713 lines
22 KiB
C++
713 lines
22 KiB
C++
//
|
|
// kqueue_reactor.hpp
|
|
// ~~~~~~~~~~~~~~~~~~
|
|
//
|
|
// Copyright (c) 2003-2008 Christopher M. Kohlhoff (chris at kohlhoff dot com)
|
|
// Copyright (c) 2005 Stefan Arentz (stefan at soze dot com)
|
|
//
|
|
// Distributed under the Boost Software License, Version 1.0. (See accompanying
|
|
// file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
|
|
//
|
|
|
|
#ifndef ASIO_DETAIL_KQUEUE_REACTOR_HPP
|
|
#define ASIO_DETAIL_KQUEUE_REACTOR_HPP
|
|
|
|
#if defined(_MSC_VER) && (_MSC_VER >= 1200)
|
|
# pragma once
|
|
#endif // defined(_MSC_VER) && (_MSC_VER >= 1200)
|
|
|
|
#include "asio/detail/push_options.hpp"
|
|
|
|
#include "asio/detail/kqueue_reactor_fwd.hpp"
|
|
|
|
#if defined(ASIO_HAS_KQUEUE)
|
|
|
|
#include "asio/detail/push_options.hpp"
|
|
#include <cstddef>
|
|
#include <vector>
|
|
#include <sys/types.h>
|
|
#include <sys/event.h>
|
|
#include <sys/time.h>
|
|
#include <boost/config.hpp>
|
|
#include <boost/date_time/posix_time/posix_time_types.hpp>
|
|
#include <boost/throw_exception.hpp>
|
|
#include "asio/detail/pop_options.hpp"
|
|
|
|
#include "asio/error.hpp"
|
|
#include "asio/io_service.hpp"
|
|
#include "asio/system_error.hpp"
|
|
#include "asio/detail/bind_handler.hpp"
|
|
#include "asio/detail/mutex.hpp"
|
|
#include "asio/detail/task_io_service.hpp"
|
|
#include "asio/detail/thread.hpp"
|
|
#include "asio/detail/reactor_op_queue.hpp"
|
|
#include "asio/detail/select_interrupter.hpp"
|
|
#include "asio/detail/service_base.hpp"
|
|
#include "asio/detail/signal_blocker.hpp"
|
|
#include "asio/detail/socket_types.hpp"
|
|
#include "asio/detail/timer_queue.hpp"
|
|
|
|
// Older versions of Mac OS X may not define EV_OOBAND.
|
|
#if !defined(EV_OOBAND)
|
|
# define EV_OOBAND EV_FLAG1
|
|
#endif // !defined(EV_OOBAND)
|
|
|
|
namespace asio {
|
|
namespace detail {
|
|
|
|
template <bool Own_Thread>
|
|
class kqueue_reactor
|
|
: public asio::detail::service_base<kqueue_reactor<Own_Thread> >
|
|
{
|
|
public:
|
|
// Per-descriptor data.
|
|
struct per_descriptor_data
|
|
{
|
|
bool allow_speculative_read;
|
|
bool allow_speculative_write;
|
|
};
|
|
|
|
// Constructor.
|
|
kqueue_reactor(asio::io_service& io_service)
|
|
: asio::detail::service_base<
|
|
kqueue_reactor<Own_Thread> >(io_service),
|
|
mutex_(),
|
|
kqueue_fd_(do_kqueue_create()),
|
|
wait_in_progress_(false),
|
|
interrupter_(),
|
|
read_op_queue_(),
|
|
write_op_queue_(),
|
|
except_op_queue_(),
|
|
pending_cancellations_(),
|
|
stop_thread_(false),
|
|
thread_(0),
|
|
shutdown_(false),
|
|
need_kqueue_wait_(true)
|
|
{
|
|
// Start the reactor's internal thread only if needed.
|
|
if (Own_Thread)
|
|
{
|
|
asio::detail::signal_blocker sb;
|
|
thread_ = new asio::detail::thread(
|
|
bind_handler(&kqueue_reactor::call_run_thread, this));
|
|
}
|
|
|
|
// Add the interrupter's descriptor to the kqueue.
|
|
struct kevent event;
|
|
EV_SET(&event, interrupter_.read_descriptor(),
|
|
EVFILT_READ, EV_ADD, 0, 0, 0);
|
|
::kevent(kqueue_fd_, &event, 1, 0, 0, 0);
|
|
}
|
|
|
|
// Destructor.
