---
date: 2025-04-29
url: https://hackmd.io/Ck3h9PJRQX-NSM1sEdb-AA
---

# Libs-API Meeting 2025-04-29

###### tags: `Libs Meetings` `Minutes`

**Meeting Link**: https://meet.jit.si/rust-libs-meeting-crxoz2at8hiccp7b3ixf89qgxfymlbwr
**Attendees**: Amanieu, Mara, David, Josh, Chris Denton, TC

## Agenda

- Triage
- Anything else?

## Triage

### FCPs

10 rust-lang/rust T-libs-api FCPs

- merge rust.tf/80437 *Tracking Issue for \`box\_into\_inner\`* - (1 checkboxes left)
- merge rust.tf/138498 *Implement Deref\<Target=ByteStr\> for CStr* - (3 checkboxes left)
- merge rust.tf/106418 *Implement \`PartialOrd\` and \`Ord\` for \`Discriminant\`* - (2 checkboxes left)
- merge rust.tf/130823 *Tracking Issue for \`non\_null\_from\_ref\`* - (3 checkboxes left)
- merge rust.tf/122661 *Change the desugaring of \`assert!\` for better error output* - (3 checkboxes left)
- merge rust.tf/137992 *Stabilise \`os\_string\_pathbuf\_leak\`* - (4 checkboxes left)
- merge rust.tf/115585 *Tracking issue for \`cfg\_match\`* - (2 checkboxes left)
- merge rust.tf/138016 *Added \`Clone\` implementation for \`ChunkBy\`* - (3 checkboxes left)
- merge rust.tf/111137 *Tracking Issue for AVX512 intrinsics* - (3 checkboxes left)
- merge rust.tf/139087 *Fallback \`{float}\` to \`f32\` when \`f32: From\<{float}\>\` and add \`impl From\<f16\> for f32\`* - (10 checkboxes left)

[BurntSushi (8)](https://rfcbot.rs/fcp/BurntSushi), [m-ou-se (6)](https://rfcbot.rs/fcp/m-ou-se), [joshtriplett (5)](https://rfcbot.rs/fcp/joshtriplett), [dtolnay (2)](https://rfcbot.rs/fcp/dtolnay), [Amanieu (6)](https://rfcbot.rs/fcp/Amanieu)

### (nominated) rust.tf/129121 *Stabilize \`tcp\_quickack\`*

FCP proposed.

David: deferaccept? unrelated? rust.tf/119639

JoshT: unrelated option, but we might as well stabilize that one too

Amanieu: I have a concern about this. The argument type for seconds is u32.

The 8472: These kind of apis tend to get a nanosecond version later. We should take a Duration

JoshT: How do we round to whole seconds? Truncating is fine.

TC: Or panic when it's not a whole number of seconds.

Amanieu: Bad for portability.

JoshT: not a question of portability, but i agree we shouldn't do it.

Options:

- `set_deferaccept(u32)` (seconds)
- `set_deferaccept(Duration)` (round down)
- `set_deferaccept(Duration)` (panic on not exact seconds)
- `set_deferaccept(Duration)` (io::ErrorKind::InvalidInput)
- `set_deferaccept_seconds(u32)`

Amanieu: let's have this discussion on the tracking issue.

JoshT to post a comment there.

### (nominated) rust.tf/129334 *Implement (part of) ACP 429: add \`DerefMut\` to \`Lazy\[Cell/Lock\]\`*

Amanieu: Seems fine

Mara: it also adds force_mut to both types. Seems fine.

### (nominated) rust.tf/130703 *Tracking Issue for secure random data generation in \`std\`*

Amanieu: Perhaps we should discuss this at the all-hands.

### (nominated) rust.tf/138023 *Add \`std::io::Seek\` instance for \`std::io::Take\`*

JoshT: Looks fine. Will propose FCP.

### (nominated) rust.tf/138340 *Remove some unsized tuple impls now that we don't support unsizing tuples anymore*

### (nominated) rust.tf/138879 *Ensure non\-empty buffers for large vectored I/O*

JoshT: Slices off the empty buffers from the start, to avoid Ok(0).

JoshT: I don't think we want to remove all empty buffers. But if we're already checking the limit anyway, it doesn't matter if we do this.

Amanieu: In nearly all cases, the first buffer is going to be nonempty, so this is going to be a O(1) check basically.

The 8472: Should we steer people towards the advance method?

Amanieu: We already do

Josht: I think this is ready for FCP.

Amanieu: Ok(0) meaning eof is guaranteed on read() but not the vectorized variants.

JoshT: It is the expectation though.

Amanieu: It does say "like read()".

### (nominated) rust.tf/139901 *Change proc\_macro::Span::byte\_range \-\> byte\_offset.*

Discussed last time.

### (nominated) rust.tf/139903 *Naming of proc\_macro::Span::local\_file()*

Discussed last time.

