…nstable

Add notes to the primitive integer and `NonZero` types clarifying the
equivalence between `highest_one` and the `ilog2`/`checked_ilog2`
methods for non-negative values.

- 调整 MaybeUninit::zeroed 文档中的 padding 字节说明

- 对齐 mem::zeroed 中不保证 padding 为零的表述

…ning

Make the linker quiet by default and add linker command line arguments
for explicit verbosity control:

- no flag: no tracing output is emitted during successful links
- `-v`: emit informational linker diagnostics
- `-vv`: emit tracing-level linker diagnostics

Also adjust the tracing formatter to produce linker-style output by disabling
ANSI color sequences and omitting timestamps, tracing levels, and tracing
targets.

* The documentation for `align_of_val()` and `Alignment::of_val()` did
  not explain its use in interacting with `dyn` types, and even
  contained the false statement that “Every reference to a value of the
  type `T` must be a multiple of this number”. This change removes that
  statement from everything except `align_of()`, and adds a mention
  of, and example code for, getting the alignment of a `dyn` value.

* The documentation for `align_of_val_raw()` did not explain how it
  relates to other functions in the family. Now it does.

* Added a caveat that the alignment of `i32` is not always 4, despite
  the examples asserting this.

Import resolution now happens in 2 phases:
1. We resolve all undetermined and collect their resolutions
2. Write all resolutions to the Resolver state.

Repeat this untill we reach a fix point.

+ Bless tests

… imports the correct traits for `aarch64`.

+ run the generator to apply change.

- take ownership of the `Vec<&Impl>` instead of copying into another alloc
- reuse `ImplString` to do natural sort ordering
- lazy formatting

…ic write_all_vectored

chore: Update annotate-snippets to 0.12.16

This PR updates `annotate-snippets` to [`0.12.16`](https://github.com/rust-lang/annotate-snippets-rs/blob/main/CHANGELOG.md#01216---2026-05-06) which fixes highlighting for indented code: rust-lang/annotate-snippets-rs#405

Also fixes rust-lang#155873 (the same problem)

…saethlin

In `copy_nonoverlapping`, use `mul nuw nsw` to compute the byte size

Seems like we might as well?  Adding these flags means the optimizer can tell the limited range on the count of items -- like how we use these flags (rust-lang#136575) when calculating `size_of_val` for a slice.

Today we use a wrapping multiplication, which mean that `copy_nonoverlapping::<u32>(src, dst, 0x40000000_00000001)` appears like 4 bytes -- a perfectly reasonable size! -- once it gets to the `memcpy` call.

If I'm understanding <https://doc.rust-lang.org/std/ptr/fn.copy_nonoverlapping.html#safety> properly, this is just exploiting existing UB, since `src` and `dst` must each be inside an allocation, and those allocations can be at most `isize::MAX` bytes.  (Plus, fundamentally, to be non-overlapping there's not enough space in the address space to be bigger than `isize::MAX`.)

cc @RalfJung to make sure this is ok, as requested last he found out I was newly exploiting some UB in codegen 🙃
r? codegen