Meeting 2025-09-30

[adamgreig]

This is the agenda for the next meeting. Please add anything you'd like to discuss below!

• Meetings: Tuesday 8pm Europe/Berlin time
• Join the Matrix chat
• IRC/Matrix logs

[0xllx0]

To help find a good path forward, I would like to discuss / receive feedback on: rust-embedded/heapless#583

Specifically, is an alternative MpMcQueue implementation warranted? If so, which alternative should be pursued?

A number of suggested potential implementations: rust-embedded/heapless#583 (comment)

[jamesmunns]

Not a Libs team member, but my $0.02: If there are better algorithms, I'd highly recommend going ahead and making a separate crate for it, and start implementing there. Once there is a good API, it will be much easier to discuss replacing the current algo, than comitting to changing the existing one without a concrete impl/API to discuss.

[datdenkikniet]

A separate crate would also potentially lend itself to tests written with loom. Especially for things like this, where exhaustively simulating/emulating contention is really the only way to ensure that the implementation is "truly correct".

We've had great success with it in rtic-sync, where it really clearly demonstrated that it is hard to correctly reason about how atomics & critical sections interact with eachother (because new edge-cases kept popping up, until there were no more gaps).

[jamesmunns]

Also clarifying, forking heapless and building a new algo there is totally a good choice too, even if you don't want to publish/maintain another crate.

I was mostly suggesting that it's better to build THEN discuss a concrete proposal, than having a lot of discussion before writing the code.

[adamgreig]

Topics from last week #871:

• Discussion on moving some crates to a new t-libs or t-docs subteam
• Discussion on linker scripts

Other topics:

• cortex-m-pac cortex-m-types crate: standard traits and result handling for Cortex-M targets cortex-m#560
• cargo-binutils release

[sirhcel]

@jannic and i updated the dependencies of sysfs-gpio and the CLI looks at least as good as before at initial testing for rust-embedded/gpio-utils#32.

[datdenkikniet]

A small but (to me & others) useful PR regarding broken submodules: rust-embedded/svd#294

[adamgreig]

thanks for the reminder!