Report undeclared lifetimes during late resolution.
First step in https://github.com/rust-lang/rust/pull/91557
We reuse the rib design of the current resolution framework. Specific `LifetimeRib` and `LifetimeRibKind` types are introduced. The most important variant is `LifetimeRibKind::Generics`, which happens each time we encounter something which may introduce generic lifetime parameters. It can be an item or a `for<...>` binder. The `LifetimeBinderKind` specifies how this rib behaves with respect to in-band lifetimes.
r? `@petrochenkov`
Remove last vestiges of skippng ident span hashing
This removes a comment that no longer applies, and properly hashes
the full ident for path segments.
Refactor HIR item-like traversal (part 1)
Issue #95004
- Create hir_crate_items query which traverses tcx.hir_crate(()).owners to return a hir::ModuleItems
- use tcx.hir_crate_items in tcx.hir().items() to return an iterator of hir::ItemId
- use tcx.hir_crate_items to introduce a tcx.hir().par_items(impl Fn(hir::ItemId)) to traverse all items in parallel;
Signed-off-by: Miguel Guarniz <mi9uel9@gmail.com>
cc `@cjgillot`
Strict provenance lint diagnostics improvements
Use `multipart_suggestion` instead of `span_suggestion` and getting a snippet for the expression. Also don't suggest unnecessary parenthesis in `lossy_provenance_casts`.
cc ``@estebank``
``@rustbot`` label A-diagnostics
rustc_metadata: Do not encode unnecessary module children
This should remove the syntax context shift and the special case for `ExternCrate` in decoder in https://github.com/rust-lang/rust/pull/95880.
This PR also shifts some work from decoding to encoding, which is typically useful for performance (but probably not much in this case).
r? `@cjgillot`
Include Refs in Valtree Creation
This adds references to `const_to_valtree`, which isn't used in the compiler yet, but after the previous changes we made to the thir and mir representations and this change we should be able to finally introduce them in the next PR.
I wasn't able to properly test this code, except indirectly by including a call of `const_to_valtree` in the code that currently creates constants (`turn_into_const_value`).
r? `@lcnr`
cc `@oli-obk` `@RalfJung`
Parse inner attributes on inline const block
According to https://github.com/rust-lang/rust/pull/84414#issuecomment-826150936, inner attributes are intended to be supported *"in all containers for statements (or some subset of statements)"*.
This PR adds inner attribute parsing and pretty-printing for inline const blocks (https://github.com/rust-lang/rust/issues/76001), which contain statements just like an unsafe block or a loop body.
```rust
let _ = const {
#![allow(...)]
let x = ();
x
};
```
resolve: Create dummy bindings for all unresolved imports
Apparently such bindings weren't previously created for all unresolved imports, causing issues like https://github.com/rust-lang/rust/issues/95879.
In this PR I'm trying to create such dummy bindings in a more centralized way by calling `import_dummy_binding` once for all imports in `finalize_imports`.
Fixes https://github.com/rust-lang/rust/issues/95879.
Allow self-profiler to only record potentially costly arguments when argument recording is turned on
As discussed [on zulip](https://rust-lang.zulipchat.com/#narrow/stream/247081-t-compiler.2Fperformance/topic/Identifying.20proc-macro.20slowdowns/near/277304909) with `@wesleywiser,` I'd like to record proc-macro expansions in the self-profiler, with some detailed data (per-expansion spans for example, to follow #95473).
At the same time, I'd also like to avoid doing expensive things when tracking a generic activity's arguments, if they were not specifically opted into the event filter mask, to allow the self-profiler to be used in hotter contexts.
This PR tries to offer:
- a way to ensure a closure to record arguments will only be called in that situation, so that potentially costly arguments can still be recorded when needed. With the additional requirement that, if possible, it would offer a way to record non-owned data without adding many `generic_activity_with_arg_{...}`-style methods. This lead to the `generic_activity_with_arg_recorder` single entry-point, and the closure parameter would offer the new methods, able to be executed in a context where costly argument could be created without disturbing the profiled piece of code.
- some facilities/patterns allowing to record more rustc specific data in this situation, without making `rustc_data_structures` where the self-profiler is defined, depend on other rustc crates (causing circular dependencies): in particular, spans. They are quite tricky to turn into strings (if the default `Debug` impl output does not match the context one needs them for), and since I'd also like to avoid the allocation there when arg recording is turned off today, that has turned into another flexibility requirement for the API in this PR (separating the span-specific recording into an extension trait). **edit**: I've removed this from the PR so that it's easier to review, and opened https://github.com/rust-lang/rust/pull/95739.
