Rollup of 10 pull requests
Successful merges:
- #118833 (Add lint against function pointer comparisons)
- #122161 (Fix suggestion when shorthand `self` has erroneous type)
- #133233 (Add context to "const in pattern" errors)
- #133761 (Update books)
- #133843 (Do not emit empty suggestion)
- #133863 (Rename `core_pattern_type` and `core_pattern_types` lib feature gates to `pattern_type_macro`)
- #133872 (No need to create placeholders for GAT args in confirm_object_candidate)
- #133874 (`fn_sig_for_fn_abi` should return a `ty::FnSig`, no need for a binder)
- #133890 (Add a new test ui/incoherent-inherent-impls/no-other-unrelated-errors to check E0116 does not cause unrelated errors)
- #133892 (Revert #133817)
r? `@ghost`
`@rustbot` modify labels: rollup
Revert #133817
This reverts commit 0585134e70, reversing changes made to 5530869e0f.
#133817 unfortunately only converted the `println!` instances to `eprintln!`, meaning that some test output (via compiletest/bootstrap) was messed up because stdout/stderr output interleaved improperly when some `println!` instances were converted to `eprintln!` instances, while some `print!` instances remain unchanged. This made reading test output annoying for contributors cc #133879.
Closes#133879 by reverting.
#133817 can be relanded in the future when `print!` instances are also matched with `println!` instances.
cc `@clubby789`
This is a clean revert so I'm going to self-approve this PR.
Add a new test ui/incoherent-inherent-impls/no-other-unrelated-errors to check E0116 does not cause unrelated errors
rustc xxx (we do not know) to 1.82.0 emits unrelated errors when E0116 is present (see #125814).
We do not know what caused and fixed it, but add a test to confirm rustc does not cause the same error in the future.
No need to create placeholders for GAT args in confirm_object_candidate
We no longer need this logic to add placeholders for GAT args since with the removal of the `gat_extended` feature gate (https://github.com/rust-lang/rust/pull/133768) we no longer allow GATs in dyn trait anyways.
r? oli-obk
Rename `core_pattern_type` and `core_pattern_types` lib feature gates to `pattern_type_macro`
That's what the gates are actually gating, and the single char difference in naming was not helpful either
fixes#128987
Do not emit empty suggestion
The `println!();` statement's span doesn't include the `;`, and the modified suggestions where trying to get the `;` by getting the differenece between the statement's and the expression's spans, which was an empty suggestion.
Fix#133833, fix#133834.
Update books
## rust-lang/book
12 commits in e16dd73690a6cc3ecdc5f5d94bbc3ce158a42e16..614c19cb4025636eb2ba68ebb3d44e3bd3a5e6e4
2024-12-02 16:22:15 UTC to 2024-11-26 21:15:51 UTC
- Fix more inverted uses of “shadowed” (rust-lang/book#4122)
- Fix a couple inverted uses of “shadowed” (rust-lang/book#4121)
- Fix confusion between variable `hello` and string `"hello"` (rust-lang/book#4118)
- Ch. 17: fix some wording issues (rust-lang/book#4117)
- Rephrase for clarity (rust-lang/book#3809)
- Link to tests section of rustc book for `cargo test -- --help` docs (rust-lang/book#4116)
- Drop a pedantry-triggering sentence about IEEE-754 (rust-lang/book#4114)
- Very small clarification on if let (rust-lang/book#4113)
- Ch17-05: Typos (rust-lang/book#4099)
- Ch20-01: Fix typos (rust-lang/book#4105)
- Add a short paragraph on editors and IDEs in installation (rust-lang/book#4112)
- Rust 2024: update Ch. 20 for new `unsafe` rules (rust-lang/book#4111)
## rust-lang/reference
8 commits in 5c86c739ec71b8bc839310ff47fa94e94635bba9..ede56d1bbe132bac476b5029cd6d7508ca9572e9
2024-11-25 17:23:35 +0000 to 2024-12-03 22:26:55 +0000
- Claim to follow Unicode 16 for lexing identifiers. (rust-lang/reference#1688)
- Clarify rules for on_unimplemented warnings (rust-lang/reference#1680)
- Enable triagebot merge-conflict notifications (rust-lang/reference#1682)
- Update default edition to 2024 for code examples (rust-lang/reference#1684)
- Fix weak keywords (rust-lang/reference#1685)
- `const` expression can borrow static items (rust-lang/reference#1610)
- Update function-pointer.md for stabilization of `extended_varargs_abi_support` (rust-lang/reference#1687)
- fix inconsistent spacing in example (rust-lang/reference#1686)
## edition-guide
1 commits in f48b0e842a3911c63240e955d042089e9e0894c7..128669297c8a7fdf771042eaec18b8adfaeaf0cd
2024-11-25 16:20:27 +0000 to 2024-12-03 22:02:43 +0000
- Fix `if_let_rescope` applicability (rust-lang/edition-guide#339)
## rust-lang/rustc-dev-guide
6 commits in 787b4166ccc67bd8f72a6e3ef6685ce9ce82909a..b21d99b770f9aceb0810c843847c52f86f45d2ed
2024-12-02 04:45:30 UTC to 2024-11-27 10:31:58 UTC
- Spell out `git submodule deinit -f --all` (rust-lang/rustc-dev-guide#2153)
- Explain how to deal with exploded git submodules (rust-lang/rustc-dev-guide#2152)
- Update `//@ proc-macro` aux build directive docs (rust-lang/rustc-dev-guide#2149)
- Remove `pretty-expanded` as it no longer exists (rust-lang/rustc-dev-guide#2147)
- Fix trivial typo (rust-lang/rustc-dev-guide#2148)
- Remove -Zfuel. (rust-lang/rustc-dev-guide#2032)
Add context to "const in pattern" errors
*Each commit addresses specific diagnostics.*
- Add primary span labels
- Point at `const` item, and `const` generic param definition
- Reword messages and notes
- Point at generic param through which an associated `const` is being referenced
- Silence const in pattern with evaluation errors when they come from `const` items that already emit a diagnostic
- On non-structural type in const used as pattern, point at the type that should derive `PartialEq`
Add lint against function pointer comparisons
This is kind of a follow-up to https://github.com/rust-lang/rust/pull/117758 where we added a lint against wide pointer comparisons for being ambiguous and unreliable; well function pointer comparisons are also unreliable. We should IMO follow a similar logic and warn people about it.
