…uwer

Rollup of 13 pull requests

Successful merges:

 - #156709 (stdarch subtree update)
 - #155006 (stabilize `feature(cfg_target_has_atomic_equal_alignment)`)
 - #156444 (Implement `OsStr::split_at`)
 - #156582 (Allow `global_asm!` in statement positions)
 - #156661 (Remove `UncheckedIterator`)
 - #152367 (Derives `Copy` for `ffi::FromBytesUntilNulError`)
 - #156443 (Fix invalid suggestion for parenthesized break)
 - #156606 (Add pext/pdep as aliases for extract_bits/deposit_bits)
 - #156630 (Replace `println!` with `assert!` in HashMap documentation examples)
 - #156644 (Widen the result of `widening_mul`.)
 - #156653 (Update `sysinfo` version to `0.39.2`)
 - #156697 (Add diagnostic items for IoBufReader and StdinLock)
 - #156704 (reduce usage of `box_patterns` in tests)

Rollup merge of #156582 - daxpedda:global-asm-statement, r=petrochenkov

Allow `global_asm!` in statement positions

This PR makes it possible to put `global_asm!` in statement positions. This is particularly useful for proc-macros, where you otherwise have to wrap them in `mod foo { global_asm!(...); }`.

I'm happy to open an ACP first (or a design meeting?).
I would also assume this needs sign-off from the lang-team?

Previously discussed on [Zulip](https://rust-lang.zulipchat.com/#narrow/channel/216763-project-inline-asm/topic/Item.20position.20global_asm/with/581784943).

r? @Amanieu

…-item, r=workingjubilee

Revert "Allow `global_asm!` in statement positions"

This reverts commit 450cdb5.

Based on discussion in [#t-lang > insta-stable &rust-lang#96;global_asm!&rust-lang#96; as inner items](https://rust-lang.zulipchat.com/#narrow/channel/213817-t-lang/topic/insta-stable.20.60global_asm.21.60.20as.20inner.20items/with/597411711), rust-lang#156582 should not have been merged without FCP.

We've also since found a bug in the implementation (or really, an assumption in the existing code that the PR invalidates), so to re-land this

- new behavior should be behind a feature gate
- the fix in rust-lang#156855 (comment) should be included (and reviewed by someone familiar with `rustc_resolve`)

Then we keep it unstable for a bit, and see if the fuzzer finds anything else. Nothing here is particularly controversial I think, but we need to do our due diligence.

r? @ghost

Rollup merge of #156884 - folkertdev:revert-global-asm-inner-item, r=workingjubilee

Revert "Allow `global_asm!` in statement positions"

This reverts commit 450cdb5.

Based on discussion in [#t-lang > insta-stable `global_asm!` as inner items](https://rust-lang.zulipchat.com/#narrow/channel/213817-t-lang/topic/insta-stable.20.60global_asm.21.60.20as.20inner.20items/with/597411711), #156582 should not have been merged without FCP.

We've also since found a bug in the implementation (or really, an assumption in the existing code that the PR invalidates), so to re-land this

- new behavior should be behind a feature gate
- the fix in #156855 (comment) should be included (and reviewed by someone familiar with `rustc_resolve`)

Then we keep it unstable for a bit, and see if the fuzzer finds anything else. Nothing here is particularly controversial I think, but we need to do our due diligence.

r? @ghost