Re: [RFC 1/2] rust: introduce abstractions for fwctl

[Danilo Krummrich]

Hi Zhi,

Additional to my other comments, some thoughts about naming.

On Thu Oct 30, 2025 at 5:03 PM CET, Zhi Wang wrote:
> +/// Trait implemented by each Rust driver that integrates with the fwctl subsystem.
> +///
> +/// Each implementation corresponds to a specific device type and provides
> +/// the vtable used by the core `fwctl` layer to manage per-FD user contexts
> +/// and handle RPC requests.
> +pub trait FwCtlOps: Sized {

Up to Jason, but I usually recommend to take the Rust module name into account,
i.e a user of the API can refer to this as fwctl::FwCtlOps.

Hence, I suggest to name this Ops or even Operations, such that users can refer
to this as fwctl::Ops or fwctl::Operations.

It would also be more consistent with existing code, e.g. pci::Device,
platform::Device, etc., but also the fwctl::Registration from this patch. :)

> +    /// Driver UCtx type.
> +    type UCtx;

I think we can affort to be a bit more verbose here, maybe just spell it out as
UserContext.

> +    /// fwctl device type, matching the C enum `fwctl_device_type`.
> +    const DEVICE_TYPE: u32;

I think it would make sense to have a new (Rust enum) type (fwctl::DeviceType)
for this, rather than using a raw u32, see [1] for reference.

[1] https://lore.kernel.org/all/20250828133323.53311-2-dakr@xxxxxxxxxx/

> +/// Represents a per-FD user context (`struct fwctl_uctx`).
> +///
> +/// Each driver embeds `struct fwctl_uctx` as the first field of its own
> +/// context type and uses this wrapper to access driver-specific data.
> +#[repr(C)]
> +#[pin_data]
> +pub struct FwCtlUCtx<T> {

What about UserContext<T>? A driver would refer to this as fwctl::UserContext.