-----
## `unpredictable_function_pointer_comparisons`
*warn-by-default*
The `unpredictable_function_pointer_comparisons` lint checks comparison of function pointer as the operands.
### Example
```rust
fn foo() {}
let a = foo as fn();
let _ = a == foo;
```
### Explanation
Function pointers comparisons do not produce meaningful result since they are never guaranteed to be unique and could vary between different code generation units. Furthermore different function could have the same address after being merged together.
----
This PR also uplift the very similar `clippy::fn_address_comparisons` lint, which only linted on if one of the operand was an `ty::FnDef` while this PR lints proposes to lint on all `ty::FnPtr` and `ty::FnDef`.
```@rustbot``` labels +I-lang-nominated
~~Edit: Blocked on https://github.com/rust-lang/libs-team/issues/323 being accepted and it's follow-up pr~~
This reverts commit 0585134e70, reversing
changes made to 5530869e0f.
The PR unfortunately only converted the `ln!` instances, meaning that
test output was messed up because stdout/stderr output interleaved when
some `println!` instances were converted to `eprintln!` instances, while
some `println!` instances remain unchanged.
Make sure to record deps from cached task in new solver on first run
We weren't actually performing a read of the dep node in `with_cached_task` in the new solver, which meant that all queries that computed a goal for the first time were just not recording the query dependencies that we call in that query.
In the incremental test, the typeck query for `fn poll` isn't being marked red even tho it's invalidated due to its writeback results changing. This happens b/c we normalize `Self::Error` into `Error`, which should call `type_of` which is a red query (since `ty::Adt` contains an `AdtDef`, and that `AdtDef`'s stable hash changes since it's ). However, since we weren't tracking deps in that normalize query, the typeck result was remaining green, and we were trying to decode a def id that no longer exists (the field that got removed).
r? lcnr
rustc xxx (we do not know) to 1.82.0 emits unrelated errors when E0116
is present (see #125814).
We do not know what caused and fixed it, but add a test to confirm rustc
does not cause the same error in the future.
Point at types that need to be marked with `#[derive(PartialEq)]`.
We use a visitor to look at a type that isn't structural, looking for all ADTs that don't derive `PartialEq`. These can either be manual `impl PartialEq`s or no `impl` at all, so we differentiate between those two cases to provide more context to the user. We also only point at types and impls from the local crate, otherwise show only a note.
```
error: constant of non-structural type `&[B]` in a pattern
--> $DIR/issue-61188-match-slice-forbidden-without-eq.rs:15:9
|
LL | struct B(i32);
| -------- must be annotated with `#[derive(PartialEq)]` to be usable in patterns
LL |
LL | const A: &[B] = &[];
| ------------- constant defined here
...
LL | A => (),
| ^ constant of non-structural type
|
= note: see https://doc.rust-lang.org/stable/std/marker/trait.StructuralPartialEq.html for details
```
- Point at type that should derive `PartialEq` to be structural.
- Point at manual `impl PartialEq`, explaining that it is not sufficient to be structural.
```
error: constant of non-structural type `MyType` in a pattern
--> $DIR/const-partial_eq-fallback-ice.rs:14:12
|
LL | struct MyType;
| ------------- `MyType` must be annotated with `#[derive(PartialEq)]` to be usable in patterns
...
LL | const CONSTANT: &&MyType = &&MyType;
| ------------------------ constant defined here
...
