Restructure `rustc_type_ir` a bit
1. Split `sty` into new `ty_kind`/`region_kind`/`const_kind` modules, so that when we uplift more kinds (e.g. `PredicateKind`, `ClauseKind`, and `ExistentialPredicate`), they can live in their own simple-to-understand files.
2. Split up the `structural_impls` module, which is a kitchen sink of random impls -- move `TypeFoldable` and `TypeVisitable` impls into existing `fold` and `visit` modules, respectively.
3. Move the `DebugWithInfcx` trait and blanket impls into a new `debug` module, and `TypeFlags` definition into a new `flags` module.
5. Move `Interner` trait into a new `interner` module. I expect this file to get a lot larger as we make the interner more powerful for the trait solver refactor.
r? `@ghost` for now, will assign once #116946 lands
Move where doc comment meant as comment check
The new place makes more sense and covers more cases beyond individual statements.
```
error: expected one of `.`, `;`, `?`, `else`, or an operator, found doc comment `//!foo
--> $DIR/doc-comment-in-stmt.rs:25:22
|
LL | let y = x.max(1) //!foo
| ^^^^^^ expected one of `.`, `;`, `?`, `else`, or an operator
|
help: add a space before `!` to write a regular comment
|
LL | let y = x.max(1) // !foo
| +
```
Fix#65329.
Uplift movability and mutability, the simple way
Just make type_ir a dependency of ast. This can be relaxed later if we want to make the dependency less heavy. Part of rust-lang/types-team#124.
r? `@lcnr` or `@jackh726`
Add a test showing failing closure signature inference in new solver
Been thinking a bit about how to make this test pass... but we don't actually have any good tests exercising this behavior in the suite.
r? lcnr
Specialize `Bytes<R>::next` when `R` is a `BufReader`.
This reduces the runtime for a simple program using `Bytes::next` to iterate through a file from 220ms to 70ms on my Linux box.
r? `@the8472`
The new place makes more sense and covers more cases beyond individual
statements.
```
error: expected one of `.`, `;`, `?`, `else`, or an operator, found doc comment `//!foo
--> $DIR/doc-comment-in-stmt.rs:25:22
|
LL | let y = x.max(1) //!foo
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ expected one of `.`, `;`, `?`, `else`, or an operator
|
help: add a space before `!` to write a regular comment
|
LL | let y = x.max(1) // !foo
| +
```
Fix#65329.
Fix duplicate labels emitted in `render_multispan_macro_backtrace()`
This PR replaces the `Vec` used to store labels with an `FxIndexSet` in order to eliminate duplicates
Fixes#116836
Implement rustc part of RFC 3127 trim-paths
This PR implements (or at least tries to) [RFC 3127 trim-paths](https://github.com/rust-lang/rust/issues/111540), the rustc part. That is `-Zremap-path-scope` with all of it's components/scopes.
`@rustbot` label: +F-trim-paths
Add FileCheck annotations to mir-opt tests.
This PR makes compiletest run LLVM `FileCheck` tool on mir-opt tests.
The check is *run by default*, except if disabled using `// skip-filecheck` comment. This ensures that we do not have a silently broken test. For now, the check is only run on the output of `--emit=mir`, ie. on PreCodegen MIR.
I give an example on `reference_prop.rs`.
r? `@oli-obk`
cc `@RalfJung`
Fixes https://github.com/rust-lang/rust/issues/85180
Rollup of 4 pull requests
Successful merges:
- #116650 (add some comments and some cleanup around Miri intptrcast)
- #116896 (Only check in a single place if a pass is enabled.)
- #116906 (Use v0.0.0 in compiler crates)
- #116921 (fix(bootstrap) info message show correct path now)
r? `@ghost`
`@rustbot` modify labels: rollup
Use v0.0.0 in compiler crates
I may be totally off base here, but my understanding is that it's conventional to use v0.0.0 to reflect the unversioned nature of the compiler crates. Fix that for some of the compiler crates that were created recently.
Make TCP connect handle EINTR correctly
According to the [POSIX](https://pubs.opengroup.org/onlinepubs/009695399/functions/connect.html) standard, if connect() is interrupted by a signal that is caught while blocked waiting to establish a connection, connect() shall fail and set errno to EINTR, but the connection request shall not be aborted, and the connection shall be established asynchronously. When the connection has been established asynchronously, select() and poll() shall indicate that the file descriptor for the socket is ready for writing.
The previous implementation differs from the recomendation: in a case of the EINTR we tried to reconnect in a loop and sometimes get EISCONN error (this problem was originally detected on MacOS).
1. More details about the problem in an [article](http://www.madore.org/~david/computers/connect-intr.html).
2. The original [issue](https://git.picodata.io/picodata/picodata/tarantool-module/-/issues/157).
