Rollup of 8 pull requests
Successful merges:
- #108754 (Retry `pred_known_to_hold_modulo_regions` with fulfillment if ambiguous)
- #108759 (1.41.1 supported 32-bit Apple targets)
- #108839 (Canonicalize root var when making response from new solver)
- #108856 (Remove DropAndReplace terminator)
- #108882 (Tweak E0740)
- #108898 (Set `LIBC_CHECK_CFG=1` when building Rust code in bootstrap)
- #108911 (Improve rustdoc-gui/tester.js code a bit)
- #108916 (Remove an unused return value in `rustc_hir_typeck`)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
This was previously needed because the indirection used to hide some unexplained lifetime errors, which it turned out were related to the `min_choice` algorithm.
Removing the indirection also solves a couple of cycle errors, large moves and makes async blocks support the `#[track_caller]` annotation.
Add the `popular-crates` binary
This program downloads crates info from <https://crates.io/> and builds a TOML file that can be fed to `lintcheck`.
I have been asked, on various pull requests, what the result of `lintcheck` was. However, the default configuration file for lintcheck is limited. This `popular-crates` program allows building a recent list of the recently most downloaded crates from <https://crates.io> and feed it to `lintcheck`. Using it, it was easy to test two new lints against the 500 recently most downloaded crates to ensure that there was no regression.
changelog: none
Do not implement HashStable for HashSet (MCP 533)
This PR removes all occurrences of `HashSet` in query results, replacing it either with `FxIndexSet` or with `UnordSet`, and then removes the `HashStable` implementation of `HashSet`. This is part of implementing [MCP 533](https://github.com/rust-lang/compiler-team/issues/533), that is, removing the `HashStable` implementations of all collection types with unstable iteration order.
The changes are mostly mechanical. The only place where additional sorting is happening is in Miri's override implementation of the `exported_symbols` query.
Add `collection_is_never_read`
Fixes#9267
`@flip1995` and `@llogiq,` I talked with you about this one at Rust Nation in London last week. :-)
This is my first contribution to Clippy, so lots of feedback would be greatly appreciated.
- \[ ] Followed [lint naming conventions][lint_naming]
- \[x] Added passing UI tests (including committed `.stderr` file)
- \[x] `cargo test` passes locally
- \[x] Executed `cargo dev update_lints`
- \[x] Added lint documentation
- \[x] Run `cargo dev fmt`
`dogfood` found one true positive (see #9509) and no false positives.
`lintcheck` found no (true or false) positives, even when running on an extended set of crates.
---
changelog: new lint [`collection_is_never_read`]
[#10415](https://github.com/rust-lang/rust-clippy/pull/10415)
<!-- changelog_checked -->
rustc_middle: Remove trait `DefIdTree`
This trait was a way to generalize over both `TyCtxt` and `Resolver`, but now `Resolver` has access to `TyCtxt`, so this trait is no longer necessary.
Lintcheck maintenance
Make `cargo lintcheck -j 0` use all threads instead of panicking, and cleanup (and shorten) lintcheck's parsing code.
changelog: none
This makes the code shorter and clearer.
The only incompatible change is that an explicit command-line argument
`--crates-toml=` will take precedence over the `LINTCHECK_TOML`
environment variable.
Using `rayon::current_num_threads()` causes a bug:
```
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value:
ThreadPoolBuildError { kind: GlobalPoolAlreadyInitialized }',
src/main.rs:632:10
```
Moreover, using the number of threads and dividing it by 2 wouldn't
return the number of physical threads on modern processors which have
a varying number of threads per core.
It makes little sense to restrict ourselves to physical threads,
especially when, in modern architectures, cores with multiple threads
are often faster (performance) while cores with a unique threads are
often slower (efficient). The Rust runtime will make a better choice.
Downgrade let_underscore_untyped to restriction
From reading #6842 I am not convinced of the cost/benefit of this lint even as a pedantic lint.
It sounds like the primary motivation was to catch cases of `fn() -> Result` being changed to `async fn() -> Result`. If the original Result was ignored by a `let _`, then the compiler wouldn't guide you to add `.await`. **However, this situation is caught in a more specific way by [let_underscore_future](https://rust-lang.github.io/rust-clippy/master/index.html#let_underscore_future) which was introduced _after_ the original suggestion (#9760).**
In #10410 it was mentioned twice that a <kbd>restriction</kbd> lint might be more appropriate for let_underscore_untyped.
changelog: Moved [`let_underscore_untyped`] to restriction
Restrict `#[rustc_box]` to `Box::new` calls
Currently, `#[rustc_box]` can be applied to any call expression with a single argument. This PR only allows it to be applied to calls to `Box::new`