- allow for extensibility in the future: other ways to record arguments, or additional data attached to them could be added in the future (e.g. recording the argument's name as well as its data).
Some areas where I'd love feedback:
- the API and names: the `EventArgRecorder` and its method for example. As well as the verbosity that comes from the increased flexibility.
- if I should convert the existing `generic_activity_with_arg{s}` to just forward to `generic_activity_with_arg_recorder` + `recorder.record_arg` (or remove them altogether ? Probably not): I've used the new API in the simple case I could find of allocating for an arg that may not be recorded, and the rest don't seem costly.
- [x] whether this API should panic if no arguments were recorded by the user-provided closure (like this PR currently does: it seems like an error to use an API dedicated to record arguments but not call the methods to then do so) or if this should just record a generic activity without arguments ?
- whether the `record_arg` function should be `#[inline(always)]`, like the `generic_activity_*` functions ?
As mentioned, r? `@wesleywiser` following our recent discussion.
Adding diagnostic data on generators to the crate metadata and using it to provide
a better diagnostic on failure to meet send bound on futures originated from a foreign crate
Rollup of 9 pull requests
Successful merges:
- #93969 (Only add codegen backend to dep info if -Zbinary-dep-depinfo is used)
- #94605 (Add missing links in platform support docs)
- #95372 (make unaligned_references lint deny-by-default)
- #95859 (Improve diagnostics for unterminated nested block comment)
- #95961 (implement SIMD gather/scatter via vector getelementptr)
- #96004 (Consider lifetimes when comparing types for equality in MIR validator)
- #96050 (Remove some now-dead code that was only relevant before deaggregation.)
- #96070 ([test] Add test cases for untested functions for BTreeMap)
- #96099 (MaybeUninit array cleanup)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Better method call error messages
Rebase/continuation of #71827
~Based on #92360~
~Based on #93118~
There's a decent description in #71827 that I won't copy here (for now at least)
In addition to rebasing, I've tried to restore most of the original suggestions for invalid arguments. Unfortunately, this does make some of the errors a bit verbose. To fix this will require a bit of refactoring to some of the generalized error suggestion functions, and I just don't have the time to go into it right now.
I think this is in a state that the error messages are overall better than before without a reduction in the suggestions given.
~I've tried to split out some of the easier and self-contained changes into separate commits (mostly in #92360, but also one here). There might be more than can be done here, but again just lacking time.~
r? `@estebank` as the original reviewer of #71827
This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes.
The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other.
We then modify that algorithm to detect 4 cases:
- A function input is missing
- An extra argument was provided
- The type of an argument is straight up invalid
- Two arguments have been swapped
- A subset of the arguments have been shuffled
(We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.)
It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site.
The basic sketch of the algorithm is as follows:
- Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input.
- If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type".
- Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument.
- Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input
- Swapped / Permuted arguments are identified with a cycle detection algorithm.
As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again.
Note that there's a lot of extra complexity:
- We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix.
- Closure arguments are wrapped in a tuple and need to be unwrapped
- We need to resolve closure types after the rest, to allow the most specific type constraints
- We need to handle imported C functions that might be variadic in their inputs.
I tried to document a lot of this in comments in the code and keep the naming clear.
Remove some now-dead code that was only relevant before deaggregation.
The code was broken anyway, if the deaggregator is disabled, it would have ICEd on any non-enum Adt
r? ```@RalfJung```
implement SIMD gather/scatter via vector getelementptr
Fixes https://github.com/rust-lang/portable-simd/issues/271
However, I don't *really* know what I am doing here... Cc ``@workingjubilee`` ``@calebzulawski``
I didn't do anything for cranelift -- ``@bjorn3`` not sure if it's okay for that backend to temporarily break. I'm happy to cherry-pick a patch that adds cranelift support. :)
Improve diagnostics for unterminated nested block comment
close#95283
(This is my first time try to messing around with rust compiler and might get a lot of things wrong... 🙇 )
make unaligned_references lint deny-by-default
This lint has been warn-by-default for a year now (since https://github.com/rust-lang/rust/pull/82525), so I think it is time to crank it up a bit. Code that triggers the lint causes UB (without `unsafe`) when executed, so we really don't want people to write code like this.
