• c++11之100行实现简单线程池


    代码从github上拷的,写了一些理解,如有错误请指正

    Threadpool.h

      1 #ifndef THREAD_POOL_H
      2 #define THREAD_POOL_H
      3 
      4 #include <vector>
      5 #include <queue>
      6 #include <memory>
      7 #include <thread>
      8 #include <mutex>
      9 #include <condition_variable>
     10 #include <future>
     11 #include <functional>
     12 #include <stdexcept>
     13 
     14 class ThreadPool {
     15 public:
     16     ThreadPool(size_t);
     17     template<class F, class... Args>
     18     auto enqueue(F&& f, Args&&... args) 
     19         -> std::future<typename std::result_of<F(Args...)>::type>;
     20     ~ThreadPool();
     21 private:
     22     // need to keep track of threads so we can join them
     23     std::vector< std::thread > workers;
     24     // the task queue
     25     std::queue< std::function<void()> > tasks;
     26     
     27     // synchronization
     28     std::mutex queue_mutex;
     29     std::condition_variable condition;
     30     bool stop;
     31 };
     32  
     33 // the constructor just launches some amount of workers
     34 inline ThreadPool::ThreadPool(size_t threads)
     35     : stop(false)
     36 {
     37     for(size_t i = 0;i<threads;++i)
     38         workers.emplace_back(                //这里一下启动threads个,即使lambda阻塞(在已启动子线程内的阻塞),主线程还是会循环
     39             [this]
     40             {
     41                 for(;;)
     42                 {
     43                     std::function<void()> task;
     44 
     45                     {
     46                         std::unique_lock<std::mutex> lock(this->queue_mutex);
     47                         
     48                         //当前线程被阻塞, 直到condition.notify_one()调用,如果lambda返回false,wait会解锁互斥元lock并置阻塞或等待状态,如果条件满足互斥元仍被锁定
     49                         //而这里锁用的是std::unique_lock而不是std::lock_guard,是因为std::lock_guard不能在wait等待中解锁,并在之后重新锁定
     50                         //如果互斥元在线程休眠期间始终被锁定,enqueue就无法锁定互斥元往下执行,则造成死锁
     51                         this->condition.wait(lock,                                
     52                             [this]{ return this->stop || !this->tasks.empty(); });    
     53                         
     54                         if(this->stop && this->tasks.empty())
     55                             return;
     56                         task = std::move(this->tasks.front());
     57                         this->tasks.pop();
     58                     }
     59 
     60                     task();
     61                     
     62                 }
     63             }
     64         );
     65 }
     66 
     67 // add new work item to the pool            //函数后跟throw()代表不抛出任何异常,跟thorw(...)代表可以抛出任何异常
     68 template<class F, class... Args>            //... Args这个代表不限类型,不限数量
     69 auto ThreadPool::enqueue(F&& f, Args&&... args)                    //&&代表右值引用(可以用常量做参数)
     70     -> std::future<typename std::result_of<F(Args...)>::type>            //<F(Args...)>代表这个是个返回类型为F,参数不确定多的函数;放入类型(如int)会报错
     71 {
     72     using return_type = std::result_of<F(Args...)>::type;                //std::result_of::type 获得函数返回类型,直接用decltype会报错
     73 
     74     auto task = std::make_shared< std::packaged_task<return_type()> >(
     75             std::bind(std::forward<F>(f), std::forward<Args>(args)...)
     76         );
     77         
     78     std::future<return_type> res = task->get_future();
     79     {                                                                //这加一个作用域的作用是出了这个作用域就解锁
     80         std::unique_lock<std::mutex> lock(queue_mutex);
     81 
     82         // don't allow enqueueing after stopping the pool
     83         if(stop)
     84             throw std::runtime_error("enqueue on stopped ThreadPool");
     85 
     86         tasks.emplace([task](){ (*task)(); });
     87     }
     88     condition.notify_one();        ////选择一个wait状态的线程进行唤醒,并使他获得对象上的锁来完成任务(即其他线程无法访问对象)
     89     return res;
     90 }
     91 
     92 // the destructor joins all threads
     93 inline ThreadPool::~ThreadPool()
     94 {
     95     {
     96         std::unique_lock<std::mutex> lock(queue_mutex);
     97         stop = true;
     98     }
     99     condition.notify_all();                        //通知所有wait状态的线程竞争对象的控制权,唤醒所有线程执行
    100     for(std::thread &worker: workers)
    101         worker.join();
    102 }
    103 
    104 #endif

    运行代码

     1 #include "stdafx.h"
     2 #include <iostream>
     3 #include <vector>
     4 #include <chrono>
     5 
     6 #include "ThreadPool.h"
     7 
     8 int main()
     9 {
    10     
    11     ThreadPool pool(4);
    12     std::vector< std::future<int> > results;
    13 
    14     for(int i = 0; i < 8; ++i) {
    15         results.emplace_back(
    16             pool.enqueue([i] {
    17                 std::cout << "hello " << i << std::endl;
    18                 std::this_thread::sleep_for(std::chrono::seconds(1));
    19                 std::cout << "world " << i << std::endl;
    20                 return i*i;
    21             })
    22         );
    23     }
    24 
    25     for(auto && result: results)
    26         std::cout << result.get() << ' ';
    27     std::cout << std::endl;
    28     system("pause");
    29     return 0;
    30 }
  • 相关阅读:
    混淆
    【虚拟DOM】√
    fn
    notebook
    sourceMappingURL
    十进制转换为十六进制
    关于this和super的区别
    类中访问级别
    Java中字符串解析
    IWorkspace操作
  • 原文地址:https://www.cnblogs.com/wangshaowei/p/8884290.html
Copyright © 2020-2023  润新知