ASoC: Exploring Rust support for codec drivers

ASoC: Exploring Rust support for codec drivers

[Jihed Chaibi]

Hi Mark, Liam,

I've been considering working on Rust support for ASoC, starting with
a small codec driver as a proof of concept.

I've previously contributed a few patches to ASoC and have been
interested in exploring Rust-for-Linux in this area. I raised the
subject with the Rust maintainers earlier this year, and they
recommended discussing it with the subsystem maintainers before
getting too far into the implementation.

I haven't found any ongoing work on Rust abstractions for ASoC
component or DAI drivers, but I may have missed something.

My initial idea would be to investigate the minimal abstractions
needed for a Rust codec driver to integrate with the existing ASoC
core, starting with component/DAI registration and callbacks, and
expanding only where needed.

Would you be open to this kind of work in ASoC? And is there a
particular approach or existing effort you'd recommend looking at
before starting?

I'd appreciate any thoughts.

Thanks,
Jihed

Re: ASoC: Exploring Rust support for codec drivers

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 915 bytes --]

On Wed, Oct 21, 2026 at 11:31:48AM +0200, Jihed Chaibi wrote:

> I haven't found any ongoing work on Rust abstractions for ASoC
> component or DAI drivers, but I may have missed something.

> My initial idea would be to investigate the minimal abstractions
> needed for a Rust codec driver to integrate with the existing ASoC
> core, starting with component/DAI registration and callbacks, and
> expanding only where needed.

> Would you be open to this kind of work in ASoC? And is there a
> particular approach or existing effort you'd recommend looking at
> before starting?

I'm not aware of anyone else working in this area, I'm certainly open to
it.  The main issues I see are that most meaningful CODEC drivers use
lower level abstractions like regmap which don't yet have rust support,
and the ongoing refactorings that Morimoto-san is doing.  I think the
CODEC drivers should be fairly stable now, though.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]