【C++】线程局部存储

KeyWords: TLS，thread_local， __thread，C11，C++11

引用阅读1-cywosp的【每天进步一点点——Linux中的线程局部存储（一）】

引用阅读2-努力努力再努力r的【__thread】

引用阅读3-whatday的【线程局部存储（TLS）】

引用阅读4-沪江塞外的【2020-09-17】

引用阅读5-cywosp的【每天进步一点点——Linux中的线程局部存储（二）】

线程局部存储实现

thread_local - cppreference.com

Storage class specifiers - cppreference.com

Storage class specifiers, except for thread_local, are not allowed on explicit 
specializations and explicit instantiations:
template <class T> struct S {
    thread_local static int tlm;
};
template <> thread_local int S<float>::tlm = 0; // "static" does not appear here


除了thread_local，存储类标识符不允许被用在显示特例化和隐式实例化中.

C语言的实现：

TARS中使用的开源代码查询：

D:\005-02-代码\016-TARS\TARS\TarsFramework\TarsBenchmark\src\comm\transport.cpp【搜索 static __thread char buf[MAX_SENDBUF_SIZE]; 】

C++的实现thread_local(多处用到)：

测试代码：

#include <iostream>
#include <string>
#include <thread>
#include <mutex>
 
thread_local unsigned int rage = 1; 
std::mutex cout_mutex;
 
void increase_rage(const std::string& thread_name)
{
    ++rage; // modifying outside a lock is okay; this is a thread-local variable
    std::lock_guard<std::mutex> lock(cout_mutex);
    std::cout << "Rage counter for " << thread_name << ": " << rage << '\n';
}
 
int main()
{
    std::thread a(increase_rage, "a"), b(increase_rage, "b"); 
    {
        std::lock_guard<std::mutex> lock(cout_mutex);
        std::cout << "Rage counter for main: " << rage << '\n';
    }
 
    a.join();
    b.join();
}

我的机器上的执行结果：

root@muten-virtual-machine:/home/muten/C++11# g++ thread_local.cpp -lpthread
root@muten-virtual-machine:/home/muten/C++11# ./a.out 
Rage counter for main: 1
Rage counter for b: 2
Rage counter for a: 2

官方提供的Possible output: 

Rage counter for a: 2
Rage counter for main: 1
Rage counter for b: 2

机器上实验的结果与官方提供的不考虑线程的输出顺序是一样的.