Re: [PATCH v6 09/16] rust: pci: add basic PCI device / driver abstractions

[Danilo Krummrich <dakr@kernel.org>]

On Sun, Dec 15, 2024 at 01:23:22PM +0100, Greg KH wrote:
> On Thu, Dec 12, 2024 at 05:33:40PM +0100, Danilo Krummrich wrote:
> > +impl DeviceId {
> > +    const PCI_ANY_ID: u32 = !0;
> > +
> > +    /// PCI_DEVICE macro.
> > +    pub const fn new(vendor: u32, device: u32) -> Self {
> > +        Self(bindings::pci_device_id {
> > +            vendor,
> > +            device,
> > +            subvendor: DeviceId::PCI_ANY_ID,
> > +            subdevice: DeviceId::PCI_ANY_ID,
> > +            class: 0,
> > +            class_mask: 0,
> > +            driver_data: 0,
> > +            override_only: 0,
> > +        })
> > +    }
> > +
> > +    /// PCI_DEVICE_CLASS macro.
> > +    pub const fn with_class(class: u32, class_mask: u32) -> Self {
> 
> I know naming is hard, and I'm not going to object to this at all, but
> using "new()" and "with_class()" feels a bit odd and mis-matched.  How
> about spelling it out, pci_device(), and pci_device_class()?

This is likely being call with module prefix, i.e. `pci::DeviceId::new`, so I'd
rather not encode "pci" again.

Maybe `pci::DeviceId::from_id` and `pci::DeviceId::from_class`?

> 
> Anyway, not a bit deal at all, let's see how this plays out with real
> drivers and we can always change it later.
> 
> > +// Allow drivers R/O access to the fields of `pci_device_id`; should we prefer accessor functions
> > +// to void exposing C structure fields?
> 
> Minor nit, do you mean "to avoid exposing..."?

Yes, but I don't think we need this comment any longer, now that we do not pass
this to probe() any longer. I'll remove the `Deref` impl.

> 
> Other than these, this looks good!  If I can get an ack from the PCI
> maintainer, I'll be glad to queue these up in my driver core tree...

Sounds good!

> 
> thanks,
> 
> greg k-h