eventfd实现线程事件通知机制

 通过eventfd实现的事件通知机制

---

eventfd的使用

eventfd系统函数

eventfd  - 事件通知文件描述符

#include <sys/eventfd.h>
int eventfd（unsigned int initval ，int flags ）;

创建一个能被用户应用程序用于时间等待唤醒机制的eventfd对象.
initval :
eventfd（）创建一个可用作事件的“eventfd对象”用户空间应用程序和内核等待/通知机制通知用户空间应用程序的事件。该对象包含一个由内核维护的无符号64位整型（uint64_t）计数器。此计数器的初始值通过initval指定。一般设0.

flags ：
以下标志中按位OR运算以更改eventfd()的行为，(文件中常用的这两个flags肯定都懂意思吧，就不翻译了,第三个信号量的不管它.)：

   EFD_CLOEXEC (since Linux 2.6.27)
          Set the close-on-exec (FD_CLOEXEC) flag on the new file
          descriptor.  See the description of the O_CLOEXEC flag in
          open(2) for reasons why this may be useful.

   EFD_NONBLOCK (since Linux 2.6.27)
          Set the O_NONBLOCK file status flag on the new open file
          description.  Using this flag saves extra calls to fcntl(2) to
          achieve the same result.

   EFD_SEMAPHORE (since Linux 2.6.30)
          Provide semaphore-like semantics for reads from the new file
          descriptor.  See below.

read（2）

成功读取返回一个8byte的整数。read（2）如果提供的缓冲区的大小小于8个字节返回错误EINVAL

write (2）

将缓冲区写入的8字节整形值加到内核计数器上。可以写入的最大值
是计数器中是最大的无符号64位值减1（即0xfffffffffffffffe）。

返回值：

On success, eventfd() returns a new eventfd file descriptor. On error, -1 is returned and errno is set to indicate the error.

使用示例

#include <iostream>
#include <assert.h>
#include <poll.h>
#include <signal.h>
#include <sys/eventfd.h>
#include <unistd.h>
#include <string.h>
#include <thread>

static int s_efd = 0;

int createEventfd()
{
  int evtfd = eventfd(0, EFD_NONBLOCK | EFD_CLOEXEC);

  std::cout << "createEventfd() fd : " << evtfd << std::endl;

  if (evtfd < 0)
  {
    std::cout << "Failed in eventfd\n";
    abort();
  }

  return evtfd;
}

void testThread()
{
  int timeout = 0;
  while(timeout < 3) {
    sleep(1);
    timeout++;
  }

  uint64_t one = 1;
  ssize_t n = write(s_efd, &one, sizeof one);
  if(n != sizeof one)
  {
    std::cout << " writes " << n << " bytes instead of 8\n";
  }
}

int main()
{
  s_efd = createEventfd();

  fd_set rdset;
  FD_ZERO(&rdset);
  FD_SET(s_efd, &rdset);

  struct timeval timeout;
  timeout.tv_sec = 1;
  timeout.tv_usec = 0;

  std::thread t(testThread);

  while(1)
  {
    if(select(s_efd + 1, &rdset, NULL, NULL, &timeout) == 0)
    {
      std::cout << "timeout\n";
      timeout.tv_sec = 1;
      timeout.tv_usec = 0;
      FD_SET(s_efd, &rdset);
        continue;
    }

    uint64_t one = 0;

    ssize_t n = read(s_efd, &one, sizeof one);
    if(n != sizeof one)
    {
      std::cout << " read " << n << " bytes instead of 8\n";
    }

    std::cout << " wakeup ！\n";

    break;
  }

  t.join();
  close(s_efd);

  return 0;
}

./test.out
createEventfd() fd : 3
timeout
timeout
timeout
 wakeup ！

eventfd 单纯的使用文件描述符实现的线程间的通知机制，可以很好的融入select、poll、epoll的I/O复用机制中.

EventLoop对eventfd的封装

所增加的接口及成员:

    typedef std::function<void()> Functor;
    void runInLoop(const Functor& cb);
    void wakeup(); //是写m_wakeupFd 通知poll 处理读事件.
    void queueInLoop(const Functor& cb);
private:
    //used to waked up
    void handleRead();
    void doPendingFunctors();
    
    int m_wakeupFd;
    std::unique_ptr<Channel> p_wakeupChannel;
    mutable MutexLock m_mutex;
    bool m_callingPendingFunctors; /* atomic */
    std::vector<Functor> m_pendingFunctors; // @GuardedBy mutex_

---

工作时序

(runInLoop() -> quueInLoop())/queueInLoop() -> wakeup() -> poll() -> handleRead() -> doPendingFunctors()

runInLoop()

如果用户在当前IO线程调用这个函数， 回调会同步进行； 如果用户在其他线程调用runInLoop()，cb会被加入队列， IO线程会被唤醒来调用这个Functor.

void EventLoop::runInLoop(const Functor&  cb)
{
  if(isInloopThread())
    cb();
  else
    queueInLoop(cb);
}

---

queueInLoop()

会将回调添加到容器，同时通过wakeup()唤醒poll()调用容器内的回调.

void EventLoop::queueInLoop(const Functor& cb)
{
  LOG_TRACE << "EventLoop::queueInLoop()";
  {
    MutexLockGuard lock(m_mutex);
    m_pendingFunctors.push_back(std::move(cb));
  }

  if(!isInloopThread())
  {
    wakeup();
  }
}

内部实现，

wakeup()

写已注册到poll的eventfd 通知poll 处理读事件.

//  m_wakeupFd(createEventfd()),
//  p_wakeupChannel(new Channel(this, m_wakeupFd)),
void EventLoop::wakeup()
{
  uint64_t one = 1;
  ssize_t n = sockets::write(m_wakeupFd, &one, sizeof one);
  if(n != sizeof one)
  {
    LOG_ERROR << "EventLoop::wakeup() writes " << n << " bytes instead of 8";
  }
}

handleRead()

poll回调读事件,处理eventfd.

void EventLoop::handleRead() //handle wakeup Fd
{
  LOG_TRACE << "EventLoop::handleRead() handle wakeup Fd";
  uint64_t one = 1;
  ssize_t n = sockets::read(m_wakeupFd, &one, sizeof one);
  if(n != sizeof one)
  {
    LOG_ERROR << "EventLoop::handleRead() reads " << n << "bytes instead of 8";
  }
  doPendingFunctors();
}

doPendingFunctors()

处理挂起的事件.

void EventLoop::doPendingFunctors()
{
  LOG_TRACE << "EventLoop::doPendingFunctors()";
  std::vector<Functor> functors;
  m_callingPendingFunctors = true;

  {
    MutexLockGuard lock(m_mutex);
    functors.swap(m_pendingFunctors);
  }

  for(size_t i = 0; i < functors.size(); ++i)
  {
    functors[i]();
  }

  m_callingPendingFunctors = false;

}