…ror-pattern`

Resolves Issue rust-lang#118752

It prevents a full rebuild of stage 1 compiler when issuing "x.py test"
with rust.lto != thin-local in config.toml.

…port

Co-authored-by: León Orell Valerian Liehr <me@fmease.dev>

…, r=davidtwco

Switch from using `//~ERROR` annotations with `--error-format` to `error-pattern`

Fixes rust-lang#118752

As noticed by `@jyn514` while working on a patch, tests failed due to `//~ERROR` annotations used in combination with the older `--error-format` which is now `error-pattern`.

…rors

custom mir: make it clear what the return block is

Custom MIR recently got support for specifying the "unwind action", so now there's two things coming after the actual call part of `Call` terminators. That's not very self-explaining so I propose we change the syntax to imitate keyword arguments:
```
Call(popped = Vec::pop(v), ReturnTo(drop), UnwindContinue())
```

Also fix some outdated docs and add some docs to `Call` and `Drop`.

Use Result::flatten in catch_with_exit_code

…=fmease

Recover parentheses in range patterns

Before:
```rs
match n {
    (0).. => (),
    _ => ()
}
```
```
error: expected one of `=>`, `if`, or `|`, found `..`
 --> src/lib.rs:3:12
  |
3 |         (0).. => (),
  |            ^^ expected one of `=>`, `if`, or `|`
  ```
 After:
```
error: range pattern bounds cannot have parentheses
 --> main.rs:3:5
  |
3 |     (0).. => (),
  |     ^ ^
  |
help: remove these parentheses
  |
3 -     (0).. => (),
3 +     0.. => (),
  |
```

This sets the groundwork for rust-lang#118625, which will extend the recovery to expressions like `(0 + 1)..` where users may tend to add parentheses to avoid dealing with precedence.

---

`@rustbot` label +A-parser +A-patterns +A-diagnostics

…ulacrum

bootstrap: Move -Clto= setting from Rustc::run to rustc_cargo

It prevents a full rebuild of stage 1 compiler when issuing "x.py test" with rust.lto != thin-local in config.toml.

…r=davidtwco

Uplift some miscellaneous coroutine-specific machinery into `check_closure`

This PR uplifts some of the logic in `check_fn` that is specific to checking coroutines, which always flows through `check_closure`.

This is just some miscellaneous clean up that I've wanted to do, especially because I'm poking around this code to make it work for async closures.

don't reexport atomic::ordering via rustc_data_structures, use std import

This looks simpler.

…rait, r=compiler-errors

Don't synthesize host effect args inside trait object types

While we were indeed emitting an error for `~const` & `const` trait bounds in trait object types, we were still synthesizing host effect args for them.

Since we don't record the original trait bound modifiers for dyn-Trait in `hir::TyKind::TraitObject` (unlike we do for let's say impl-Trait, `hir::TyKind::OpaqueTy`), AstConv just assumes `ty::BoundConstness::NotConst` in `conv_object_ty_poly_trait_ref` which given `<host> dyn ~const NonConstTrait` resulted in us not realizing that `~const` was used on a non-const trait which lead to a failed assertion in the end.

Instead of updating `hir::TyKind::TraitObject` to track this kind of information, just strip the user-provided constness (similar to rust-lang#119505).

Fixes rust-lang#119524.