|
|
~kqueue_reactor()
|
|
{
|
|
shutdown_service();
|
|
close(kqueue_fd_);
|
|
}
|
|
|
|
// Destroy all user-defined handler objects owned by the service.
|
|
void shutdown_service()
|
|
{
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
shutdown_ = true;
|
|
stop_thread_ = true;
|
|
lock.unlock();
|
|
|
|
if (thread_)
|
|
{
|
|
interrupter_.interrupt();
|
|
thread_->join();
|
|
delete thread_;
|
|
thread_ = 0;
|
|
}
|
|
|
|
read_op_queue_.destroy_operations();
|
|
write_op_queue_.destroy_operations();
|
|
except_op_queue_.destroy_operations();
|
|
|
|
for (std::size_t i = 0; i < timer_queues_.size(); ++i)
|
|
timer_queues_[i]->destroy_timers();
|
|
timer_queues_.clear();
|
|
}
|
|
|
|
// Initialise the task, but only if the reactor is not in its own thread.
|
|
void init_task()
|
|
{
|
|
if (!Own_Thread)
|
|
{
|
|
typedef task_io_service<kqueue_reactor<Own_Thread> > task_io_service_type;
|
|
use_service<task_io_service_type>(this->get_io_service()).init_task();
|
|
}
|
|
}
|
|
|
|
// Register a socket with the reactor. Returns 0 on success, system error
|
|
// code on failure.
|
|
int register_descriptor(socket_type, per_descriptor_data& descriptor_data)
|
|
{
|
|
descriptor_data.allow_speculative_read = true;
|
|
descriptor_data.allow_speculative_write = true;
|
|
|
|
return 0;
|
|
}
|
|
|
|
// Start a new read operation. The handler object will be invoked when the
|
|
// given descriptor is ready to be read, or an error has occurred.
|
|
template <typename Handler>
|
|
void start_read_op(socket_type descriptor,
|
|
per_descriptor_data& descriptor_data, Handler handler,
|
|
bool allow_speculative_read = true)
|
|
{
|
|
if (allow_speculative_read && descriptor_data.allow_speculative_read)
|
|
{
|
|
asio::error_code ec;
|
|
std::size_t bytes_transferred = 0;
|
|
if (handler.perform(ec, bytes_transferred))
|
|
{
|
|
handler.complete(ec, bytes_transferred);
|
|
return;
|
|
}
|
|
|
|
// We only get one shot at a speculative read in this function.
|
|
allow_speculative_read = false;
|
|
}
|
|
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
|
|
if (shutdown_)
|
|
return;
|
|
|
|
if (!allow_speculative_read)
|
|
need_kqueue_wait_ = true;
|
|
else if (!read_op_queue_.has_operation(descriptor))
|
|
{
|
|
// Speculative reads are ok as there are no queued read operations.
|
|
descriptor_data.allow_speculative_read = true;
|
|
|
|
asio::error_code ec;
|
|
std::size_t bytes_transferred = 0;
|
|
if (handler.perform(ec, bytes_transferred))
|
|
{
|
|
handler.complete(ec, bytes_transferred);
|
|
return;
|
|
}
|
|
}
|
|
|
|
// Speculative reads are not ok as there will be queued read operations.
|
|
descriptor_data.allow_speculative_read = false;
|
|
|
|
if (read_op_queue_.enqueue_operation(descriptor, handler))
|
|
{
|
|
struct kevent event;
|
|
EV_SET(&event, descriptor, EVFILT_READ, EV_ADD, 0, 0, 0);
|
|
if (::kevent(kqueue_fd_, &event, 1, 0, 0, 0) == -1)
|
|
{
|
|
asio::error_code ec(errno,
|
|
asio::error::get_system_category());
|
|
read_op_queue_.perform_all_operations(descriptor, ec);
|
|
}
|
|
}
|
|
}
|
|
|
|
// Start a new write operation. The handler object will be invoked when the
|
|
// given descriptor is ready to be written, or an error has occurred.
|
|
template <typename Handler>
|
|
void start_write_op(socket_type descriptor,
|
|
per_descriptor_data& descriptor_data, Handler handler,
|
|
bool allow_speculative_write = true)
|
|
{
|
|
if (allow_speculative_write && descriptor_data.allow_speculative_write)
|
|
{
|
|
asio::error_code ec;
|
|
std::size_t bytes_transferred = 0;
|
|
if (handler.perform(ec, bytes_transferred))
|
|
{
|
|
handler.complete(ec, bytes_transferred);
|
|
return;
|
|
}
|
|
|
|
// We only get one shot at a speculative write in this function.
