Adds `#[rustc_force_inline]` which is similar to always inlining but
reports an error if the inlining was not possible, and which always
attempts to inline annotated items, regardless of optimisation levels.
It can only be applied to free functions to guarantee that the MIR
inliner will be able to resolve calls.
Subtree sync for rustc_codegen_cranelift
This has a couple of changes that will conflict with https://github.com/rust-lang/rust/pull/134338.
r? `@ghost`
`@rustbot` label +A-codegen +A-cranelift +T-compiler
Rollup of 7 pull requests
Successful merges:
- #132607 (Used pthread name functions returning result for FreeBSD and DragonFly)
- #134693 (proc_macro: Use `ToTokens` trait in `quote` macro)
- #134732 (Unify conditional-const error reporting with non-const error reporting)
- #135083 (Do not ICE when encountering predicates from other items in method error reporting)
- #135251 (Only treat plain literal patterns as short)
- #135320 (Fix typo in `#[coroutine]` gating error)
- #135321 (remove more redundant into() conversions)
r? `@ghost`
`@rustbot` modify labels: rollup
Do not ICE when encountering predicates from other items in method error reporting
See the comments I left in the code and the test file.
Fixes https://github.com/rust-lang/rust/issues/124350
Unify conditional-const error reporting with non-const error reporting
This PR unifies the error reporting between `ConditionallyConstCall` and `FnCallNonConst` so that the former will refer to syntactical sugar like operators by their sugared name, rather than calling all operators "methods". We achieve this by making the "non-const" part of the error message generic over the "non" part so we can plug in "conditionally" instead.
This should ensure that as we constify traits in the standard library, we don't regress error messages for things like `==`.
r? fmease or reassign
Remove special-casing for argument patterns in MIR typeck (attempt to fix perf regression of #133858)
See [my comment](https://github.com/rust-lang/rust/pull/133858#issuecomment-2579029618) on #133858 for more information. This is just a guess as to what went wrong, and I haven't been able to get the profiler running locally, so I'll need a perf run to make sure this actually helps.
There's one test's stderr that suffers a bit, but this was just papering over the issue anyway. Making region errors point to the correct constraints in the presence of invariance/contravariance is a broader problem; the current way it's handled is mostly based on guesswork, luck, and hoping it works out. Properly handling that (somehow) would improve the test's stderr without the hack that this PR reverts.
Make sure to walk into nested const blocks in `RegionResolutionVisitor`
Fixes https://github.com/rust-lang/rust/issues/135306
I tried auditing the rest of the visitors that called `.visit_body`, and it seems like this is the only one that was missing it. I wonder if we should modify intravisit (specifcially, that `NestedBodyFilter` stuff) to make this less likely to happen, tho...
r? oli-obk
`-Zrandomize-layout` harder. `Foo<T> != Foo<U>`
Tracking issue: #106764
Previously randomize-layout only used a deterministic shuffle based on the seed stored in an Adt's ReprOptions, meaning that `Foo<T>` and `Foo<U>` were shuffled by the same seed. This change adds a similar seed to each calculated LayoutData so that a struct can be randomized both based on the layout of its fields and its per-type seed.
Primitives start with simple seed derived from some of their properties. Though some types can no longer be distinguished at that point, e.g. usize and u64 will still be treated the same.
previously field ordering was using the same seed for all instances of Foo,
now we pass seed values through the layout tree so that not only
the struct itself affects layout but also its fields
Creating a "trimmed DefID path" when no error is being emitted is an ICE (on purpose). If we create a trimmed path for a lint that is then silenced before being emitted causes a known ICE. This side-steps the issue by always using `with_no_trimmed_path!`.
This was verified to fix https://github.com/quinn-rs/quinn/, but couldn't write a repro case for the test suite.
Fix#135289.
Rollup of 3 pull requests
Successful merges:
- #134898 (Make it easier to run CI jobs locally)
- #135195 (Make `lit_to_mir_constant` and `lit_to_const` infallible)
- #135261 (Account for identity substituted items in symbol mangling)
r? `@ghost`
`@rustbot` modify labels: rollup
Make `lit_to_mir_constant` and `lit_to_const` infallible
My motivation for this change is just that it's annoying to check everywhere, especially since all but one call site was just ICEing on errors anyway right there.
They can still fail, but now just return an error constant instead of having the caller handle the error.
fixes#114317fixes#126182
Rollup of 5 pull requests
Successful merges:
- #135212 (Remove outdated information in the `unreachable_pub` lint description)
- #135225 (Explicitly build proc macro test with panic=unwind)
- #135242 (add missing provenance APIs on NonNull)
- #135247 (Add a list of symbols for stable standard library crates)
- #135269 (Remove some unnecessary `.into()` calls)
r? `@ghost`
`@rustbot` modify labels: rollup
`Ty::new` wasn't used anywhere outside this module
`Ty::new_adt` shouldn't ever be used for anything but adts. This hasn't caught any bugs, but seems good to check anyway
Add a list of symbols for stable standard library crates
There are a few locations where the crate name is checked against an enumerated list of `std`, `core`, `alloc`, and `proc_macro`, or some subset thereof. In most cases when we are looking for any "standard library" crate, all four crates should be treated the same. Change this so the crates are listed in one place, and that list is used wherever a list of `std` crates is needed.
`test` could be considered relevant in some of these cases, but generally treating it separate from the others seems preferable while it is unstable.
There are also a few places that Clippy will be able to use this.
Explicitly build proc macro test with panic=unwind
Fuchsia explicitly builds rust and all rust targets with `-C panic=abort` to minimize code generation size. However, when compiling a proc-macro with this setting it can cause a warning to be emitted, which breaks `tests/ui/invalid-compile-flags/crate-type-flag.rs`. This hasn't been a problem in the past for us since we compile our proc macros on host, rather than inside Fuchsia.
This attempts to fix the issue by explicitly requiring that we're using the unwinder when compiling this test to avoid the warning being emitted.
Fixes#135223
Remove outdated information in the `unreachable_pub` lint description
As far as I understand the `unreachable_pub` lint hasn't had false-positives since it started using "effective visibilities". Let's remove that warning from the lint description.
r? `@petrochenkov`
[mir-opt] GVN some more transmute cases
We already did `Transmute`-then-`PtrToPtr`; this adds the nearly-identical `PtrToPtr`-then-`Transmute`.
It also adds `transmute(Foo(x))` → `transmute(x)`, when `Foo` is a single-field transparent type. That's useful for things like `NonNull { pointer: p }.as_ptr()`. It also detects when a `Transmute` is just an identity-for-the-value `PtrCast` between different raw pointer types, to help such things fold with other GVN passes.
Found these as I was looking at <https://github.com/rust-lang/compiler-team/issues/807>-related changes. This also removes the questionably-useful "turn a transmute into a field projection" part of instsimplify (which I added ages ago without an obvious need for it) since that would just put back the field projections that MCP807 is trying to ban.
r? mir-opt
Implement `const Destruct` in old solver
Self-explanatory. Not totally settled that this is the best structure for built-in trait impls for effect goals in the new solver, but it's almost certainly the simplest.
r? lcnr or re-roll
Add new `{x86_64,i686}-win7-windows-gnu` targets
These are in symmetry with `{x86_64,i686}-win7-windows-msvc`.
> ## Tier 3 target policy
>
> At this tier, the Rust project provides no official support for a target, so we
> place minimal requirements on the introduction of targets.
>
> A proposed new tier 3 target must be reviewed and approved by a member of the
> compiler team based on these requirements. The reviewer may choose to gauge
> broader compiler team consensus via a [Major Change Proposal (MCP)][https://forge.rust-lang.org/compiler/mcp.html].
>
> A proposed target or target-specific patch that substantially changes code
> shared with other targets (not just target-specific code) must be reviewed and
> approved by the appropriate team for that shared code before acceptance.
>
> - A tier 3 target must have a designated developer or developers (the "target
> maintainers") on record to be CCed when issues arise regarding the target.
> (The mechanism to track and CC such developers may evolve over time.)
This is me, `@tbu-` on github.
