Ongoing Optimization

Now it’s time to put the theory into practice. The job is relatively easy. A small program should undergo ongoing optimization.

The program

// ongoingOptimization.cpp

#include <iostream>
#include <thread>

int x= 0;
int y= 0;

void writing(){
  x= 2000;
  y= 11;
}

void reading(){ 
  std::cout << "y: " << y << " ";
  std::cout << "x: " << x << std::endl;
}

int main(){
  std::thread thread1(writing);
  std::thread thread2(reading);
  thread1.join();
  thread2.join();
}

The program is quite simple. It consists of two threads thread1 and thread2.thread1 writes the values x and y. thread 2 reads the values x and y in the opposite direction. The execution of the program is not so thrilling either.

But you see, even this simple program has different results. So I finally came to my key idea.

In the process of the ongoing optimization of the program, I have two questions.

Is the program well-defined? Precisely, is there a data race?
Which values for x and y are possible?

The first question is often very challenging to answer. So I will reason in the first step about the program and verify in the second step my reasoning with CppMem. If I have answered the first question, the second answer can easily be derived. I will provide the possible values in a table.

Modernes C++ Mentoring

"Fundamentals for C++ Professionals" (open)

"Design Patterns and Architectural Patterns with C++" (open)

"C++20: Get the Details" (open)

"Concurrency with Modern C++" (open)

"Generic Programming (Templates) with C++": October 2024

"Embedded Programming with Modern C++": October 2024

"Clean Code: Best Practices for Modern C++": March 2025

Do you want to stay informed: Subscribe.

But still, one question isn’t answered yet. What do I mean by ongoing optimization? I mean, by ongoing optimization, that I try to improve the program by weakening the C++ memory model. So in my journey, I discuss the following stations.

Non-atomic variables
Locks
Atomics with sequential consistency
Atomics with acquire-release semantic
Atomics with relaxed semantic
Volatile variables

What’s next?

Of course, this was a very short post. But this post should provide you only the context of my ongoing optimization. In case you are bored, reason about the small program and answer the two question.

Is the program well-defined?
Which values for x and y are possible?

Still bored? Then reason about the weakening of the program’s memory model and the consequences. In the next post, I will analyze the unsynchronized access with the help of CppMem. Stay tuned.

Post Views: 3,551

Thanks a lot to my Patreon Supporters: Matt Braun, Roman Postanciuc, Tobias Zindl, G Prvulovic, Reinhold Dröge, Abernitzke, Frank Grimm, Sakib, Broeserl, António Pina, Sergey Agafyin, Андрей Бурмистров, Jake, GS, Lawton Shoemake, Jozo Leko, John Breland, Venkat Nandam, Jose Francisco, Douglas Tinkham, Kuchlong Kuchlong, Robert Blanch, Truels Wissneth, Mario Luoni, Friedrich Huber, lennonli, Pramod Tikare Muralidhara, Peter Ware, Daniel Hufschläger, Alessandro Pezzato, Bob Perry, Satish Vangipuram, Andi Ireland, Richard Ohnemus, Michael Dunsky, Leo Goodstadt, John Wiederhirn, Yacob Cohen-Arazi, Florian Tischler, Robin Furness, Michael Young, Holger Detering, Bernd Mühlhaus, Stephen Kelley, Kyle Dean, Tusar Palauri, Juan Dent, George Liao, Daniel Ceperley, Jon T Hess, Stephen Totten, Wolfgang Fütterer, Matthias Grün, Phillip Diekmann, Ben Atakora, Ann Shatoff, Rob North, Bhavith C Achar, Marco Parri Empoli, Philipp Lenk, Charles-Jianye Chen, Keith Jeffery, Matt Godbolt, and Honey Sukesan.

Thanks, in particular, to Jon Hess, Lakshman, Christian Wittenhorst, Sherhy Pyton, Dendi Suhubdy, Sudhakar Belagurusamy, Richard Sargeant, Rusty Fleming, John Nebel, Mipko, Alicja Kaminska, Slavko Radman, and David Poole.

My special thanks to Embarcadero
My special thanks to PVS-Studio
My special thanks to Tipi.build
My special thanks to Take Up Code
My special thanks to SHAVEDYAKS