### (nominated) rust.tf/139916 *make std::intrinsic functions actually be intrinsics*

Separate crater run?

Seems unlikely this breaks anything. Let's rely on the beta crate run.

### (nominated) rust.tf/139994 *\`impl Display for CStr{,ing}\`*

Rejected in ACP. Amanieu to reply.

### (nominated) rust.tf/140151 *remove intrinsics::drop\_in\_place*

Accidentally stable. Crater detected no regressions.

FCP

### (nominated) rust.tf/140250 *u16 format precision change, breach of the Rust stability guarantee*

JoshT: Agree with Mara's analysis

### (waiting on team) rust.tf/136687 *Improve the documentation of \`Display\` and \`FromStr\`, and their interactions*

### (new change proposal) rust.tf/libs577 *Option::try\_get\_or\_insert\_with*

### (new change proposal) rust.tf/libs576 *Add \`Chars::single(&self) \-\> Option\<char\>\` for getting exactly one character*

### (new change proposal) rust.tf/libs575 *ACP: \`Option::update\_or\_default\`*

### (new change proposal) rust.tf/libs574 *ACP: str::advance\_char*

### (new change proposal) rust.tf/libs573 *ACP: \[perf\] \`proc\-macro\` \`Ident\` should have method \`as\_str\`*

### (new change proposal) rust.tf/libs572 *Add \`Cell::get\` for \`Rc\`\-like shallow\-copying types*

### (new change proposal) rust.tf/libs570 *actually checked bitshifts*

Right shifting a 1-bit out is overflow?

Amanieu: yes?

JoshT: No?

TC: underflow?

Amanieu: checked_lossless_shift_left?

Amanieu: deprecate all the overflowing shift methods. bad api design.

No consensus to add these methods.

---

TC: Description of what LLVM has exactly:

https://github.com/rust-lang/libs-team/issues/570#issuecomment-2839591488

Amanieu: Some consideration for deprecating some methods, such as `overflowing_shl`.

Current shift methods:
- shl (Shl trait)
- checked_shl
- overflowing_shl
- strict_shl (unstable)
- unbounded_shl (stabilized in beta)
- unchecked_shl
- wrapping_shl

Proposed new methods:
- checked_exact_shl -> Option
- unchecked_exact_shl -> T

Preconditions:
- shift in range 0..BITS
- value preserved when shifted back

TC: Alternate names?:

- exact_shl -> Option
- exact_shl_unchecked -> T, unsafe

The 8472: We should update this for `exact_div` then also.

Amanieu: I'm happy with that.

exact_div tracking issue: https://github.com/rust-lang/rust/issues/139911

### (new change proposal) rust.tf/libs569 *ACP: abstract \`BufReader::peek\`*

The 8472 to comment on BufReader::peek.

Amanieu to accept with some renaming and API changes.

### (new change proposal) rust.tf/libs568 *Implement \`Read\` and \`Write\` traits for \`BorrowedFd\` and \`OwnedFd\`*

### (new change proposal) rust.tf/libs562 *Add ability to customize STARTUPINFO structure in Command API*

### (stalled change proposal) rust.tf/libs210 *Add associated consts to f32, f64 for mathematical constants*

TC: This is more of a lang thing, but in the context of:

https://github.com/rust-lang/rust/issues/134691

We could imagine wanting:

trait Tr {
    const C: u8;
}

use Tr::C;

// And also:

struct S;

impl Tr for S {
    const C: u8 = 0;
}

use <S as Tr>::C;

// And also:

impl S {
    const C: u8 = 0;
}

use S::C;

### (stalled change proposal) rust.tf/libs366 *Implementing UnixSocketExt traits for solaris/illumos ; starting with the unix\_socket\_exclbind feature*

### (stalled change proposal) rust.tf/libs194 *Fix:Introduce an interface to expose the current \`Command\` captured env var logic*

### (stalled change proposal) rust.tf/libs111 *Restructure ptr\_metadata to minimal support*

### (stalled change proposal) rust.tf/libs296 *ACP: Designing an alternative \`FromStr\`*

### (stalled change proposal) rust.tf/libs383 *Add \`FRAC\_1\_SQRT\_2PI\` constant*

Re-pinged Mara.

### (stalled change proposal) rust.tf/libs335 *ACP: \`PathBuf::has\_dir\_suffix\`*

### (stalled change proposal) rust.tf/libs331 *Support permanently skipping tests on a specific system*

### (stalled change proposal) rust.tf/libs207 *\`parse\_line\` method for \`Stdin\`*

Josh: There's a new RFC for something like this:

https://github.com/rust-lang/rfcs/pull/3799

### (stalled change proposal) rust.tf/libs380 *\`Result::ignore\`*

_Generated by [fully-automatic-rust-libs-team-triage-meeting-agenda-generator](https://github.com/rust-lang/libs-team/tree/main/tools/agenda-generator)_