|
|
allow_speculative_write = false;
|
|
}
|
|
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
|
|
if (shutdown_)
|
|
return;
|
|
|
|
if (!allow_speculative_write)
|
|
need_kqueue_wait_ = true;
|
|
else if (!write_op_queue_.has_operation(descriptor))
|
|
{
|
|
// Speculative writes are ok as there are no queued write operations.
|
|
descriptor_data.allow_speculative_write = true;
|
|
|
|
asio::error_code ec;
|
|
std::size_t bytes_transferred = 0;
|
|
if (handler.perform(ec, bytes_transferred))
|
|
{
|
|
handler.complete(ec, bytes_transferred);
|
|
return;
|
|
}
|
|
}
|
|
|
|
// Speculative writes are not ok as there will be queued write operations.
|
|
descriptor_data.allow_speculative_write = false;
|
|
|
|
if (write_op_queue_.enqueue_operation(descriptor, handler))
|
|
{
|
|
struct kevent event;
|
|
EV_SET(&event, descriptor, EVFILT_WRITE, EV_ADD, 0, 0, 0);
|
|
if (::kevent(kqueue_fd_, &event, 1, 0, 0, 0) == -1)
|
|
{
|
|
asio::error_code ec(errno,
|
|
asio::error::get_system_category());
|
|
write_op_queue_.perform_all_operations(descriptor, ec);
|
|
}
|
|
}
|
|
}
|
|
|
|
// Start a new exception operation. The handler object will be invoked when
|
|
// the given descriptor has exception information, or an error has occurred.
|
|
template <typename Handler>
|
|
void start_except_op(socket_type descriptor,
|
|
per_descriptor_data&, Handler handler)
|
|
{
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
|
|
if (shutdown_)
|
|
return;
|
|
|
|
if (except_op_queue_.enqueue_operation(descriptor, handler))
|
|
{
|
|
struct kevent event;
|
|
if (read_op_queue_.has_operation(descriptor))
|
|
EV_SET(&event, descriptor, EVFILT_READ, EV_ADD, 0, 0, 0);
|
|
else
|
|
EV_SET(&event, descriptor, EVFILT_READ, EV_ADD, EV_OOBAND, 0, 0);
|
|
if (::kevent(kqueue_fd_, &event, 1, 0, 0, 0) == -1)
|
|
{
|
|
asio::error_code ec(errno,
|
|
asio::error::get_system_category());
|
|
except_op_queue_.perform_all_operations(descriptor, ec);
|
|
}
|
|
}
|
|
}
|
|
|
|
// Start a new write operation. The handler object will be invoked when the
|
|
// given descriptor is ready to be written, or an error has occurred.
|
|
template <typename Handler>
|
|
void start_connect_op(socket_type descriptor,
|
|
per_descriptor_data& descriptor_data, Handler handler)
|
|
{
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
|
|
if (shutdown_)
|
|
return;
|
|
|
|
// Speculative writes are not ok as there will be queued write operations.
|
|
descriptor_data.allow_speculative_write = false;
|
|
|
|
if (write_op_queue_.enqueue_operation(descriptor, handler))
|
|
{
|
|
struct kevent event;
|
|
EV_SET(&event, descriptor, EVFILT_WRITE, EV_ADD, 0, 0, 0);
|
|
if (::kevent(kqueue_fd_, &event, 1, 0, 0, 0) == -1)
|
|
{
|
|
asio::error_code ec(errno,
|
|
asio::error::get_system_category());
|
|
write_op_queue_.perform_all_operations(descriptor, ec);
|
|
}
|
|
}
|
|
}
|
|
|
|
// Cancel all operations associated with the given descriptor. The
|
|
// handlers associated with the descriptor will be invoked with the
|
|
// operation_aborted error.
|
|
void cancel_ops(socket_type descriptor, per_descriptor_data&)
|
|
{
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
cancel_ops_unlocked(descriptor);
|
|
}
|
|
|
|
// Cancel any operations that are running against the descriptor and remove
|
|
// its registration from the reactor.
|
|
void close_descriptor(socket_type descriptor, per_descriptor_data&)
|
|
{
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
|
|
// Remove the descriptor from kqueue.
