uefi-raw: Add binding for EFI_PCI_ROOT_BRIDGE_IO_PROTOCOL

[seijikun]

Checklist

• Sensible git history (for example, squash "typo" or "fix" commits). See the Rewriting History guide for help.
• Update the changelog (if necessary)

[seijikun]

No idea what the CI is complaining about, tbh.
Is it because the variants are called Uint8, Uint16, etc.?

I'm still quite unsure about how we should handle this: *mut Self here, because a normal read may (depending on the device) change the PCI device's internal state just as much as a write.
Should we just make both take a mutable pointer?

[nicholasbishop]

No idea what the CI is complaining about, tbh.

Currently just a couple minor clippy things, add must_use and const: https://github.com/rust-osdev/uefi-rs/actions/runs/14888659415/job/41814946880?pr=1658

Should we just make both take a mutable pointer?

Yep, when in doubt let's go with *mut in uefi-raw.

[seijikun]

Aight, when in doubt, *mut it out.
I made every method of which I think might change the device's state/memory in any way as *mut.

This is the error I don't really understand:

[phip1611]

The error reporting of the check-raw step is rather poor, I agree. You can run cargo xtask check-raw locally (sources are in <repo>/xtask/src) and debug into to look up what's wrong; feel free to add an improvement to the error reporting as well

[nicholasbishop]

Ah sorry I missed that error. It's because you're using a Rust enum, which is not quite compatible with a C enum (because it's undefined behavior to create a Rust enum from an invalid value).

This can be fixed by using the newtype_enum macro.

(I put up #1660 to improve this error message.)

[seijikun]

This can be fixed by using the newtype_enum macro.

Oof, that's embarrassing, the last merge request was only a few weeks ago and I've already forgotten about it.

I needed some iterations for the addressing stuff (PciIoAddress), but now I'm quite happy with it.

[nicholasbishop]

We did a release recently, so these changes should now move to a new ## Added section under [Unreleased].

[nicholasbishop]

The uefi-raw crate doesn't normally use generics. I think in the spec the address is treated as u64 in all cases, even though for for pci specifically the u64 is broken down into subfields, so let's do that and leave higher-level abstractions to the uefi crate.