1. 线程的移动语义实现
基于前面几章介绍的移动语义,我们用它来实现线程。#include <iostream> #include <thread> #include <vector> #include <algorithm> #include <cassert> int main() { std::vector<std::thread> workers; for (int i = 0; i < 5; i++) { auto t = std::thread([i]() { std::cout << "thread function: " << i << "\n"; }); workers.push_back(std::move(t)); } std::cout << "main thread\n"; std::for_each(workers.begin(), workers.end(), [](std::thread &t) { assert(t.joinable()); t.join(); }); return 0; }运行结果为:
thread function: thread function: thread function: 01
thread function: 3
2
main thread
thread function: 4
接下来对代码进行一些改变。将线程函数task()独立出来。
#include <iostream> #include <thread> #include <vector> #include <algorithm> #include <cassert> void task(int i) { std::cout<<"worker : "<<i<<std::endl; } int main() { std::vector<std::thread> workers; for (int i = 0; i < 5; i++) { auto t = std::thread(&task, i); workers.push_back(std::move(t)); } std::cout << "main thread" << std::endl; std::for_each(workers.begin(), workers.end(), [](std::thread &t) { assert(t.joinable()); t.join(); }); return 0; }运行结果为:
worker : worker : 02
worker : 3
worker : 1
main thread
worker : 4
2. 通过引用传递线程参数
下面将线程函数的参数改为引用void task(int &i) { std::cout << "worker : " << i << "\n"; }与此同时,线程的构造函数同样需要相应的修改
auto t = std::thread(&task, std::ref(i));
但是,上面程序看似可以工作,不过不是一个好的设计,因为多个线程在同时使用一个指向相同对象的引用。