This project shows how std::call_once magically restrict only one thread to call a block of code and avoid wrong use of mutex to protect data, thus increase efficiency. - View it on GitHub