Define more ioctl codes on riscv32gc-unknown-linux-gnu

[sunfishcode]

Description

Define ioctl codes including FICLONE and FS_IOC32_GETVERSION on riscv32gc-unknown-linux-gnu.

Sources

The values match the values in the linux-raw-sys crate, which is autogenerated from Linux sources:

https://github.com/sunfishcode/linux-raw-sys/blob/21a3b3b1fada619179810ca92404690d45890f29/src/riscv32/ioctl.rs#L248

Checklist

• Relevant tests in libc-test/semver have been updated
• No placeholder or unstable values like *LAST or *MAX are
  included (see #3131)
• Tested locally (cd libc-test && cargo test --target mytarget);
  especially relevant for platforms that may not be checked in CI

[rustbot]

r? @tgross35

rustbot has assigned @tgross35.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[kraj]

lgtm

[tgross35]

Any idea why these are gated in the first place? All of the constants are in https://github.com/torvalds/linux/blob/bec7dcbc242c6c087cede1a6fdfaeb5d6eaf25bf/include/uapi/linux/fs.h, I don't know why any platforms wouldn't have them unless they have no fs support.

In any case LGTM, blocked on CI issues.

[kraj]

Any idea why these are gated in the first place? All of the constants are in https://github.com/torvalds/linux/blob/bec7dcbc242c6c087cede1a6fdfaeb5d6eaf25bf/include/uapi/linux/fs.h, I don't know why any platforms wouldn't have them unless they have no fs support.

they all do exist but values are different sadly

In any case LGTM, blocked on CI issues.

[tgross35]

they all do exist but values are different sadly

Oh, from differences in sizes passed to _IOR and _IOW? We actually have these functions now, the cfg could probably just be deleted and replaced with calls

libc/src/unix/linux_like/linux/mod.rs

Lines 5991 to 6025 in 30f03b2

	/// Build an ioctl number, analogous to the C macro of the same name.
	const fn _IOC(dir: u32, ty: u32, nr: u32, size: usize) -> u32 {
	// FIXME(ctest) the `garando_syntax` crate (used by ctest in the CI test suite)
	// cannot currently parse these `debug_assert!`s
	//
	// debug_assert!(dir <= _IOC_DIRMASK);
	// debug_assert!(ty <= _IOC_TYPEMASK);
	// debug_assert!(nr <= _IOC_NRMASK);
	// debug_assert!(size <= (_IOC_SIZEMASK as usize));
	(dir << _IOC_DIRSHIFT)
	| (ty << _IOC_TYPESHIFT)
	| (nr << _IOC_NRSHIFT)
	| ((size as u32) << _IOC_SIZESHIFT)
	}
	/// Build an ioctl number for an argumentless ioctl.
	pub const fn _IO(ty: u32, nr: u32) -> u32 {
	_IOC(_IOC_NONE, ty, nr, 0)
	}
	/// Build an ioctl number for an read-only ioctl.
	pub const fn _IOR<T>(ty: u32, nr: u32) -> u32 {
	_IOC(_IOC_READ, ty, nr, size_of::<T>())
	}
	/// Build an ioctl number for an write-only ioctl.
	pub const fn _IOW<T>(ty: u32, nr: u32) -> u32 {
	_IOC(_IOC_WRITE, ty, nr, size_of::<T>())
	}
	/// Build an ioctl number for a read-write ioctl.
	pub const fn _IOWR<T>(ty: u32, nr: u32) -> u32 {
	_IOC(_IOC_READ | _IOC_WRITE, ty, nr, size_of::<T>())
	}

.

[tgross35]

Actually @sunfishcode would you mind just changing the relevant blocks to make use of those instead? I think we may as well use this as an excuse to get rid of some pre-const fn cruft.