This is an interesting and deeply buried optimization for the GNU C++ STL implementation. I didn't expect anything like this.

Since atomics are really a hard topic, this article is welcome. It does a good job explaining what memory ordering does. It helps to debunk some common misconceptions.

Not a huge fan of the writing style. Still this gives a good idea of what you have to deal with when you're trying to build lock free data structures. Here it's illustrated with Rust, but it's not Rust specific.

Very interesting tools for testing and verifying concurrent code.

If you're still struggling with atomics and memory fences (and you should) this is a good piece to read. It really goes through it all.