Only add codegen backend to dep info if -Zbinary-dep-depinfo is used
I am currently migrating the cg_clif build system from using a binary linked to the codegen backend as rustc replacement to passing `-Zcodegen-backend` instead. Without this PR this would force cargo to rebuild the sysroot on any change to the codegen backend even if I explicitly specify that I want it to be preserved, which would make development of cg_clif a lot slower. If you still want to have changes to the codegen backend invalidate the cargo build cache you can explicitly specify `-Zbinary-dep-depinfo`.
cc ``@eddyb`` as the codegen backend was initially added to the depinfo for rust-gpu.
Implement sym operands for global_asm!
Tracking issue: #93333
This PR is pretty much a complete rewrite of `sym` operand support for inline assembly so that the same implementation can be shared by `asm!` and `global_asm!`. The main changes are:
- At the AST level, `sym` is represented as a special `InlineAsmSym` AST node containing a path instead of an `Expr`.
- At the HIR level, `sym` is split into `SymStatic` and `SymFn` depending on whether the path resolves to a static during AST lowering (defaults to `SynFn` if `get_early_res` fails).
- `SymFn` is just an `AnonConst`. It runs through typeck and we just collect the resulting type at the end. An error is emitted if the type is not a `FnDef`.
- `SymStatic` directly holds a path and the `DefId` of the `static` that it is pointing to.
- The representation at the MIR level is mostly unchanged. There is a minor change to THIR where `SymFn` is a constant instead of an expression.
- At the codegen level we need to apply the target's symbol mangling to the result of `tcx.symbol_name()` depending on the target. This is done by calling the LLVM name mangler, which handles all of the details.
- On Mach-O, all symbols have a leading underscore.
- On x86 Windows, different mangling is used for cdecl, stdcall, fastcall and vectorcall.
- No mangling is needed on other platforms.
r? `@nagisa`
cc `@eddyb`
only downgrade selection Error -> Ambiguous if type error is in predicate
That is, we don't care if there's a TypeError type in the ParamEnv.
Fixes#95408
remove find_use_placement
A more robust solution to finding where to place use suggestions was added in #94584.
The algorithm uses the AST to find the span for the suggestion so we pass this span
down to the HIR during lowering and use it instead of calling `find_use_placement`
Fixes#94941
Check var scope if it exist
Fixes#92893.
Added helper function to check the scope of a variable, if it doesn't have a scope call delay_span_bug, which avoids us trying to get a block/scope that doesn't exist.
Had to increase `ROOT_ENTRY_LIMIT` was getting tidy error
Create (unstable) 2024 edition
[On Zulip](https://rust-lang.zulipchat.com/#narrow/stream/213817-t-lang/topic/Deprecating.20macro.20scoping.20shenanigans/near/272860652), there was a small aside regarding creating the 2024 edition now as opposed to later. There was a reasonable amount of support and no stated opposition.
This change creates the 2024 edition in the compiler and creates a prelude for the 2024 edition. There is no current difference between the 2021 and 2024 editions. Cargo and other tools will need to be updated separately, as it's not in the same repository. This change permits the vast majority of work towards the next edition to proceed _now_ instead of waiting until 2024.
For sanity purposes, I've merged the "hello" UI tests into a single file with multiple revisions. Otherwise we'd end up with a file per edition, despite them being essentially identical.
````@rustbot```` label +T-lang +S-waiting-on-review
Not sure on the relevant team, to be honest.
Stabilize `derive_default_enum`
This stabilizes `#![feature(derive_default_enum)]`, as proposed in [RFC 3107](https://github.com/rust-lang/rfcs/pull/3107) and tracked in #87517. In short, it permits you to `#[derive(Default)]` on `enum`s, indicating what the default should be by placing a `#[default]` attribute on the desired variant (which must be a unit variant in the interest of forward compatibility).
```````@rustbot``````` label +S-waiting-on-review +T-lang
`--extern-location` was an experiment to investigate the best way to
generate useful diagnostics for unused dependency warnings by enabling a
build system to identify the corresponding build config.
While I did successfully use this, I've since been convinced the
alternative `--json unused-externs` mechanism is the way to go, and
there's no point in having two mechanisms with basically the same
functionality.
