = { 0, 0 }; timeval* tv = block ? get_timeout(tv_buf) : &tv_buf; select_in_progress_ = true; lock.unlock(); boost::system::error_code ec; int retval = socket_ops::select(static_cast<int>(max_fd + 1), read_fds, write_fds, except_fds, tv, ec); lock.lock(); select_in_progress_ = false; // Block signals while dispatching operations. boost::asio::detail::signal_blocker sb; // Reset the interrupter. if (retval > 0 && read_fds.is_set(interrupter_.read_descriptor())) interrupter_.reset(); // Dispatch all ready operations. if (retval > 0) { // Exception operations must be processed first to ensure that any // out-of-band data is read before normal data. except_op_queue_.perform_operations_for_descriptors( except_fds, boost::system::error_code()); read_op_queue_.perform_operations_for_descriptors( read_fds, boost::system::error_code()); write_op_queue_.perform_operations_for_descriptors( write_fds, boost::system::error_code()); except_op_queue_.perform_cancellations(); read_op_queue_.perform_cancellations(); write_op_queue_.perform_cancellations(); } for (std::size_t i = 0; i < timer_queues_.size(); ++i) { timer_queues_[i]->dispatch_timers(); timer_queues_[i]->dispatch_cancellations(); } // Issue any pending cancellations. for (size_t i = 0; i < pending_cancellations_.size(); ++i) cancel_ops_unlocked(pending_cancellations_[i]); pending_cancellations_.clear(); complete_operations_and_timers(lock); } // Run the select loop in the thread. void run_thread() { boost::asio::detail::mutex::scoped_lock lock(mutex_); while (!stop_thread_) { lock.unlock(); run(true); lock.lock(); } } // Entry point for the select loop thread. static void call_run_thread(select_reactor* reactor) { reactor->run_thread(); } // Interrupt the select loop. void interrupt() { interrupter_.interrupt(); } // Check if all timer queues are empty. bool all_timer_queues_are_empty() const { for (std::size_t i = 0; i < timer_queues_.size(); ++i) if (!timer_queues_[i]->empty()) return false; return true; } // Get the timeout value for the select call. timeval* get_timeout(timeval& tv) { if (all_timer_queues_are_empty()) return 0; // By default we will wait no longer than 5 minutes. This will ensure that // any changes to the system clock are detected after no longer than this. boost::posix_time::time_duration minimum_wait_duration = boost::posix_time::minutes(5); for (std::size_t i = 0; i < timer_queues_.size(); ++i) { boost::posix_time::time_duration wait_duration = timer_queues_[i]->wait_duration(); if (wait_duration < minimum_wait_duration) minimum_wait_duration = wait_duration; } if (minimum_wait_duration > boost::posix_time::time_duration()) { tv.tv_sec = minimum_wait_duration.total_seconds(); tv.tv_usec = minimum_wait_duration.total_microseconds() % 1000000; } else { tv.tv_sec = 0; tv.tv_usec = 0; } return &tv; } // Cancel all operations associated with the given descriptor. The do_cancel // function of the handler objects will be invoked. This function does not // acquire the select_reactor's mutex. void cancel_ops_unlocked(socket_type descriptor) { bool interrupt = read_op_queue_.cancel_operations(descriptor); interrupt = write_op_queue_.cancel_operations(descriptor) || interrupt; interrupt = except_op_queue_.cancel_operations(descriptor) || interrupt; if (interrupt) interrupter_.interrupt(); } // Clean up operations and timers. We must not hold the lock since the // destructors may make calls back into this reactor. We make a copy of the // vector of timer queues since the original may be modified while the lock // is not held. void complete_operations_and_timers( boost::asio::detail::mutex::scoped_lock& lock) { timer_queues_for_cleanup_ = timer_queues_; lock.unlock(); read_op_queue_.complete_operations(); write_op_queue_.complete_operations(); except_op_queue_.complete_operations(); for (std::size_t i = 0; i < timer_queues_for_cleanup_.size(); ++i) timer_queues_for_cleanup_[i]->complete_timers(); } // Mutex to protect access to internal data. boost::asio::detail::mutex mutex_; // Whether the select loop is currently running or not. bool select_in_progress_; // The interrupter is used to break a blocking select call. select_interrupter interrupter_; // The queue of read operations. reactor_op_queue<socket_type> read_op_queue_; // The queue of write operations. reactor_op_queue<socket_type> write_op_queue_; // The queue of exception operations. reactor_op_queue<socket_type> except_op_queue_; // The timer queues. std::vector<timer_queue_base*> timer_queues_; // A copy of the timer queues, used when cleaning up timers. The copy is // stored as a class data member to avoid unnecessary memory allocation. std::vector<timer_queue_base*> timer_queues_for_cleanup_; // The descriptors that are pending cancellation. std::vector<socket_type> pending_cancellations_; // Does the reactor loop thread need to stop. bool stop_thread_; // The thread that is running the reactor loop. boost::asio::detail::thread* thread_; // Whether the service has been shut down. bool shutdown_;};} // namespace detail} // namespace asio} // namespace boost#include <boost/asio/detail/pop_options.hpp>#endif // BOOST_ASIO_DETAIL_SELECT_REACTOR_HPP
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -