lauf VS boba

I have a VM lying around: https://github.com/foonathan/lauf

Yeah, that's what I'm doing in lauf. When returning, the stack segments stay allocated and can be reused when it reaches the end again.

I'm working on lauf, which is a low-level bytecode Interpreter: https://github.com/foonathan/lauf

I've been working on a generic bytecode interpreter the last couple of months. It reached a point where I needed to test it in a real compiler, so I've started writing a C interpreter.

I'm currently writing a C interpreter using my lauf bytecode interpreter. It's developed live on Youtube where I'm streaming every Tuesday and Thursday at 17:00 CEST/CET: https://www.youtube.com/c/foonathan

I've done a small implementation, used in type inference, in my language Boba. And you are correct, I used the linear equation solving method.

The vast majority of October's improvements on Boba were type system and runtime bug fixes. In particular, the effect handler/delimited continuation semantics were hopelessly busted beyond a few simple examples I'd fixated on.

September was another productive month for Boba, which is starting to get more 'quality of life' improvements rather than broad new features. That doesn't make the work less important: one of the bug fixes to the type inference engine last month caught a previously unseen bug in the core Boba libraries!

Also worth checking out is Adam Gundry's work on type inference for UoM types. Or, if you want an example implementation of the Abelian unification used in standard type inference extended with UoM types, you can reference my implementation, based on solving linear equations.

August was a surprisingly productive month for the Boba compiler. A few highlights:

The next large feature for Boba (a general-purpose concatenative language) is language integrated property tests.

That's a good question! I wrote up some of my thoughts on the benefits of Go as a backend, but there's also a historical component here. The first backend I was experimenting with was compile-to-C plus a C-based runtime. Go was closer to C than C# for what I needed at the time and I thought had a nicer concurrency story as a backend.