This effectively reverts https://github.com/rust-lang/rust/pull/72603
when checking pointee metadata, canonicalize the `Sized` check
Use `infcx.predicate_must_hold_modulo_regions` with a `Sized` obligation instead of just calling `ty.is_sized`, because the latter does not canonicalize region and type vars (and in the test case I added in this PR, there's a region var in the `ParamEnv`).
Fixes#95311
Remove `<mbe::TokenTree as Clone>`
`mbe::TokenTree` doesn't really need to implement `Clone`, and getting rid of that impl leads to some speed-ups.
r? `@petrochenkov`
errors: lazily load fallback fluent bundle
Addresses (hopefully) https://github.com/rust-lang/rust/pull/95667#issuecomment-1094794087.
Loading the fallback bundle in compilation sessions that won't go on to emit any errors unnecessarily degrades compile time performance, so lazily create the Fluent bundle when it is first required.
r? `@ghost` (just for perf initially)
fix: wrong trait import suggestion for T:
The suggestion to bound `T` had an extra `:`.
```rust
fn foo<T:>(t: T) {
t.clone();
}
```
```
error[E0599]: no method named `clone` found for type parameter `T` in the current scope
--> src/lib.rs:2:7
|
2 | t.clone();
| ^^^^^ method not found in `T`
|
= help: items from traits can only be used if the type parameter is bounded by the trait
help: the following trait defines an item `clone`, perhaps you need to restrict type parameter `T` with it:
|
1 | fn foo<T: Clone:>(t: T) {
| ~~~~~~~~
```
Fixes: #95898
Use mir constant in thir instead of ty::Const
This is blocked on https://github.com/rust-lang/rust/pull/94059 (does include its changes, the first two commits in this PR correspond to those changes) and https://github.com/rust-lang/rust/pull/93800 being reinstated (which had to be reverted). Mainly opening since `@lcnr` offered to give some feedback and maybe also for a perf-run (if necessary).
This currently contains a lot of duplication since some of the logic of `ty::Const` had to be copied to `mir::ConstantKind`, but with the introduction of valtrees a lot of that functionality will disappear from `ty::Const`.
Only the last commit contains changes that need to be reviewed here. Did leave some `FIXME` comments regarding future implementation decisions and some things that might be incorrectly implemented.
r? `@oli-obk`
Loading the fallback bundle in compilation sessions that won't go on to
emit any errors unnecessarily degrades compile time performance, so
lazily create the Fluent bundle when it is first required.
Signed-off-by: David Wood <david.wood@huawei.com>
prevent opaque types from appearing in impl headers
cc `@lqd`
opaque types are not distinguishable from their hidden type at the codegen stage. So we could either end up with cases where the hidden type doesn't implement the trait (which will thus ICE) or where the hidden type does implement the trait (so we'd be using its impl instead of the one written for the opaque type). This can even lead to unsound behaviour without unsafe code.
Fixes https://github.com/rust-lang/rust/issues/86411.
Fixes https://github.com/rust-lang/rust/issues/84660.
rebase of #87382 plus some diagnostic tweaks
Fix suggestions in case of `T:` bounds
This PR fixes a corner case in `suggest_constraining_type_params` that was causing incorrect suggestions.
For the following functions:
```rust
fn a<T:>(t: T) { [t, t]; }
fn b<T>(t: T) where T: { [t, t]; }
```
We previously suggested the following:
```text
...
help: consider restricting type parameter `T`
|
1 | fn a<T: Copy:>(t: T) { [t, t]; }
| ++++++
...
help: consider further restricting this bound
|
2 | fn b<T>(t: T) where T: + Copy { [t, t]; }
| ++++++
```
Note that neither `T: Copy:` not `where T: + Copy` is a correct bound.
With this commit the suggestions are correct:
```text
...
help: consider restricting type parameter `T`
|
1 | fn a<T: Copy>(t: T) { [t, t]; }
| ++++
...
help: consider further restricting this bound
|
2 | fn b<T>(t: T) where T: Copy { [t, t]; }
| ++++
```
r? `@compiler-errors`
I've tried fixing #95898 here too, but got too confused with how `suggest_traits_to_import` works and what it does 😅
Move name resolution logic to a dedicated file
The code resolution logic from an Ident is scattered between several files.
The first commits creates `rustc_resolve::probe` module to hold the different mutually recursive functions together. Just a move, no code change.
The following commits attempt to make the logic a bit more readable.
The two fields `last_import_segment` and `unusable_binding` are replaced by function parameters.