> - Targets must use naming consistent with any existing targets; for instance, a
> target for the same CPU or OS as an existing Rust target should use the same
> name for that CPU or OS. Targets should normally use the same names and
> naming conventions as used elsewhere in the broader ecosystem beyond Rust
> (such as in other toolchains), unless they have a very good reason to
> diverge. Changing the name of a target can be highly disruptive, especially
> once the target reaches a higher tier, so getting the name right is important
> even for a tier 3 target.
> - Target names should not introduce undue confusion or ambiguity unless
> absolutely necessary to maintain ecosystem compatibility. For example, if
> the name of the target makes people extremely likely to form incorrect
> beliefs about what it targets, the name should be changed or augmented to
> disambiguate it.
> - If possible, use only letters, numbers, dashes and underscores for the name.
> Periods (`.`) are known to cause issues in Cargo.
Consistent with `{x86_64,i686}-win7-windows-msvc`, see also #118150.
> - Tier 3 targets may have unusual requirements to build or use, but must not
> create legal issues or impose onerous legal terms for the Rust project or for
> Rust developers or users.
> - The target must not introduce license incompatibilities.
> - Anything added to the Rust repository must be under the standard Rust
> license (`MIT OR Apache-2.0`).
> - The target must not cause the Rust tools or libraries built for any other
> host (even when supporting cross-compilation to the target) to depend
> on any new dependency less permissive than the Rust licensing policy. This
> applies whether the dependency is a Rust crate that would require adding
> new license exceptions (as specified by the `tidy` tool in the
> rust-lang/rust repository), or whether the dependency is a native library
> or binary. In other words, the introduction of the target must not cause a
> user installing or running a version of Rust or the Rust tools to be
> subject to any new license requirements.
> - Compiling, linking, and emitting functional binaries, libraries, or other
> code for the target (whether hosted on the target itself or cross-compiling
> from another target) must not depend on proprietary (non-FOSS) libraries.
> Host tools built for the target itself may depend on the ordinary runtime
> libraries supplied by the platform and commonly used by other applications
> built for the target, but those libraries must not be required for code
> generation for the target; cross-compilation to the target must not require
> such libraries at all. For instance, `rustc` built for the target may
> depend on a common proprietary C runtime library or console output library,
> but must not depend on a proprietary code generation library or code
> optimization library. Rust's license permits such combinations, but the
> Rust project has no interest in maintaining such combinations within the
> scope of Rust itself, even at tier 3.
> - "onerous" here is an intentionally subjective term. At a minimum, "onerous"
> legal/licensing terms include but are *not* limited to: non-disclosure
> requirements, non-compete requirements, contributor license agreements
> (CLAs) or equivalent, "non-commercial"/"research-only"/etc terms,
> requirements conditional on the employer or employment of any particular
> Rust developers, revocable terms, any requirements that create liability
> for the Rust project or its developers or users, or any requirements that
> adversely affect the livelihood or prospects of the Rust project or its
> developers or users.
AFAICT, it's the same legal situation as the tier 1 `{x86_64,i686}-pc-windows-gnu`.
> - Neither this policy nor any decisions made regarding targets shall create any
> binding agreement or estoppel by any party. If any member of an approving
> Rust team serves as one of the maintainers of a target, or has any legal or
> employment requirement (explicit or implicit) that might affect their
> decisions regarding a target, they must recuse themselves from any approval
> decisions regarding the target's tier status, though they may otherwise
> participate in discussions.
> - This requirement does not prevent part or all of this policy from being
> cited in an explicit contract or work agreement (e.g. to implement or
> maintain support for a target). This requirement exists to ensure that a
> developer or team responsible for reviewing and approving a target does not
> face any legal threats or obligations that would prevent them from freely
> exercising their judgment in such approval, even if such judgment involves
> subjective matters or goes beyond the letter of these requirements.
Understood.
> - Tier 3 targets should attempt to implement as much of the standard libraries
> as possible and appropriate (`core` for most targets, `alloc` for targets
> that can support dynamic memory allocation, `std` for targets with an
> operating system or equivalent layer of system-provided functionality), but
> may leave some code unimplemented (either unavailable or stubbed out as
> appropriate), whether because the target makes it impossible to implement or
> challenging to implement. The authors of pull requests are not obligated to
> avoid calling any portions of the standard library on the basis of a tier 3
> target not implementing those portions.
This target supports the whole libstd surface, since it's essentially reusing all of the x86_64-pc-windows-gnu target. Understood.
> - The target must provide documentation for the Rust community explaining how
> to build for the target, using cross-compilation if possible. If the target
> supports running binaries, or running tests (even if they do not pass), the
> documentation must explain how to run such binaries or tests for the target,
> using emulation if possible or dedicated hardware if necessary.
I tried to write some documentation on that.
> - Tier 3 targets must not impose burden on the authors of pull requests, or
> other developers in the community, to maintain the target. In particular,
> do not post comments (automated or manual) on a PR that derail or suggest a
> block on the PR based on a tier 3 target. Do not send automated messages or
> notifications (via any medium, including via ``@`)` to a PR author or others
> involved with a PR regarding a tier 3 target, unless they have opted into
> such messages.
> - Backlinks such as those generated by the issue/PR tracker when linking to
> an issue or PR are not considered a violation of this policy, within
> reason. However, such messages (even on a separate repository) must not
> generate notifications to anyone involved with a PR who has not requested
> such notifications.
Understood.
> - Patches adding or updating tier 3 targets must not break any existing tier 2
> or tier 1 target, and must not knowingly break another tier 3 target without
> approval of either the compiler team or the maintainers of the other tier 3
> target.
> - In particular, this may come up when working on closely related targets,
> such as variations of the same architecture with different features. Avoid
> introducing unconditional uses of features that another variation of the
> target may not have; use conditional compilation or runtime detection, as
> appropriate, to let each target run code supported by that target.
> - Tier 3 targets must be able to produce assembly using at least one of
> rustc's supported backends from any host target. (Having support in a fork
> of the backend is not sufficient, it must be upstream.)
Understood.
> If a tier 3 target stops meeting these requirements, or the target maintainers
> no longer have interest or time, or the target shows no signs of activity and
> has not built for some time, or removing the target would improve the quality
> of the Rust codebase, we may post a PR to remove it; any such PR will be CCed
> to the target maintainers (and potentially other people who have previously
> worked on the target), to check potential interest in improving the situation.
>
Understood.
r? compiler-team
Suggest Replacing Comma with Semicolon in Incorrect Repeat Expressions
Fixes#80173
This PR detects typos in repeat expressions like `["_", 10]` and `vec![String::new(), 10]` and suggests replacing comma with semicolon.
Also, improves code in other place by adding doc comments and making use of a helper function to check if a type implements `Clone`.
References:
1. For `vec![T; N]`: https://doc.rust-lang.org/std/macro.vec.html
2. For `[T; N]`: https://doc.rust-lang.org/std/primitive.array.html
We already did `Transmute`-then-`PtrToPtr`; this adds the nearly-identical `PtrToPtr`-then-`Transmute`.
It also adds `transmute(Foo(x))` → `transmute(x)`, when `Foo` is a single-field transparent type. That's useful for things like `NonNull { pointer: p }.as_ptr()`.
Found these as I was looking at MCP807-related changes.
There are a few locations where the crate name is checked against an
enumerated list of `std`, `core`, `alloc`, and `proc_macro`, or some
subset thereof. In most of these cases, all four crates should likely be
treated the same. Change this so the crates are listed in one place, and
that list is used wherever a list of `std` crates is needed.
`test` could be considered relevant in some of these cases, but
generally treating it separate from the others seems preferable while it
is unstable.
There are also a few places that Clippy will be able to use this.
Fuchsia explicitly builds rust and all rust targets with `-C
panic=abort` to minimize code generation size. However, when compiling a
proc-macro with this setting it can cause a warning to be emitted, which
breaks `tests/ui/invalid-compile-flags/crate-type-flag.rs`. This hasn't
been a problem in the past for us since we compile our proc macros on
host, rather than inside Fuchsia.
This attempts to fix the issue by explicitly requiring that we're using
the unwinder when compiling this test to avoid the warning being
emitted.
