Skip to content

Conversation

@afeistel
Copy link
Contributor

@afeistel afeistel commented Sep 21, 2025

abi_stable seems to have been abandoned

I don't yet know what the implications of this are in terms of how stabby lays stuff out in memory (especially enums) (nvm, checked, shouldn't affect since we're doing repr(C)), still need to at least run tests lol

(I had this change made something like 6 months ago but then forgot to push it anywhere, tried to fix abi_stable first instead)

had to rearrange code somewhat, unfortunately, since stabby wants stuff to be pub

@afeistel
Copy link
Contributor Author

afeistel commented Sep 21, 2025

stabby has its own Future stuff but it has a somewhat different approach (which, on >1.78 stable, allegedly has performance issues)

(and I kind of prefer the more direct control over the layout)

@oxalica
Copy link
Owner

oxalica commented Sep 30, 2025

abi_stable seems to have been abandoned

The crate does lack updates recently. Would you mind opening an issue upstream to ask about maintenance status, and probably report it to RUSTSEC?

I didn't know stabby before. I'll take a look when I got time.

Also that, I think we do not need to raise MSRV because of an default-disabled optional dependency.

@afeistel
Copy link
Contributor Author

Would you mind opening an issue upstream to ask about maintenance status

I'd already assume there would be no response, but I guess worth doing to explicitly ask

probably report it to RUSTSEC

will do after I go over my PR with fixes to it once more, since I think there are even some soundness-related:

  • #[repr(C, u8)] on enums. The way enums are currently handled seems to be potentially unsound.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants