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 …

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.

[image] vlaviron: 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 this thread was influenced by this paper or other public works on the subject, and if not, if knowing a…

[image] gasche: I wonder if your performance analysis could also suggest ways to speedup the C runtime a bit. (For example: maybe you could play with explicit tail-calls as a replacement for computed gotos in C, and see how that works. Or maybe looking at the parts of the runtime that are faster…

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

Learning

Topic		Replies	Views
OCaml Compiler Implemented in C++: It's Not Just Faster—It Also Increases Trust Ecosystem ai	11	1103	October 12, 2026
Advantages of OCaml over Rust Community	76	33563	September 4, 2019
Is there ever a time when it makes sense to choose Go over OCaml? Ecosystem	129	15136	February 9, 2023
On interoperability between Rust and OCaml Ecosystem	33	4392	August 21, 2020
Interesting OCaml Articles Community web , blog , video , podcast , social-media	107	20980	October 18, 2023

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

Related topics