liugan528的博客

Updating…… 1.每个线程都必须具有一个初始函数(initial function)。例如：#include <iostream>#include &amp

// condition_variable::notify_all#include <iostream> // std::cout#include <thread> // std::thread#include <mutex> // std::mutex, std::unique_lock#include <condition_variable> // std::condition_

std::lock()、std::try_lock()、std::call_once()的用法

1. 像std::lock_guard那样使用接受一个互斥体引用的构造函数，这个构造函数尝试获得互斥体的锁，并且阻塞直到获得锁。#include <iostream> // std::cout#include <thread> // std::thread#include <mutex> // std::mutex, std::lock_guard, std::adopt_lockstd::mutex mtx;

// lock_guard example#include <iostream> // std::cout#include <thread> // std::thread#include <mutex> // std::mutex, std::lock_guardstd::mutex mtx;void print_thread_id(int idx) { std::lock_guard<std::m

标准库提供了一个std::timed_mutex类和一个std::recursive_timed_mutex类，这两个类都是支持普通lock()、try_lock()和unlock()方法的定时互斥体类，此外还支持以下两个方法：try_lock_for调用线程将尝试在一个给定的相对时间内试图获得这个锁。如果不能在给定超时之后获得这个锁，这个调用失败并返回false。如果在超时时间之内获得了这个锁，这个调用成功并返回true。#include <iostream>#include <

std::recursive_mutex的行为几乎和std::mutex一致，区别在于已经获得了一个递归互斥体所有权的线程允许在同一个互斥体上再次调用loci()和try_lock()。调用线程调用unlock()方法的次数应该等于获得这个递归互斥体锁的次数。#include <iostream>#include <string>#include <mutex>#include <thread>std::recursive_mutex mu; /

#include <iostream>#include <thread>#include <vector>void func() { thread_local int a = 0; int b = 0; a++; b++; std::cout << std::this_thread::get_id() << " a = " << a << " b = " << ..

通过函数指针创建线程// 通过函数指针创建线程#include <iostream>#include <thread>void counter(int id, int num_iterations){ for (int i = 0; i < num_iterations; ++i) { std::cout << "counter " << id << " has value ";

#include <iostream>#include <atomic>#include <thread>#include <vector>void promote(std::atomic_int &a, int &b){ for (int i = 0; i < 1000000; ++i) { ++a; ++b; }}int main(){ std::ve

如果实际线程数大于硬件支持的线程数，则出现超额订阅oversubscription，上下文切换将意味着更多的线程会降低性能。#include <thread>#include <functional>#include <numeric>#include <algorithm> // std::for_each#include <vector>#include <iostream>template<typena

生成一批线程并等待它们完成#include <thread>#include <functional>#include <algorithm>#include <vector>#include <iostream>void Print(int i) { std::cout << i << std::endl;}int main(){ std::vector<std::thread&

std::thread和std::unique_ptr，std::fstream、std::ifstream、std::ofstream等流一样只可转移move所有权，不可复制copy。#include <thread>#include <utility> // std::move#include <iostream>void Foo() { std::cout << "foo" << std::endl;}int mai

任何callable target – functions, lambda expressions, bind expressions, or other function objects, as well as pointers to member functions and pointers to data members都可以作为线程的入口函数。#include <thread>#include <iostream>#include <functional>

创建线程时函数参数默认直接拷贝？不一定按照参考文献[1]的说法，传递参数给线程函数的时候参数会以默认的方式被复制到内部存储空间。实际上，在Microsoft Visual Studio Professional 2017实验环境下其表现并非如此，如下程序是可以正常运行的。也许参考文献[1]采用的是旧标准，而最新的编译器已经更“智能”了。#include <string>#include <thread>#include <iostream>void foo(c

#include <thread>#include <iostream>class Stomach{public: void Dislike() { std::cout << "I can't stomach his speeches." << std::endl; }};class ThreadGuard{public: ThreadGuard(std::thread &t) : t_(t)