The Rust community owes the OCaml community a debt of gratitude, the first Rust compiler was written in OCaml. In this post we rewrite the OCaml runtime in Rust. Consider it our down payment on repaying the debt. :saluting_face: No Marks Shinwell were harmed[1] More seriously The OCaml runtime is …

[image] gasche: Some small/local niceties, such as the ability to return several things from a function. (I wouldn’t mind proper algebraic datatypes either.) Obviously C is missing on the ADT front, but to me it is one of the languages which makes it easiest to return several things from a f…

It seems https://github.com/mbacarella/rustcaml leads to a 404?

You are thinking from an efficiency perspective, but the hassle of having to declare a struct is too much for me most of the time, so instead I use out-variables (which are also slightly annoying to use in C) with some decision fatigue on what to return. I would prefer to return a tuple or an anonym…

There’s always the option of declaring a struct inline: struct point { int x, y; } make_point (int x, int y) { return (struct point){ .x = x, .y =y }; }

[image] gasche: The availability of formal-methods tools for very advanced testing (say model-checking) or even formal verification of parts of the runtime. [image] gasche: We probably want a language that is portable, widely available. [image] gasche: We don’t have the maintena…

Like with OCaml the strict typing discipline of Ada can save one from lots of errors even without SPARK, and the compiler-writers of GNAT have been increasing including the ability to warn statically of more and more things via static analyses built in to the compiler. Ada does not have multiple re…

[image] gasche: we don’t have, for example, a concurrent hash table I’m not sure what you had in mind. I searched for concurrent hash tables for Rust, and I saw that there are two approaches, sharding and lock-free. I’m not aware of any lock-free hash table implementations for Ada. Ada has b…

[image] gasche: (For example: maybe you could play with explicit tail-calls as a replacement for computed gotos in C, and see how that works. Note that tail-calls are not a replacement for computed gotos. It is tall-calls + a very adhoc ABI. Indeed, if you are using the standard calling conv…

Coming back to the original subject, I recently stumbled (again) upon this paper: https://dl.acm.org/doi/10.1145/3737696 . It dates from November, and explores the challenges of using automated tools (including LLMs) to translate C code to Rust. I would be curious to know if the work presented in th…

[image] amaro: That’s very interesting! I’d like to try it, but the GitHub - mbacarella/rustcaml: The OCaml runtime ported from C to Rust · GitHub link is giving me a 404 error. Maybe the repository is currently private? Yes accidentally still had it private. Should have been fixed.

A line-by-line translation of the OCaml runtime from C to Rust

Learning

Topic		Replies	Views
Advantages of OCaml over Rust Community	76	33023	September 4, 2019
Interesting OCaml Articles Community web , blog , video , podcast , social-media	107	20622	October 18, 2023
Is there ever a time when it makes sense to choose Go over OCaml? Ecosystem	129	14448	February 9, 2023
OCaml compiler design and development Learning compiler	18	5119	May 25, 2020
My Thoughts on OCaml vs Haskell/Rust in 2023 Ecosystem blog	91	28436	September 13, 2023

A line-by-line translation of the OCaml runtime from C to Rust

Related topics