In order to manage the fallout, `maybe_` variants of the function are added, dedicated to speculative resolution.
r? `@petrochenkov`
Rollup of 4 pull requests
Successful merges:
- #95783 (rustdoc doctest: include signal number in exit status)
- #95794 (`parse_tt`: a few more tweaks)
- #95963 ([bootstrap] Grab the right FileCheck binary for dist when cross-compiling.)
- #95975 (Don't test -Cdefault-linker-libraries=yes when cross compiling.)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Skip `Lazy` for some metadata tables
Some metadata tables encode their entries indirectly, through the Lazy construct. This is useful when dealing with variable length encoding, but incurs the extra cost of one u32.
Meanwhile, some fields can be encoded in a single u8, or can use a short fixed-length encoding. This PR proposes to do so, and avoid the overhead.
`impl const Default for Box<[T]>` and `Box<str>`
The unstable `const_default_impls` (#87864) already include empty `Vec<T>` and `String`. Now we extend that concept to `Box<[T]>` and `Box<str>` as well.
This obviates a hack in `rustc_ast`'s `P::<[T]>::new`.
Instead of checking only the user provided sysroot or the default (when
no sysroot is provided), search user provided sysroot and then check
default sysroots for locale requested by the user.
Signed-off-by: David Wood <david.wood@huawei.com>
use `Span::find_ancestor_inside` to get right span in CastCheck
This is a quick fix. This bad suggestion likely lives in other places... but thought it would be useful to fix all of the CastCheck ones first.
Let me know if reviewer would prefer I add more tests for each of the diagnostics in CastCheck, or would like to do a more thorough review of other suggestions that use spans in typeck. I would also be open to further suggestions on how to better expose an API that gives us the "best" span for a diagnostic suggestion.
Fixed#95919
Fix crate_type attribute to not warn on duplicates
In #88681 I accidentally marked the `crate_type` attribute as only allowing a single attribute. However, multiple attributes are allowed (they are joined together [here](027a232755/compiler/rustc_interface/src/util.rs (L530-L542))). This fixes it to not report a warning if duplicates are found.
Closes#95902
Document the current MIR semantics that are clear from existing code
This PR adds documentation to places, operands, rvalues, statementkinds, and terminatorkinds that describes their existing semantics and requirements. In many places the semantics depend on the Rust memory model or other T-Lang decisions - when this is the case, it is just noted as such with links to UCG issues where possible. I'm hopeful that none of the documentation added here can be used to justify optimizations that depend on the memory model. The documentation for places and operands probably comes closest to running afoul of this - if people think that it cannot be merged as is, it can definitely also be taken out.
The goal here is to only document parts of MIR that seem to be decided already, or are at least depended on by existing code. That leaves quite a number of open questions - those are marked as "needs clarification." I'm not sure what to do with those in this PR - we obviously can't decide all these questions here. Should I just leave them in as is? Take them out? Keep them in but as `//` instead of `///` comments?
If this is too big to review at once, I can split this up.
r? rust-lang/mir-opt
Respect -Z verify-llvm-ir and other flags that add extra passes when combined with -C no-prepopulate-passes in the new LLVM Pass Manager.
As part of the switch to the new LLVM Pass Manager the behaviour of flags such as `-Z verify-llvm-ir` (e.g. sanitizer, instrumentation) was modified when combined with `-C no-prepopulate-passes`. With the old PM, rustc was the one manually constructing the pipeline and respected those flags but in the new pass manager, those flags are used to build a list of callbacks that get invoked at certain extension points in the pipeline. Unfortunately, `-C no-prepopulate-passes` would skip building the pipeline altogether meaning we'd never add the corresponding passes. The fix here is to just manually invoke those callbacks as needed.
Fixes#95874
Demonstrating the current vs fixed behaviour using the bug in #95864
```console
$ rustc +nightly asm-miscompile.rs --edition 2021 --emit=llvm-ir -C no-prepopulate-passes -Z verify-llvm-ir
$ echo $?
0
$ rustc +stage1 asm-miscompile.rs --edition 2021 --emit=llvm-ir -C no-prepopulate-passes -Z verify-llvm-ir
Basic Block in function '_ZN14asm_miscompile3foo28_$u7b$$u7b$closure$u7d$$u7d$17h360e2f7eee1275c5E' does not have terminator!
label %bb1
LLVM ERROR: Broken module found, compilation aborted!
```
