On 8/28/25 6:25 AM, Alexandre Courbot wrote:
> On Wed Aug 27, 2025 at 8:12 AM JST, John Hubbard wrote:
...
>> +macro_rules! define_all_pci_vendors {
>> + (
>> + $($variant:ident = $binding:expr,)+
>> + ) => {
>> +
>> + impl Vendor {
>
> Why the blank line here? (same for the `define_all_pci_classes` in the
> previous patch).
Removed in both places, thanks!
...
>> +impl fmt::Display for Vendor {
>> + #[inline]
>> + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
>> + write!(f, "0x{:04x}", self.0)
>> + }
>> +}
>
> Possibly an exercice for a future patch, but do we want to display the
> vendor name if it is defined, rather than its hex code (which is more
> the job of `Debug`)? We could leverage the macro above to do that. The
> same should be doable for the PCI classes.
>
> I suspect strings for all the names already exist on the C side, in
> which case we would want to reuse them instead of defining new ones.
A reasonable suspicion, but actually, they do not yet exist. So I think
we could just add them on the Rust side.
>
> Note that I don't think this needs to be done for this series - it's
> just a thought as I was looking at this `Display` implementation that
> looks more like a `Debug` one.
Good points. Yes, I can do a follow-up patch.
thanks,
--
John Hubbard