|
|
struct kevent event[2];
|
|
EV_SET(&event[0], descriptor, EVFILT_READ, EV_DELETE, 0, 0, 0);
|
|
EV_SET(&event[1], descriptor, EVFILT_WRITE, EV_DELETE, 0, 0, 0);
|
|
::kevent(kqueue_fd_, event, 2, 0, 0, 0);
|
|
|
|
// Cancel any outstanding operations associated with the descriptor.
|
|
cancel_ops_unlocked(descriptor);
|
|
}
|
|
|
|
// Add a new timer queue to the reactor.
|
|
template <typename Time_Traits>
|
|
void add_timer_queue(timer_queue<Time_Traits>& timer_queue)
|
|
{
|
|
asio::detail::mutex::scoped_lock lock(mutex_);
|
|
timer_queues_.push_back(&timer_queue);
|
|
}
|
|
|
|
// Remove a timer queue from the reactor.
|
|
template <typename Time_Traits>
|
|
void remove_timer_queue(timer_queue<Time_Traits>& timer_queue)
|
|
{
|
|
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)
|
|
{
|
|
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)
|
|
{
|
|
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<kqueue_reactor<Own_Thread> >;
|
|
|
|
// Run the kqueue loop.
|
|
void run(bool block)
|
|
{
|
|
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;
|
|
}
|
|
|
|
// Determine how long to block while waiting for events.
|
|
timespec timeout_buf = { 0, 0 };
|
|
timespec* timeout = block ? get_timeout(timeout_buf) : &timeout_buf;
|
|
|
|
wait_in_progress_ = true;
|
|
lock.unlock();
|
|
|
|
// Block on the kqueue descriptor.
|
|
struct kevent events[128];
|
|
int num_events = (block || need_kqueue_wait_)
|
|
? kevent(kqueue_fd_, 0, 0, events, 128, timeout)
|
|
: 0;
|
|
|
|
lock.lock();
|
|
wait_in_progress_ = false;
|
|
|
|
// Block signals while performing operations.
|
|
asio::detail::signal_blocker sb;
|
|
|
|
// Dispatch the waiting events.
|
|
for (int i = 0; i < num_events; ++i)
|
|
{
|
|
int descriptor = events[i].ident;
|
|
if (descriptor == interrupter_.read_descriptor())
|
|
{
|
|
interrupter_.reset();
|
|
}
|
|
else if (events[i].filter == EVFILT_READ)
|
|
{
|
|
// Dispatch operations associated with the descriptor.
|
|
bool more_reads = false;
|
|
bool more_except = false;
|
|
if (events[i].flags & EV_ERROR)
|
|
{
|
|
asio::error_code error(
|
|
events[i].data, asio::error::get_system_category());
|
|
except_op_queue_.perform_all_operations(descriptor, error);
|
|
read_op_queue_.perform_all_operations(descriptor, error);
|
|
}
|
|
else if (events[i].flags & EV_OOBAND)
|
|
{
|
|
asio::error_code error;
|
|
more_except = except_op_queue_.perform_operation(descriptor, error);
|
|
if (events[i].data > 0)
|
|
more_reads = read_op_queue_.perform_operation(descriptor, error);
|
|
else
|
|
more_reads = read_op_queue_.has_operation(descriptor);
|
|
}
|
|
else
|
|
{
|
|
asio::error_code error;
|
|
more_reads = read_op_queue_.perform_operation(descriptor, error);
|
|
more_except = except_op_queue_.has_operation(descriptor);
|
|
}
|
|
|
|
// Update the descriptor in the kqueue.
|
|
struct kevent event;
|
|
if (more_reads)
|
|
EV_SET(&event, descriptor, EVFILT_READ, EV_ADD, 0, 0, 0);
|
|
else if (more_except)
|
|
EV_SET(&event, descriptor, EVFILT_READ, EV_ADD, EV_OOBAND, 0, 0);
|
|
else
|
|
EV_SET(&event, descriptor, EVFILT_READ, EV_DELETE, 0, 0, 0);
|
|
if (::kevent(kqueue_fd_, &event, 1, 0, 0, 0) == -1)
|
|
{
|
|
asio::error_code error(errno,
|
|
asio::error::get_system_category());
|
|
except_op_queue_.perform_all_operations(descriptor, error);
|
|
read_op_queue_.perform_all_operations(descriptor, error);
|
|
}
|
|
}
|
|
else if (events[i].filter == EVFILT_WRITE)
|
|
{
|
|
// Dispatch operations associated with the descriptor.