LL | if let CONSTANT = &&MyType {
| ^^^^^^^^ constant of non-structural type
|
note: the `PartialEq` trait must be derived, manual `impl`s are not sufficient; see https://doc.rust-lang.org/stable/std/marker/trait.StructuralPartialEq.html for details
--> $DIR/const-partial_eq-fallback-ice.rs:5:1
|
LL | impl PartialEq<usize> for MyType {
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
```
```
error: trait object `dyn Send` cannot be used in patterns
--> $DIR/issue-70972-dyn-trait.rs:6:9
|
LL | const F: &'static dyn Send = &7u32;
| -------------------------- constant defined here
...
LL | F => panic!(),
| ^ trait object can't be used in patterns
```
- Add primary span labels.
- Point at const generic parameter used as pattern.
- Point at statics used as pattern.
- Point at let bindings used in const pattern.
Centralize emitting an error in `const_to_pat` so that all errors from that evaluating a `const` in a pattern can add addditional information. With this, now point at the `const` item's definition:
```
error[E0158]: constant pattern depends on a generic parameter
--> $DIR/associated-const-type-parameter-pattern.rs:20:9
|
LL | pub trait Foo {
| -------------
LL | const X: EFoo;
| ------------- constant defined here
...
LL | A::X => println!("A::X"),
| ^^^^
```
The `println!();` statement's span doesn't include the `;`, and the modified suggestions where trying to get the `;` by getting the differenece between the statement's and the expression's spans, which was an empty suggestion.
Fix#133833, fix#133834.
Rollup of 8 pull requests
Successful merges:
- #133737 (Include LLDB and GDB visualizers in MSVC distribution)
- #133774 (Make CoercePointee errors translatable)
- #133831 (Don't try and handle unfed `type_of` on anon consts)
- #133847 (Remove `-Zshow-span`.)
- #133849 (coverage: Use a separate counter type and simplification step during counter creation)
- #133850 (Avoid `opaque type not constrained` errors in the presence of other errors)
- #133851 (Stop git from merging generated files)
- #133856 (Update sysinfo version to 0.33.0)
r? `@ghost`
`@rustbot` modify labels: rollup
Stop git from merging generated files
See the relevant documentation for details: https://git-scm.com/docs/gitattributes#Documentation/gitattributes.txt-Unset-1-1-1
This will cause `.stderr`, `.stdout`, `.mir`, and `.fixed` files to stop generating conflict markers and instead just pick the version from your branch, not the one you're rebasing over. Git will still raise a conflict and thus stop a rebase at that commit, but it avoids having to deal with conflict markers in files where we just bless them away anyway.
Avoid `opaque type not constrained` errors in the presence of other errors
pulled out of https://github.com/rust-lang/rust/pull/128440
These errors carry no new information if the opaque type was actually used in a constraining (but erroneous) way somewhere.
coverage: Use a separate counter type and simplification step during counter creation
When instrumenting a function's MIR for coverage, there is a point where we need to decide, for each node in the control-flow graph, whether its execution count will be tracked by a physical counter, or by an expression that combines physical counters from other parts of the graph.
Currently the code for doing that is heavily tied to the final form of the LLVM coverage mapping format, and performs some important simplification steps on-the-fly. These factors make the code extremely difficult to modify without breaking or massively worsening the resulting coverage-instrumentation metadata.
---
This PR aims to improve that situation somewhat by adding an extra intermediate representation between the code that chooses how each node will be counted, and the code that converts those decisions into actual tables of physical counters and trees of counter expressions.
As part of doing that, some of the simplifications that are currently performed during the main counter creation step have been pulled out into a separate step.
In most cases the resulting coverage metadata is equivalent, slightly better, or slightly worse. The biggest outlier is `counters.rs`, where the coverage metadata ends up about 10% larger. This seems to be the result of the new approach having less subexpression sharing (because it relies on flatten-sort-cancel), and therefore being less effective at taking advantage of MIR optimizations to replace counters for unused control-flow with zeroes. I think the modest downside is acceptable in light of the future possibilities opened up by this decoupling.
Remove `-Zshow-span`.
It's very old (added in #12087). It's strange, and it's not clear what its use cases are. It only works with the crate root file because it runs before expansion. I suspect it won't be missed.
r? `@estebank`
Don't try and handle unfed `type_of` on anon consts
The `type_of` query for anon consts in the type system is actually implemented by feeding the return value during hir ty lowering, not the hir-based logic in `const_arg_anon_type_of`. The HIR based logic is incomplete (doesn't handle all hir nodes) and also generally wrong to call (re-lowers HIR or invokes typeck which can result in query cycles).
r? `@compiler-errors`
Make CoercePointee errors translatable
Tracked by #123430
Just in case that a translatable error message would become a blocker to stabilization, this PR switches over to fluent error messages, which also slightly improve the wordings and use more accurate span information.
cc `@Darksonn` `@traviscross`
Include LLDB and GDB visualizers in MSVC distribution
MSVC distributions currently don't include the lldb or GDB python files. MSVC and LLDB/GDB are not mutually exclusive (and end up being a common case with vscode + codelldb/lldb-dap), so they should probably be included.
the existing visualizers currently only partially work on MSVC due to the differences in how the debug info is generated, but they also only partially work on GNU anyway - both of which are actively being fixed.