Fixes#135223
A couple simple borrowck cleanups
This PR has a couple simple renamings:
- it's been a long time since the mapping from `Location`s to `PointIndex`es was extracted from `RegionElements` into the `DenseLocationMap`, but only the types were renamed at the time. borrowck still refers to this map as `elements`. That's confusing, especially since sometimes we also use the mapping via `LivenessValues`, and makes more sense as `location_map` instead.
- to clarify `LocationTable` is not as general as it sounds, and is only for datalog polonius. In this branch I didn't rename the handful of `location_table` fields and params to `polonius_table`, but can do that to differentiate it even more from `location_map`. I did try it locally and it looks worthwhile, so if you'd prefer I can also push it here. (Or we could even switch these datalog types and fields to even more explicit names)
- to clarify the incomprehensible `AllFacts`, it is renamed to `PoloniusFacts`. These can be referred to as `facts` within the legacy polonius module, but as `polonius_facts` outside of it to make it clear that they're not about NLLs (nor are they about in-tree polonius but that'll be magically fixed when they're removed in the future)
r? `@matthewjasper`
Exhaustively handle expressions in patterns
We currently have this invariant in HIR that a `PatKind::Lit` or a `PatKind::Range` only contains
* `ExprKind::Lit`
* `ExprKind::UnOp(Neg, ExprKind::Lit)`
* `ExprKind::Path`
* `ExprKind::ConstBlock`
So I made `PatKind::Lit` and `PatKind::Range` stop containing `Expr`, and instead created a `PatLit` type whose `kind` enum only contains those variants.
The only place code got more complicated was in clippy, as it couldn't share as much anymore with `Expr` handling
It may be interesting on merging `ExprKind::{Path,Lit,ConstBlock}` in the future and using the same `PatLit` type (under a new name).
Then it should also be easier to eliminate any and all `UnOp(Neg, Lit) | Lit` matching that we have across the compiler. Some day we should fold the negation into the literal itself and just store it on the numeric literals
Its original naming hides the fact that it's related to datalog
polonius, and bound to be deleted in the near future.
It also conflicts with the expected name for the actual NLL location
map, and prefixing it with its use will make the differentiation
possible.
`best_blame_constraint`: Blame better constraints when the region graph has cycles from invariance or `'static`
This fixes#132749 by changing which constraint is blamed for region errors in several cases. `best_blame_constraint` had a heuristic that tried to pinpoint the constraint causing an error by filtering out any constraints where the outliving region is unified with the ultimate target region being outlived. However, it used the SCCs of the region graph to do this, which is unreliable; in particular, if the target region is `'static`, or if there are cycles from the presence of invariant types, it was skipping over the constraints it should be blaming. As is the case in that issue, this could lead to confusing diagnostics. The simplest fix seems to work decently, judging by test stderr: this makes `best_blame_constraint` no longer filter constraints by their outliving region's SCC.
There are admittedly some quirks in the test output. In many cases, subdiagnostics that depend on the particular constraint being blamed have either started or stopped being emitted. After starting at this for quite a while, I think anything too fickle about whether it outputs based on the particular constraint being blamed should instead be looking at the constraint path as a whole, similar to what's done for [the placeholder-from-predicate note](https://github.com/rust-lang/rust/compare/master...dianne:rust:better-blame-constraints-for-static#diff-3c0de6462469af483c9ecdf2c4b00cb26192218ef2d5c62a0fde75107a74caaeR506).
Very many tests involving invariant types gained a note pointing out the types' invariance, but in a few cases it was lost. A particularly illustrative example is [tests/ui/lifetimes/copy_modulo_regions.stderr](https://github.com/rust-lang/rust/compare/master...dianne:rust:better-blame-constraints-for-static?expand=1#diff-96e1f8b29789b3c4ce2f77a5e0fba248829b97ef9d1ce39e7d2b4aa57b2cf4f0); I'd argue the new constraint is a better one to blame, but it lacks the variance diagnostic information that's elsewhere in the constraint path. If desired, I can try making that note check the whole path rather than just the blamed constraint.
The subdiagnostic [`BorrowExplanation::add_object_lifetime_default_note`](https://doc.rust-lang.org/nightly/nightly-rustc/rustc_borrowck/diagnostics/explain_borrow/enum.BorrowExplanation.html#method.add_object_lifetime_default_note) depends on a `Cast` being blamed, so [a special case](364ca7f99c) was necessary to keep it from disappearing from tests specifically testing for it. However, see the FIXME comment in that commit; I think the special case should be removed once that subdiagnostic works properly, but it's nontrivial enough to warrant a separate PR. Incidentally, this removes the note from a test where it was being added erroneously: in [tests/ui/borrowck/two-phase-surprise-no-conflict.stderr](https://github.com/rust-lang/rust/compare/master...dianne:rust:better-blame-constraints-for-static?expand=1#diff-8cf085af8203677de6575a45458c9e6b03412a927df879412adec7e4f7ff5e14), the object lifetime is explicitly provided and it's not `'static`.
"Elements" are `RegionElement`s. The dense location mapping was removed
from the element containers a while ago but didn't rename its use-sites.
Most of the old naming only used the mapping, and are better named
`location_map`.
The goal of this cleanup is to make it more apparent which feature gates
correspond to which typing rules, and which typing rules correspond to
what code. My intent is for calls to the "which typing rules do we
have?" functions to be replaced by comments (and edition checks, as
appropriate), but as long as we're experimenting with multiple rulesets,
this seemed to me to be the easiest to document and read. There's still
some nontrivial control flow, but I've added comments to try and make it
clearer.
There's some logic that looks like it could be de-duplicated across
different ways of matching against inherited references; however, the
duplication is intentional. Once we choose which rulesets we want, we
can make this more clever, but until then, my priorities are clarity and
ease of modification/extension. That said, I think the diagnostics could
use some work; factoring out commonalities there (and separating them
from the typing logic) would be ideal. I've opted not to include that
here both since it'd make this refactor less obvious and since it
affects test output.
Also, this doesn't get quite as fine-grained as Typing Rust Patterns, so
there's some instances where certain rules are conflated. I'd prefer to
minimize dead/untested codepaths for rulesets we're not interested in,
so as a compromise I've added comments wherever some aspect of the
typing rules is assumed from another. I'm not totally happy with it, but
I think it's at least better than plain checks against the feature gates
and edition.
If Rules 3 or 5 are adopted in any edition, we'll need to track the
`MutblCap` in all editions for macro hygiene purposes. Previously, the
check for whether to track it was conflated with the checks for whether
to apply Rules 3 and 5, so to make it a bit clearer, this always tracks
the `MutblCap`. If needed, we could check if Rules 3 or 5 are present in
any edition before tracking the `MutblCap`, but since it's not that much
more expensive to always track it, I've figured that's simplest.
My main concern with removing the checks is that it may not be clear
that the `MutblCap` is tracked for those specific purposes. To try and
mitigate this, I've made its doc comment a bit more precise regarding
the extent of how and why it's used.
This leaves the condition untouched on the `cap_to_weakly_not` call
needed for Rule 5, since it's only needed for that and it can affect
diagnostics.
As far as I can tell, the assignment removed here will never do
anything. `pat_info.max_ref_mutbl` starts at `MutblCap::Mut` for the
top-level pattern and is only changed if feature gates are enabled that
would result in the statement not being executed. Regardless of what new
pattern typing rules are adopted, I don't imagine we want to
conditionally reset `max_ref_mutbl` based on edition either, since it'd
have consequences for subpatterns in other editions.
This aims to reduce the complexity needed in the boolean logic for telling which
rules we're using to type patterns. If we still want the functionality this
removes, we can re-add it later, after some cleanup to pattern typing.
arm: add unstable soft-float target feature
This has an actual usecase as mentioned [here](https://github.com/rust-lang/rust/issues/116344#issuecomment-2575324988), and with my recent ARM float ABI changes there shouldn't be any soundness concerns any more. We will reject enabling this feature on `hf` targets, but disabling it on non-`hf` targets is entirely fine -- the target feature refers to whether softfloat emulation is used for float instructions, and is independent of the ABI which we set separately via `llvm_floatabi`.
Cc ``@workingjubilee``
Convert typeck constraints in location-sensitive polonius
In this PR, we do a big chunk of the work of localizing regular outlives constraints.