|
|
bool more_writes = false;
|
|
if (events[i].flags & EV_ERROR)
|
|
{
|
|
asio::error_code error(
|
|
events[i].data, asio::error::get_system_category());
|
|
write_op_queue_.perform_all_operations(descriptor, error);
|
|
}
|
|
else
|
|
{
|
|
asio::error_code error;
|
|
more_writes = write_op_queue_.perform_operation(descriptor, error);
|
|
}
|
|
|
|
// Update the descriptor in the kqueue.
|
|
struct kevent event;
|
|
if (more_writes)
|
|
EV_SET(&event, descriptor, EVFILT_WRITE, EV_ADD, 0, 0, 0);
|
|
else
|
|
EV_SET(&event, descriptor, EVFILT_WRITE, EV_DELETE, 0, 0, 0);
|
|
if (::kevent(kqueue_fd_, &event, 1, 0, 0, 0) == -1)
|
|
{
|
|
asio::error_code error(errno,
|
|
asio::error::get_system_category());
|
|
write_op_queue_.perform_all_operations(descriptor, error);
|
|
}
|
|
}
|
|
}
|
|
|
|
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_timers();
|
|
timer_queues_[i]->dispatch_cancellations();
|
|
}
|
|
|
|
// Issue any pending cancellations.
|
|
for (std::size_t i = 0; i < pending_cancellations_.size(); ++i)
|
|
cancel_ops_unlocked(pending_cancellations_[i]);
|
|
pending_cancellations_.clear();
|
|
|
|
// Determine whether kqueue needs to be called next time the reactor is run.
|
|
need_kqueue_wait_ = !read_op_queue_.empty()
|
|
|| !write_op_queue_.empty() || !except_op_queue_.empty();
|
|
|
|
complete_operations_and_timers(lock);
|
|
}
|
|
|
|
// Run the select loop in the thread.
|
|
void run_thread()
|
|
{
|
|
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(kqueue_reactor* reactor)
|
|
{
|
|
reactor->run_thread();
|
|
}
|
|
|
|
// Interrupt the select loop.
|
|
void interrupt()
|
|
{
|
|
interrupter_.interrupt();
|
|
}
|
|
|
|
// Create the kqueue file descriptor. Throws an exception if the descriptor
|
|
// cannot be created.
|
|
static int do_kqueue_create()
|
|
{
|
|
int fd = kqueue();
|
|
if (fd == -1)
|
|
{
|
|
boost::throw_exception(
|
|
asio::system_error(
|
|
asio::error_code(errno,
|
|
asio::error::get_system_category()),
|
|
"kqueue"));
|
|
}
|
|
return fd;
|
|
}
|
|
|
|
// 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 kevent call.
|
|
timespec* get_timeout(timespec& ts)
|
|
{
|
|
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())
|
|
{
|
|
ts.tv_sec = minimum_wait_duration.total_seconds();
|
|
ts.tv_nsec = minimum_wait_duration.total_nanoseconds() % 1000000000;
|
|
}
|
|
else
|
|
{
|
|
ts.tv_sec = 0;
|
|
ts.tv_nsec = 0;
|
|
}
|
|
|
|
return &ts;
|
|
}
|
|
|
|
// 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 kqueue_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(
|
|
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.
|
|
asio::detail::mutex mutex_;
|
|
|
|
// The kqueue file descriptor.
|
|
int kqueue_fd_;
|
|
|
|
// Whether the kqueue wait call is currently in progress
|
|
bool wait_in_progress_;
|
|
|
|
// The interrupter is used to break a blocking kevent 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 except 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.
|
|
asio::detail::thread* thread_;
|
|
|
|
// Whether the service has been shut down.
|
|
bool shutdown_;
|
|
|
|
// Whether we need to call kqueue the next time the reactor is run.
|
|
bool need_kqueue_wait_;
|
|
};
|
|
|
|
} // namespace detail
|
|
} // namespace asio
|
|
|
|
#endif // defined(ASIO_HAS_KQUEUE)
|
|
|
|
#include "asio/detail/pop_options.hpp"
|
|
|
|
#endif // ASIO_DETAIL_KQUEUE_REACTOR_HPP
|