Do you have any suggestions for someone that doesn't know much about the subject but would be interested in learning more about it?
Long-term I'd like to be able to dive into stuff like V8 and actually understand its internals.
Long-term I'd like to be able to dive into stuff like V8 and actually understand its internals.
https://news.ycombinator.com/item?id=11322060
https://news.ycombinator.com/item?id=2588696
http://lambda-the-ultimate.org/node/3851
The thing that makes luajit hard to understand is a combination of two things. First, the jit compiler uses advanced techniques that aren't easy to learn just looking at the code. You really need to know a lot of compiler theory to just get started. Additionally, mike pall wrote a large chunk of luajit in assembly language and hand-optimized lots of things, which gives the code that dense demoscene feeling that was mentioned before.
If you want to learn more about luajit one crucial thing would be learning about tracing jits for dynamic languages (luajit goes all in on tracing jit). The wikipedia article is pretty good. For papers make sure you read "Trace-based Just-in-Time Type Specialization for Dynamic Languages"
https://en.wikipedia.org/wiki/Tracing_just-in-time_compilati...
If you want to dive into a jit codebase one that might be interesting is the Higgs JIT. I haven't looked at the code itself very closely but I read Maxine's PhD thesis and her presentations on jit and I liked her approach of trying to make a jit that is as simple as possible. Most jits these days are massively complex beasts.
The use of assembly is actually very tame in luajit. I was actually referring more to things like the fact that it spits out native code by iterating backwards so it can do register allocation in reverse (simultaneously and with minimal memory footprint!). Because of a lot of neat details of how the SSA IR is laid out and stored and how tracing JITs work, this results in more optimal register allocation in a way that is actually very straightforward to implement.
It's very, very clever and overall very clean with minimal code, but understanding why things are done in a particular way requires a lot of experience. The demoscene is similar: very neat 'hacks' that combine low-level programming knowledge with very deep domain knowledge motivated by the need to write code that is both fast and small.