The slightly annoying thing is handling effectful statements: usually the subset graph propagates loans at a single point between regions, and liveness propagates loans between points within a single region, but some statements have effects applied on exit.
This was also a problem before, in datalog polonius terms and Niko's solution at the time, this is about: the mid-point. The idea was to duplicate all MIR locations into two physical points, and orchestrate the effects with that. Somewhat easier to do, but double the CFG.
We've always believed we didn't _need_ midpoints in principle, as we can represent changes on exit as on happening entry to the successor, but there's some difficulty in tracking the position information at sufficient granularity through outlives relation (especially since we also have bidirectional edges and time-traveling now).
Now, that is surely what we should be doing in the future. In the mean time, I infer this from the kind of statement/terminator where an outlives constraint arose. It's not particularly complicated but some explanation will help clarify the code.
Assignments (in their various forms) are the quintessential example of these crossover cases: loans that would flow into the LHS would not be visible on entry to the point but on exit -- so we'll localize these edges to the successor. Let's look at a real-world example, involving invariance for bidirectional edges:
```rust
let mut _1: HashMap<i32, &'7 i32>;
let mut _3: &'9 mut HashMap<i32, &'10 i32>;
...
/* at bb1[3]: */ _3 = &'3 mut _1;
```
Here, typeck expectedly produces 3 outlives constraints today:
1. `'3 -> '9`
2. `'7 -> '10`
3. `'10 -> '7`
And we localize them like so,
1. `'3 -> '9` flows into the LHS and becomes: `3_bb1_3 -> 9_bb1_4`
2. `'7 -> '10` flows into the LHS and becomes: `7_bb1_3 -> 10_bb1_4`
3. `'10 -> '7` flows from the LHS and becomes: `10_bb1_4 -> 7_bb1_3` (time traveling 👌)
---
r? ``@jackh726``
To keep you entertained during the holidays I also threw in a couple of small changes removing cruft in the borrow checker.
We're actually getting there. The next PR will be the last one needed to get end-to-end tests working.
Use a post-monomorphization typing env when mangling components that come from impls
When mangling associated methods of impls, we were previously using the wrong param-env. Instead of using a fully monomorphized param-env like we usually do in codegen, we were taking the post-analysis param-env, and treating it as an early binder to *re-substitute* the impl args. I've pointed out the problematic old code in an inline comment.
This would give us param-envs with possibly trivial predicates that would prevent normalization via param-env shadowing.
In the example test linked below, `tests/ui/symbol-names/normalize-in-param-env.rs`, this happens when we mangle the impl `impl<P: Point2> MyFrom<P::S> for P` with the substitution `P = Vec2`. Because the where clause of the impl is `P: Point2`, which elaborates to `[P: Point2, P: Point, <P as Point>::S projects-to <P as Point2>::S2]` and the fact that `impl Point2 for Vec2` normalizes `Vec2::S2` to `Vec2::S`, this causes a cycle.
The proper fix here is to use a fully monomorphized param-env for the case where the impl is properly substituted.
Fixes#135143
While #134081 uncovered this bug for legacy symbol mangling, it was preexisting for v0 symbol mangling. This PR fixes both. The test requires a "hack" because we strip the args of the instance we're printing for legacy symbol mangling except for drop glue, so we box a closure to ensure we generate drop glue.
r? oli-obk
Normalize each signature input/output in `typeck_with_fallback` with its own span
Applies the same hack as #106582 but to the args in typeck. Greatly improves normalization error spans from a signature.
remove unnecessary `eval_verify_bound`
This does not impact any tests. I feel like any cases where this could useful should instead be fixed by a general improvement to `eval_verify_bound` to avoid having to promote this `TypeTest` in the first place 🤔
r? types cc ``@nikomatsakis``
Avoid naming variables `str`
This renames variables named `str` to other names, to make sure `str`
always refers to a type.
It's confusing to read code where `str` (or another standard type name)
is used as an identifier. It also produces misleading syntax
highlighting.
llvm: Ignore error value that is always false
See llvm/llvm-project#121851
For LLVM 20+, this function (`renameModuleForThinLTO`) has no return value. For prior versions of LLVM, this never failed, but had a signature which allowed an error value people were handling.
`@rustbot` label: +llvm-main
r? `@nikic`
Wait a moment before approving while the llvm-main infrastructure picks it up.
This renames variables named `str` to other names, to make sure `str`
always refers to a type.
It's confusing to read code where `str` (or another standard type name)
is used as an identifier. It also produces misleading syntax
highlighting.
Avoid replacing the definition of `CURRENT_RUSTC_VERSION`
Before this PR, replace-version-placeholder hardcoded the path defining CURRENT_RUSTC_VERSION (to avoid replacing it). After a refactor moved the file defining it without changing the hardcoded path, the tool started replacing the constant itself with the version number.
To avoid this from happening in the future, this changes the definition of the constant to avoid the tool from ever matching it.
r? `@workingjubilee`
mark deprecated option as deprecated in rustc_session to remove copypasta and small refactor
This marks deprecated options as deprecated via flag in options table in rustc_session, which removes copypasted deprecation text from rustc_driver_impl.
This also adds warning for deprecated `-C ar` option, which didn't emitted any warnings before.
Makes `inline_threshold` `[UNTRACKED]`, as it do nothing.
Adds few tests.
See individual commits.
Suggest to replace tuple constructor through projection
See the code example. when `Self::Assoc` normalizes to a struct that has a tuple constructor, you cannot construct the type via `Self::Assoc(field, field)`. Instead, suggest to replace it with the correct named struct.
Fixes#120871
Don't ice on bad transmute in typeck in new solver
Old trait solver ends up getting its infcx tainted because we try to normalize the type, but the new trait solver doesn't. This means we try to compute the stalled transmute obligations, which tries to normalize a type an ICEs. Let's make this a delayed bug.
r? lcnr
Improve diagnostics for `HostEffectPredicate` in the new solver
Adds derived cause for host effect predicates. Some diagnostics regress, but that's connected to the fact that our predicate visitor doesn't play well with aliases just yet.
Add support for wasm exception handling to Emscripten target
This is a draft because we need some additional setting for the Emscripten target to select between the old exception handling and the new exception handling. I don't know how to add a setting like that, would appreciate advice from Rust folks. We could maybe choose to use the new exception handling if `Ctarget-feature=+exception-handling` is passed? I tried this but I get errors from llvm so I'm not doing it right.
See llvm/llvm-project#121851
For LLVM 20+, this function (`renameModuleForThinLTO`) has no return
value. For prior versions of LLVM, this never failed, but had a
signature which allowed an error value people were handling.
The SCCs of the region graph are not a reliable heuristic to use for blaming an interesting
constraint for diagnostics. For region errors, if the outlived region is `'static`, or the involved
types are invariant in their lifetiems, there will be cycles in the constraint graph containing both
the target region and the most interesting constraints to blame. To get better diagnostics in these
cases, this commit removes that heuristic.
`ExtraConstraintInfo` was used only for a single subdiagnostic, so this moves the logic for that
to its own function and eliminates the indirection. In order to do so cleanly, this also changes
the arguments to `BorrowExplanation::add_explanation_to_diagnostic`, which happens to simplify its
call sites.
Before this commit, replace-version-placeholder hardcoded the path
defining CURRENT_RUSTC_VERSION (to avoid replacing it). After a refactor
moved the file defining it without changing the hardcoded path, the tool
started replacing the constant itself with the version number.
To avoid this from happening in the future, this changes the definition
of the constant to avoid the tool from ever matching it.
Suppress host effect predicates if underlying trait doesn't hold
Don't report two errors for when the (`HostEffectPredicate`) `T: const Trait` isn't implemented because (`TraitPredicate`) `T: Trait` doesn't even hold.
Use `PostBorrowckAnalysis` in `check_coroutine_obligations`
This currently errors with:
```
error: concrete type differs from previous defining opaque type use
--> tests/ui/coroutine/issue-52304.rs:10:21
|
10 | pub fn example() -> impl Coroutine {
| ^^^^^^^^^^^^^^ expected `{example::{closure#0} upvar_tys=() resume_ty=() yield_ty=&'{erased} i32 return_ty=() witness={example::{closure#0}}}`, got `{example::{closure#0} upvar_tys=() resume_ty=() yield_ty=&'static i32 return_ty=() witness={example::{closure#0}}}`
|
= note: previous use here
```
This is because we end up redefining the opaque in `check_coroutine_obligations` but with the `yield_ty = &'erased i32` from hir typeck, which causes the *equality* check for opaques to fail.
