Search⌘ K
Home
Courses
Modern C++ Concurrency: Get the most out of any machine

Multithreading: Shared Data

Explore best practices for managing shared data in multithreaded C++ programs. Learn techniques like minimizing locked sections, using scoped locks, and avoiding deadlocks to ensure thread safety and efficient synchronization.

We'll cover the following...

Data Sharing

With data sharing, the challenges in multithreading programming start.

Pass data per default by copy

C++
#include <iostream>
#include <thread>
#include <string>
int main(){
std::string s{"C++11"};
std::thread t1([s]{std::cout << s << std::endl; }); // do something with s
t1.join();
std::thread t2([&s]{ std::cout << s << std::endl; }); // do something with s
t2.join();
// do something with s
  
s.replace(s.begin(), s.end(), 'C', 'Z'); 
}

If you pass data such as the std::string s to a thread t1 by copy, the creator thread and the created thread t1 will use independent data; this is in contrast to the thread t2. It gets its std::string s by reference. This means you have to synchronize the access to s in the creator thread and the created thread t2 preventively. This is error-prone and expensive.

...