delegation: fix cycles during delayed lowering
This PR forces lowering of delayed owners after `hir_crate_items`, as some diagnostics use `hir_crate_items` which results in query cycle which is then hangs calling `def_path_str` again and again. Fixesrust-lang/rust#154169. Part of rust-lang/rust#118212.
r? @petrochenkov
Fix pattern types rendering in rustdoc
Closesrust-lang/rust#150889 .
`core` was rendering local pattern-type impls through `clean_ty`, which formatted `rustc_hir::TyPat` with derived Debug, while inlined docs go through `clean_middle_ty` and get the pretty-printed `rustc_middle::ty::Pattern`.
Lower local HIR pattern types before formatting so both paths share the same canonical pattern printer and stop exposing `TyPat { ... }` in implementors.
#### Current result:
<img width="1133" height="738" alt="截屏2026-04-07 22 47 26" src="https://github.com/user-attachments/assets/bb5f8942-30b1-400a-a0a0-f581da5bb59f" />
Use fine grained component-wise span tracking in use trees
This often produces nicer spans and even doesn't need a Span field anymore (not that I expect the unused field to affect any perf, but still neat).
Cargo submodule update
11 commits in a357df4c26fc14514e66aae2a269456b5545c7db..101549dddbd2b08e806f50154e3aa4cb3374cc21
2026-04-03 16:47:15 +0000 to 2026-04-08 12:51:20 +0000
- Never include use extra-filename in build scripts (rust-lang/cargo#16855)
- fix(toml): Force script edition warnings on quiet (rust-lang/cargo#16848)
- GitHub fast path uses `http_async` (rust-lang/cargo#16847)
- feat(manifest): allow git dependency alongside alternate registry (rust-lang/cargo#16810)
- fix(auth): add auth scheme hint to token rejected error for alt registries (rust-lang/cargo#16794)
- Warn on invalid jobserver file descriptors (rust-lang/cargo#16843)
- docs(unstable): List the minimum required MSRV for 'public' field (rust-lang/cargo#16841)
- feat(lints): Emit unused_dependencies lint (rust-lang/cargo#16600)
- fix(tree): clarify error message when `-i` is used without a package name (rust-lang/cargo#16818)
- fix: Typo in target.<cfg>.linker (rust-lang/cargo#16839)
- Send Content-Type header with cargo publish requests (rust-lang/cargo#16832)
r? ghost
rustdoc: Inherit inline attributes for declarative macros
When explicitly re-exporting a declarative macro by name, rustdoc previously bypassed intermediate re-exports and dropped `#[doc(inline)]` attributes, causing the macro to be incorrectly stripped if the original definition was `#[doc(hidden)]`.
This updates `generate_item_with_correct_attrs` to walk the `reexport_chain` specifically for declarative macros, allowing them to inherit inline attributes exactly as glob imports do, while preserving strict visibility rules for standard items.
Fixesrust-lang/rust#154694
Update Fira Mono License Information
Update license information for Fira Mono font.
The license itself seems to be identical as `FiraSans` license, but didn't reuse it's file as it was explicitly named `FiraSans-LICENSE`.
Added a new LICENSE from https://github.com/mozilla/Fira/blob/master/LICENSE and updated `build` and `static` files setup.
`./x doc library` worked fine locally and was able to see the mono license being minified and served as well.
Anything that can change there can also be changed in bootstrap.toml.
We have a separate check to make sure that the local config is
compatable with CI's config.
Example output:
```
$ x b compiler
Building bootstrap
Finished `dev` profile [unoptimized] target(s) in 0.02s
NOTE: detected 3 modifications that could affect a build of rustc
- src/bootstrap/src/core/config/config.rs
- src/bootstrap/src/core/download.rs
- src/build_helper/src/git.rs
skipping rustc download due to `download-rustc = 'if-unchanged'`
Building stage1 compiler artifacts (stage0 -> stage1, aarch64-apple-darwin)
Finished `release` profile [optimized] target(s) in 0.72s
Creating a sysroot for stage1 compiler (use `rustup toolchain link 'name' build/host/stage1`)
Build completed successfully in 0:00:05
```
`ty::Alias` refactor
This PR changes the following alias-related types from this:
```rust
pub enum AliasTyKind {
Projection,
Inherent,
Opaque,
Free,
}
pub struct AliasTy<I: Interner> {
pub args: I::GenericArgs,
pub def_id: I::DefId,
}
pub enum TyKind<I: Interner> {
...
Alias(AliasTyKind, AliasTy<I>),
}
```
Into this:
```rust
pub enum AliasTyKind<I: Interner> {
Projection { def_id: I::DefId },
Inherent { def_id: I::DefId },
Opaque { def_id: I::DefId },
Free { def_id: I::DefId },
}
pub struct AliasTy<I: Interner> {
pub args: I::GenericArgs,
pub kind: AliasTyKind<I>,
}
pub enum TyKind<I: Interner> {
...
Alias(AliasTy<I>),
}
```
... and then does a thousand other changes to accommodate for this change everywhere.
This brings us closer to being able to have `AliasTyKind`s which don't require a `DefId` (and thus can be more easily created, etc). Although notably we depend on both `AliasTyKind -> DefId` and `DefId -> AliasTyKind` conversions in a bunch of places still.
r? lcnr
----
A lot of these changes were done either by search & replace (via `ast-grep`) or on auto pilot, so I'm not quite sure I didn't mess up somewhere, but at least tests pass...
Lewis McClelland