The coroutine obligtions in question (when `-Znext-solver` is enabled) are:
```
Binder { value: TraitPredicate(<Opaque(DefId(0:5 ~ issue_52304[4c6d]::example::{opaque#0}), []) as std::marker::Sized>, polarity:Positive), bound_vars: [] }
Binder { value: AliasRelate(Term::Ty(Alias(Opaque, AliasTy { args: [], def_id: DefId(0:5 ~ issue_52304[4c6d]::example::{opaque#0}), .. })), Equate, Term::Ty(Coroutine(DefId(0:6 ~ issue_52304[4c6d]::example::{closure#0}), [(), (), &'{erased} i32, (), CoroutineWitness(DefId(0:6 ~ issue_52304[4c6d]::example::{closure#0}), []), ()]))), bound_vars: [] }
Binder { value: AliasRelate(Term::Ty(Coroutine(DefId(0:6 ~ issue_52304[4c6d]::example::{closure#0}), [(), (), &'{erased} i32, (), CoroutineWitness(DefId(0:6 ~ issue_52304[4c6d]::example::{closure#0}), []), ()])), Subtype, Term::Ty(Alias(Opaque, AliasTy { args: [], def_id: DefId(0:5 ~ issue_52304[4c6d]::example::{opaque#0}), .. }))), bound_vars: [] }
```
Ignoring the fact that we end up stalling some really dumb obligations here (lol), I think it makes more sense for us to be using post borrowck analysis for this check anyways.
r? lcnr
inline_threshold mark deprecated
no-stack-check
print deprecation message for -Car too
inline_threshold deprecated and do nothing: make in untracked
make OptionDesc struct from tuple
A few borrowck tweaks to improve 2024 edition migration lints
See first two commits' changes to test outputs. Test coverage in this area is kinda weak, but I think it affects more cases than this (like the craters that will begin to trigger the `tail_expr_drop_order` tests in #134523).
Third commit is a drive-by change that removes a deref hack from `UseSpans` which doesn't really improve diagnostics much.
Mention `unnameable_types` in `unreachable_pub` documentation.
This link makes sense because someone who wishes to avoid unusable `pub` is likely, but not guaranteed, to be interested in avoiding unnameable types.
Also fixed some grammar problems I noticed in the area.
Fixes#116604.
r? Urgau
add m68k-unknown-none-elf target
r? `@workingjubilee`
The existing `m68k-unknown-linux-gnu` target builds `std` by default, requires atomics, and has a base cpu with an fpu. A smaller/more embedded target is desirable both to have a baseline target for the ISA, as well to make debugging easier for working on the llvm backend. Currently this target is using the `M68010` as the minimum CPU due, but as missing features are merged into the `M68k` llvm backend I am hoping to lower this further.
I have been able to build very small crates using a toolchain built against this target (together with a later version of `object`) using the configuration described in the target platform-support documentation, although getting anything of substantial complexity to build quickly hits errors in the llvm backend
This link makes sense because someone who wishes to avoid unusable `pub`
is likely, but not guaranteed, to be interested in avoiding unnameable
types.
Also fixed some grammar problems I noticed in the area.
Fixes#116604.
cg_llvm: Use constants for DWARF opcodes, instead of FFI calls
Split off from #134009 to incorporate feedback from https://github.com/rust-lang/rust/pull/134009#discussion_r1903133906.
Most of the constant values now come from gimli, which is already a compiler dependency.
I noticed that `DW_OP_LLVM_fragment` is an LLVM detail that is not defined by DWARF and could hypothetically change, so I added a static assertion on the C++ side to detect that if it ever happens.
r? workingjubilee
Add a notion of "some ABIs require certain target features"
I think I finally found the right shape for the data and checks that I recently added in https://github.com/rust-lang/rust/pull/133099, https://github.com/rust-lang/rust/pull/133417, https://github.com/rust-lang/rust/pull/134337: we have a notion of "this ABI requires the following list of target features, and it is incompatible with the following list of target features". Both `-Ctarget-feature` and `#[target_feature]` are updated to ensure we follow the rules of the ABI. This removes all the "toggleability" stuff introduced before, though we do keep the notion of a fully "forbidden" target feature -- this is needed to deal with target features that are actual ABI switches, and hence are needed to even compute the list of required target features.
We always explicitly (un)set all required and in-conflict features, just to avoid potential trouble caused by the default features of whatever the base CPU is. We do this *before* applying `-Ctarget-feature` to maintain backward compatibility; this poses a slight risk of missing some implicit feature dependencies in LLVM but has the advantage of not breaking users that deliberately toggle ABI-relevant target features. They get a warning but the feature does get toggled the way they requested.
For now, our logic supports x86, ARM, and RISC-V (just like the previous logic did). Unsurprisingly, RISC-V is the nicest. ;)
As a side-effect this also (unstably) allows *enabling* `x87` when that is harmless. I used the opportunity to mark SSE2 as required on x86-64, to better match the actual logic in LLVM and because all x86-64 chips do have SSE2. This infrastructure also prepares us for requiring SSE on x86-32 when we want to use that for our ABI (and for float semantics sanity), see https://github.com/rust-lang/rust/issues/133611, but no such change is happening in this PR.
r? `@workingjubilee`
Target: Add mips mti baremetal support
Do the same thing as gcc, which use the vendor `mti` to mark the toolchain as MIPS32r2 default.
We support both big endian and little endian flavor:
mips-mti-none-elf
mipsel-mti-none-elf
[Debuginfo] Force enum `DISCR_*` to `static const u64` to allow for inspection via LLDB
see [here](https://rust-lang.zulipchat.com/#narrow/channel/317568-t-compiler.2Fwg-debugging/topic/Revamping.20Debuginfo/near/486614878) for more info.
This change mainly helps `*-msvc` debugged with LLDB. Currently, LLDB cannot inspect `static` struct fields, so the intended visualization for enums is only borderline functional, and niche enums with ranges of discriminant cannot be determined at all .
LLDB *can* inspect `static const` values (though for whatever reason, non-enum/non-u64 consts don't work).
This change adds the `LLVMRustDIBuilderCreateQualifiedType` to the rust FFI layer to wrap the discr type with a `const` modifier, as well as forcing all generated integer enum `DISCR_*` values to be u64's. Those values will only ever be used by debugger visualizers anyway, so it shouldn't be a huge deal, but I left a fixme comment for it just in case.. The `tag` also still properly reflects the discriminant type, so no information is lost.
turn rustc_box into an intrinsic
I am not entirely sure why this was made a special magic attribute, but an intrinsic seems like a more natural way to add magic expressions to the language.
Pass the arch rather than full target name to windows_registry::find_tool
The full target name can be anything with custom target specs. Passing just the arch wasn't possible before cc 1.2, but is now thanks to https://github.com/rust-lang/cc-rs/pull/1285.
try-job: i686-msvc
rustc_intrinsic: support functions without body
We synthesize a HIR body `loop {}` but such bodyless intrinsics.
Most of the diff is due to turning `ItemKind::Fn` into a brace (named-field) enum variant, because it carries a `bool`-typed field now. This is to remember whether the function has a body. MIR building panics to avoid ever translating the fake `loop {}` body, and the intrinsic logic uses the lack of a body to implicitly mark that intrinsic as must-be-overridden.
