Brendan Zabarauskas on Twitter: "@porglezomp @jonmsterling Yeah, I'm not an expert in this stuf, but it seems like it's the opposite case here? Rust has a chance to make this nicer for Rust users, where as people writing the runtime and C extensions for OCaml would have to go to the people designing the C memory model?" / Twitter

You’re unable to view this Tweet because this account owner limits who can view their Tweets. Learn more

cassie

@porglezomp

Mar 31

The part of the memory model being discussed here doesn’t really impact safe Rust, it’s changes that inherently only impact unsafe Rust. It affects safe Rust in mostly the same ways that C extensions could violate OCaml’s memory model no matter what bug fixes you make to it.

2

1

3

cassie

@porglezomp

Mar 31

The model that applies to references as opposed to raw pointers is already stricter than this experimental change to raw pointers

1

Replying to

and

Yeah, I'm not an expert in this stuf, but it seems like it's the opposite case here? Rust has a chance to make this nicer for Rust users, where as people writing the runtime and C extensions for OCaml would have to go to the people designing the C memory model?

8:52 PM · Mar 31, 2022Twitter Web App

Replying to

and

The Rust community is already much more keen to think about this stuff, and the desire to help people write trusted/unsafe code that does not violate the expectations of safe code encourages that, where as it's easier for C people to ignore seeing as they don't have that subset.

Conversation