I first tried actually having no body rather than generating the fake body, but there's a *lot* of code that assumes that all function items have HIR and MIR, so this didn't work very well. Then I noticed that even `rustc_intrinsic_must_be_overridden` intrinsics have MIR generated (they are filled with an `Unreachable` terminator) so I guess I am not the first to discover this. ;)
r? `@oli-obk`
Rollup of 7 pull requests
Successful merges:
- #133964 (core: implement `bool::select_unpredictable`)
- #135001 (Allow using self-contained LLD in bootstrap)
- #135055 (Report impl method has stricter requirements even when RPITIT inference gets in the way)
- #135064 (const-in-pattern: test that the PartialEq impl does not need to be const)
- #135066 (bootstrap: support `./x check run-make-support`)
- #135069 (remove unused function params)
- #135084 (Update carrying_mul_add test to tolerate `nuw`)
r? `@ghost`
`@rustbot` modify labels: rollup
const-in-pattern: test that the PartialEq impl does not need to be const
Fixes https://github.com/rust-lang/rust/issues/119398 by adding a test.
`@compiler-errors` is there some place in the code where we could add a comment saying "as a backcompat hack, here we only require `PartialEq` and not `const PartialEq`"?
r? `@compiler-errors`
Project to `TyKind::Error` when there are unconstrained non-lifetime (ty/const) impl params
It splits the `enforce_impl_params_are_constrained` function into lifetime/non-lifetime, and queryfies the latter. We can then use the result of the latter query (`Result<(), ErrorGuaranteed>`) to intercept projection and constrain the projected type to `TyKind::Error`, which ensures that we leak no ty or const vars to places that don't expect them, like `normalize_erasing_regions`.
The reason we split `enforce_impl_params_are_constrained` into two parts is because we only error for *lifetimes* if the lifetime ends up showing up in any of the associated types of the impl (e.g. we allow `impl<'a> Foo { type Assoc = (); }`). However, in order to compute the `type_of` query for the anonymous associated type of an RPITIT, we need to do trait solving (in `query collect_return_position_impl_trait_in_trait_tys`). That would induce cycles. Luckily, it turns out for lifetimes we don't even care about if they're unconstrained, since they're erased in all contexts that we are trying to fix ICEs. So it's sufficient to keep this check separated out of the query.
I think this is a bit less invasive of an approach compared to #127973. The major difference between this PR and that PR is that we queryify the check instead of merging it into the `explicit_predicates_of` query, and we use the result to taint just projection goals, rather than trait goals too. This doesn't require a lot of new tracking in `ItemCtxt` and `GenericPredicates`, and it also seems to not require any other changes to typeck like that PR did.
Fixes#123141Fixes#125874Fixes#126942Fixes#127804Fixes#130967
r? oli-obk
Improve infer (`_`) suggestions in `const`s and `static`s
Fixes https://github.com/rust-lang/rust/issues/135010.
This PR does a few things to (imo) greatly improve the error message when users write something like `static FOO: [i32; _] = [1, 2, 3]`.
Firstly, it adapts the recovery code for when we encounter `_` in a const/static to work a bit more like `fn foo() -> _`, and removes the somewhat redundant query `diagnostic_only_typeck`.
Secondly, it changes the lowering for `[T; _]` to always lower under the `feature(generic_arg_infer)` logic to `ConstArgKind::Infer`. We still issue the feature error, so it's not doing anything *observable* on the good path, but it does mean that we no longer erroneously interpret `[T; _]`'s array length as a `_` **wildcard expression** (à la destructuring assignment, like `(_, y) = expr`).
Lastly it makes the suggestions verbose and fixes (well, suppresses) a bug with stashing and suggestions.
r? oli-obk
Some type-outlives computation tweaks
Some tweaks that I wrote when investigating https://github.com/rust-lang/rust/issues/135006.
The only commit that's probably interesting here is f3646748cd (the first commit). For some reason it was concerned with filtering out param-env outlives clauses when they matched item-bound outlives clauses. However, if you look at the rest of the control flow for that function, not filtering out those bounds doesn't actually affect the behavior materially.
Pass objcopy args for stripping on OSX
When `-Cstrip` was changed in #131405 to use the bundled rust-objcopy instead of /usr/bin/strip on OSX, strip-like arguments were preserved.
But strip and objcopy are, while being the same binary, different, they have different defaults depending on which binary they are. Notably, strip strips everything by default, and objcopy doesn't strip anything by default.
Additionally, `-S` actually means `--strip-all`, so debuginfo stripped everything and symbols didn't strip anything.
We now correctly pass `--strip-debug` and `--strip-all`.
fixes#135028
try-job: aarch64-apple
try-job: dist-aarch64-apple
Do the same thing as gcc, which use the vendor `mti` to mark
the toolchain as MIPS32r2 default.
We support both big endian and little endian flavor:
mips-mti-none-elf
mipsel-mti-none-elf
taint fcx on selection errors during unsizing
With `feature(dyn_compatible_for_dispatch)` we only check for dyn-compatibility by checking the `T: Unsize<dyn Trait>` predicate during the unsizing coercions checks. If the predicate doesn't hold, we emit an error, but pretend the coercion succeeded to prevent further errors. To prevent const eval from attempting to actually perform this coercion, we need to taint the fcx after reporting the trait errors in the coercion check.
fixes https://github.com/rust-lang/rust/issues/135021
fixes https://github.com/rust-lang/rust/issues/130521
When `-Cstrip` was changed to use the bundled rust-objcopy instead of
/usr/bin/strip on OSX, strip-like arguments were preserved.
But strip and objcopy are, while being the same binary, different, they
have different defaults depending on which binary they are.
Notably, strip strips everything by default, and objcopy doesn't strip
anything by default.
Additionally, `-S` actually means `--strip-all`, so debuginfo stripped
everything and symbols didn't strip anything.
We now correctly pass `--strip-debug` and `--strip-all`.
Range metadata was disabled for amdgpu due to a backend bug. I did not
encounter any problems when removing the workaround to enable range
metadata (tried compiling `core` and `alloc`), so I assume this has
been fixed in LLVM in the last years.
Remove the workaround to re-enable range metadata.
Pointers for variables all need to be in the same address space for
correct compilation. Therefore ensure that even if a global variable is
created in a different address space, it is casted to the default
address space before its value is used.
This is necessary for the amdgpu target and others where the default
address space for global variables is not 0.
For example `core` does not compile in debug mode when not casting the
address space to the default one because it tries to emit the following
(simplified) LLVM IR, containing a type mismatch:
```llvm
@alloc_0 = addrspace(1) constant <{ [6 x i8] }> <{ [6 x i8] c"bit.rs" }>, align 1
@alloc_1 = addrspace(1) constant <{ ptr }> <{ ptr addrspace(1) @alloc_0 }>, align 8
; ^ here a struct containing a `ptr` is needed, but it is created using a `ptr addrspace(1)`
```
For this to compile, we need to insert a constant `addrspacecast` before
we use a global variable:
```llvm
@alloc_0 = addrspace(1) constant <{ [6 x i8] }> <{ [6 x i8] c"bit.rs" }>, align 1
@alloc_1 = addrspace(1) constant <{ ptr }> <{ ptr addrspacecast (ptr addrspace(1) @alloc_0 to ptr) }>, align 8
```
As vtables are global variables as well, they are also created with an
`addrspacecast`. In the SSA backend, after a vtable global is created,
metadata is added to it. To add metadata, we need the non-casted global
variable. Therefore we strip away an addrspacecast if there is one, to
get the underlying global.
Autodiff Upstreaming - rustc_codegen_llvm changes
Now that the autodiff/Enzyme backend is merged, this is an upstream PR for the `rustc_codegen_llvm` changes.
It also includes small changes to three files under `compiler/rustc_ast`, which overlap with my frontend PR (https://github.com/rust-lang/rust/pull/129458).
Here I only include minimal definitions of structs and enums to be able to build this backend code.
The same goes for minimal changes to `compiler/rustc_codegen_ssa`, the majority of changes there will be in another PR, once either this or the frontend gets merged.
We currently have 68 files left to merge, 19 in the frontend PR, 21 (+3 from the frontend) in this PR, and then ~30 in the middle-end.
This PR is large because it includes two of my three large files (~800 loc each). I could also first only upstream enzyme_ffi.rs, but I think people might want to see some use of these bindings in the same PR?
To already highlight the things which reviewers might want to discuss:
1) `enzyme_ffi.rs`: I do have a fallback module to make sure that we don't link rustc against Enzyme when we build rustc without autodiff support.
2) `add_panic_msg_to_global` was a pain to write and I currently can't even use it. Enzyme writes gradients into shadow memory. Pass in one float scalar? We'll allocate and return an extra float telling you how this float affected the output. Pass in a slice of floats? We'll let you allocate the vector and pass in a mutable reference to a float slice, we'll then write the gradient into that slice. It should be at least as large as your original slice, so we check that and panic if not. Currently we panic silently, but I already generate a nicer panic message with this function. I just don't know how to print it to the user. yet. I discussed this with a few rustc devs and the best we could come up with (for now), was to look for mangled panic calls in the IR and pick one, which works surprisingly reliably. If someone knows a good way to clean this up and print the panic message I'm all in, otherwise I can remove the code that writes the nicer panic message and keep the silent panic, since it's enough for soundness. Especially since this PR is already a bit larger.
3) `SanitizeHWAddress`: When differentiating C++, Enzyme can use TBAA to "understand" enums/unions, but for Rust we don't have this information. LLVM might to speculative loads which (without TBAA) confuse Enzyme, so we disable those with this attribute. This attribute is only set during the first opt run before Enzyme differentiates code. We then remove it again once we are done with autodiff and run the opt pipeline a second time. Since enums are everywhere in Rust, support for them is crucial, but if this looks too cursed I can remove these ~100 lines and keep them in my fork for now, we can then discuss them separately to make this PR simpler?
4) Duplicated llvm-opt runs: Differentiating already optimized code (and being able to do additional optimizations on the fly, e.g. for GPU code) is _the_ reason why Enzyme is so fast, so the compile time is acceptable for autodiff users: https://enzyme.mit.edu/talks/Publications/ (There are also algorithmic issues in Enzyme core which are more serious than running opt twice).
5) I assume that if we merge these minimal cg_ssa changes here already, I also need to fix the other backends (GCC and cliff) to have dummy implementations, correct?
6) *I'm happy to split this PR up further if reviewers have recommendations on how to.*
For the full implementation, see: https://github.com/rust-lang/rust/pull/129175
Tracking:
- https://github.com/rust-lang/rust/issues/124509
Provide structured suggestion for `impl Default` of type where all fields have defaults
```
error: `Default` impl doesn't use the declared default field values
--> $DIR/manual-default-impl-could-be-derived.rs:28:1
|
LL | / impl Default for B {
LL | | fn default() -> Self {
LL | | B {
LL | | x: s(),
| | --- this field has a default value
LL | | y: 0,
| | - this field has a default value
... |
LL | | }
| |_^
|
help: to avoid divergence in behavior between `Struct { .. }` and `<Struct as Default>::default()`, derive the `Default`
|
LL ~ #[derive(Default)] struct B {
|
```
Note that above the structured suggestion also includes completely removing the manual `impl`, but the rendering doesn't.
Some small nits to the borrowck suggestions for mutating a map through index
1. Suggesting users to either use `.insert` or `.get_mut` (which do totally different things) can be a bit of a footgun, so let's make that a bit more nuanced.
2. I find the suggestion of `.get_mut(|val| { *val = whatever; })` to be a bit awkward. I changed this to be an if-let instead.
3. Fix a bug which was suppressing the structured suggestion for some mutations via the index operator on `HashMap`/`BTreeMap`.
r? estebank or reassign
E0277: suggest dereferencing function arguments in more cases
This unifies and generalizes some of the logic in `TypeErrCtxt::suggest_dereferences` so that it will suggest dereferencing arguments to function/method calls in order to satisfy trait bounds in more cases.
Previously it would only fire on reference types, and it had two separate cases (one specifically to get through custom `Deref` impls when passing by-reference, and one specifically to catch #87437). I've based the new checks loosely on what's done for `E0308` in `FnCtxt::suggest_deref_or_ref`: it will suggest dereferences to satisfy trait bounds whenever the referent is `Copy`, is boxed (& so can be moved out of the boxes), or is being passed by reference.
This doesn't make the suggestion fire in contexts other than function arguments or binary operators (which are in a separate case that this doesn't touch), and doesn't make it suggest a combination of `&`-removal and dereferences. Those would require a bit more restructuring, so I figured just doing this would be a decent first step.
Closes#90997
borrowck diagnostics: make `add_move_error_suggestions` use the HIR rather than `SourceMap`
This PR aims to fix#132806 by rewriting `add_move_error_suggestions`[^1]. Previously, it manually scanned the source text to find a leading `&`, which isn't always going to produce a correct result (see: that issue). Admittedly, the HIR visitor in this PR introduces a lot of boilerplate, but hopefully the logic at its core isn't too complicated (I go over it in the comments). I also tried a simpler version that didn't use a HIR visitor and suggested adding `ref` always, but the `&ref x` suggestions really didn't look good. As a bonus for the added complexity though, it's now able to produce nice `&`-removing suggestions in more cases.
I tried to do this such that it avoids edition-dependent checks and its suggestions can be applied together with those from the match ergonomics 2024 migration lint. I haven't added tests for that since the details of match ergonomics 2024 are still being sorted out, but I can try if desired once that's finalized.
[^1]: In brief, it fires on patterns where users try to bind by-value in such a way that moves out of a reference to a non-Copy type (including slice references with non-copy elements). The suggestions are to change the binding's mode to be by-reference, either by removing[^2] an enclosing `&`/`&mut` or adding `ref` to the binding.
[^2]: Incidentally, I find the terminology of "consider removing the borrow" a bit confusing for a suggestion to remove a `&` pattern in order to make bindings borrow rather than move. I'm not sure what a good, concise way to explain that would be though, and that should go in a separate PR anyway.
```
error: `Default` impl doesn't use the declared default field values
--> $DIR/manual-default-impl-could-be-derived.rs:28:1
|
LL | / impl Default for B {
LL | | fn default() -> Self {
LL | | B {
LL | | x: s(),
| | --- this field has a default value
LL | | y: 0,
| | - this field has a default value
... |
LL | | }
| |_^
|
help: to avoid divergence in behavior between `Struct { .. }` and `<Struct as Default>::default()`, derive the `Default`
|
LL ~ #[derive(Default)] struct B {
|
```
Note that above the structured suggestion also includes completely removing the manual `impl`, but the rendering doesn't.
Account for C string literals and `format_args` in `HiddenUnicodeCodepoints` lint
This is stacked on #134955, and either that can land first or both of them can land together here. I split this out because this is a bit more involved of an impl.
Fixes#94945
compiler: Add a statement-of-intent to `rustc_abi`
This just documents the most basic idea of what the crate is even for in my view, rather than leaving that strewn about GitHub issues, PR reviews, and Zulip streams. In particular, I hope to make it clearer what code should go in `rustc_abi` and what should not, which is of immediate relevance to contributors.
I considered going even further and explaining ideas like "ABI compatibility", prologues, and so on. However, because of the cross-cutting nature of ABI, I think such explanations should probably live in the place for cross-cutting documents: the rustc dev guide. This is only meant to be a quick "by the way".
explicitly set float ABI for all ARM targets
We currently always set the `FloatABIType` field in the LLVM target machine to `Default`, which means LLVM infers the ARM float ABI (hard vs soft) from the LLVM target triple. This causes problems such as having to set the LLVM triple to `*-gnueabi` for our `musleabi` targets to ensure they get correctly inferred as soft-float targets. It also means rustc doesn't really know which float ABI ends up being used, which is a blocker for https://github.com/rust-lang/rust/pull/134794. So I think we should stop doing that and instead explicitly control that value. That's what this PR implements.
See [Zulip](https://rust-lang.zulipchat.com/#narrow/channel/187780-t-compiler.2Fwg-llvm/topic/Softfloat.20ABI.2C.20hardfloat.20instructions) for more context.
Best reviewed commit-by-commit. I hope I got all those `llvm_floatabi` values right...
stabilize const_swap
libs-api FCP passed in https://github.com/rust-lang/rust/issues/83163.
However, I only just realized that this actually involves an intrinsic. The intrinsic could be implemented entirely with existing stable const functionality, but we choose to make it a primitive to be able to detect more UB. So nominating for `@rust-lang/lang` to make sure they are aware; I leave it up to them whether they want to FCP this.
While at it I also renamed the intrinsic to make the "nonoverlapping" constraint more clear.
Fixes#83163
A couple datalog/borrowck cleanups
As discussed on zulip, here's a chill one in between slightly more interesting PRs:
- I hadn't noticed there still were a couple of datalog-related modules outside of their dedicated `polonius` module (go to horn-clause jail, bonk!).
- there somehow was both a `diags` module and a `diagnostics` module.
- a couple other tiny things being renamed -- let me know what you think.
As requested I've tried to have somewhat granular commits to ease review, but the last two or three could be squashed together, since they're all related to the `diags` module (but moving its contents is less tedious to check in its own commit).
r? `@jackh726`
rustc_codegen_ssa: Buffer file writes in link_rlib
This makes this step take ~25ms on my machine (M3 Max 64GB) for Zed repo instead of ~150ms (on editor crate). Additionally it takes down the time needed for a clean cargo build of ripgrep from ~6.1s to 5.9s.
This change is mostly relevant for dev builds of crates with multiple large CGUs.
I imagine it could be quite relevant for dev scenarios on Windows, but sadly I have no way to measure that myself.
Compute liveness constraints in location-sensitive polonius
This continues the location-sensitive prototype. In this episode, we build the liveness constraints.
Reminder of the approach we're taking: we need variance data to create liveness edges in the forward/backward/both directions (respectively in the cases of covariance, contravariance, invariance) in the localized constraint graph.
This PR:
- introduces the holder for that, and for the liveness data in the correct shape: the transpose of what we're using today, "live regions per points".
- records use/drop live region variance during tracing
- records regular live region variance at the end of liveness
- records the correctly shaped live region per point matrix
- uses all of the above to compute the liveness constraints
(There's still technically one tiny part of the liveness owl left to do, but I'll leave it for a future PR: we also need to disable the NLL optimization that avoids computing liveness for locals whose types contain a region outliving a free region -- the existing constraints make it effectively live at all points; this doesn't work under polonius)
r? `@jackh726` cc `@matthewjasper`
Rollup of 4 pull requests
Successful merges:
- #134870 (Fix sentence fragment in `pin` module docs)
- #134884 (Fix typos)
- #134892 (Added codegen test for elidings bounds check when indexes are manually checked)
- #134894 (Document how to run the split Docker pipelines)
r? `@ghost`
`@rustbot` modify labels: rollup
Improve default target options for x86_64-unknown-linux-none
Without a standard library, we cannot unwind, so it should be panic=abort by default.
Additionally, it does not have std because while it is Linux, it cannot use libc, which std uses today for Linux.
Using PIE by default may be surprising to users, as shown in #134763, so I've documented it explicitly. I'm not sure if we want to count that as fixing the issue or not.
cc `@morr0ne,` as you added the target (and are the maintainer), and `@Noratrieb,` who reviewed that PR (:D).
- add a FIXME when looking for the region variance of unexpected regions
- drive-by: fix a doc comment link
- drive-by: simplify the variance match using exported variants instead
This context struct will hold data to help creating localized
constraints:
- the live regions, with the shape matching a CFG walk, indexed per
point
- the variance of these live regions, represented as the direction we'll
add the appropriate
We also add this structure to the mir typeck to record liveness data,
and make it responsible for localized constraint creation.
Avoid ICE in borrowck
Provide a fallback in `best_blame_constraint` when `find_constraint_paths_between_regions` doesn't have a result. This code is due a rework to avoid the letf-over `unwrap()`, but avoids the ICE caused by the repro.
Fix#133252.
This makes this step take ~25ms on my machine (M3 Max 64GB) for Zed repo instead of ~150ms. Additionally it takes down the time needed for a clean cargo build of ripgrep from ~6.1s to 5.9s.
This change is mostly relevant for crates with multiple large CGUs.
Implement `default_overrides_default_fields` lint
Detect when a manual `Default` implementation isn't using the existing default field values and suggest using `..` instead:
```
error: `Default` impl doesn't use the declared default field values
--> $DIR/manual-default-impl-could-be-derived.rs:14:1
|
LL | / impl Default for A {
LL | | fn default() -> Self {
LL | | A {
LL | | y: 0,
| | - this field has a default value
... |
LL | | }
| |_^
|
= help: use the default values in the `impl` with `Struct { mandatory_field, .. }` to avoid them diverging over time
note: the lint level is defined here
--> $DIR/manual-default-impl-could-be-derived.rs:5:9
|
LL | #![deny(default_overrides_default_fields)]
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
```
r? `@compiler-errors`
This is a simpler version of #134441, detecting the simpler case when a field with a default should have not been specified in the manual `Default::default()`, instead using `..` for it. It doesn't provide any suggestions, nor the checks for "equivalences" nor whether the value used in the imp being used would be suitable as a default field value.
Skip parenthesis around tuple struct field calls
The pretty-printer previously did not distinguish between named vs unnamed fields when printing a function call containing a struct field. It would print the call as `(self.fun)()` for a named field which is correct, and `(self.0)()` for an unnamed field which is redundant.
This PR changes function calls of tuple struct fields to print without parens.
**Before:**
```rust
struct Tuple(fn());
fn main() {
let tuple = Tuple(|| {});
(tuple.0)();
}
```
**After:**
```rust
struct Tuple(fn());
fn main() {
let tuple = Tuple(|| {});
tuple.0();
}
```
Skip parenthesis if `.` makes statement boundary unambiguous
There is a rule in the parser that statements and match-arms never end in front of a `.` or `?` token (except when the `.` is really `..` or `..=` or `...`). So some of the leading subexpressions that need parentheses inserted when followed by some other operator like `-` or `+`, do not need parentheses when followed by `.` or `?`.
Example:
```rust
fn main() {
loop {}.to_string() + "";
match () {
_ => loop {}.to_string() + "",
};
}
```
`-Zunpretty=expanded` before:
```console
#![feature(prelude_import)]
#[prelude_import]
use std::prelude::rust_2021::*;
#[macro_use]
extern crate std;
fn main() {
(loop {}).to_string() + "";
match () { _ => (loop {}).to_string() + "", };
}
```
After:
```console
#![feature(prelude_import)]
#[prelude_import]
use std::prelude::rust_2021::*;
#[macro_use]
extern crate std;
fn main() {
loop {}.to_string() + "";
match () { _ => loop {}.to_string() + "", };
}
```
Make `ty::Error` implement all auto traits
I have no idea what's up with the crashes test I fixed--I really don't want to look into it since it has to do something with borrowck and multiple layers of opaques. I think the underlying idea of allowing error types to implement all auto traits is justified though.
Fixes#134796Fixes#131050
r? lcnr
Add a compiler intrinsic to back `bigint_helper_methods`
cc https://github.com/rust-lang/rust/issues/85532
This adds a new `carrying_mul_add` intrinsic, to implement `wide_mul` and `carrying_mul`.
It has fallback MIR for all types -- including `u128`, which isn't currently supported on nightly -- so that it'll continue to work on all backends, including CTFE.
Then it's overridden in `cg_llvm` to use wider intermediate types, including `i256` for `u128::carrying_mul`.
Spruce up the docs of several queries related to the type/trait system and const eval
- Editorial
- Proper rustdoc summary/synopsis line by making use of extra paragraphs: Leads to better rendered output on module pages, in search result lists and overall, too
- Use rustdoc warning blocks for admonitions of the form "do not call / avoid calling this query directly"
- Use intra-doc links of the form ``[`Self::$query`]`` to cross-link queries. Indeed, such links are generally a bit brittle due to the existence of `TyCtxtFeed` which only contains a subset of queries. Therefore the docs of `feedable` queries cannot cross-link to non-`feedable` ones. I'd say it's fine to use intra-doc links despite the potential/unlikely occasional future breakage (if a query with the aforementioned characteristics becomes `feedable`). `Self::` is nicer than `TyCtxt::` (which would be more stable) since it accounts for other contexts like `TyCtxt{Feed,At,Ensure{,WithValue}}`
- Informative
- Generally add, flesh out and correct some doc comments
- Add *Panic* sections (to a few selected queries only). The lists of panics aren't necessarily exhaustive and focus on the more "obvious" or "important" panics.
- Where applicable add a paragraph calling attention to the relevant [`#[rustc_*]` TEST attribute](https://rustc-dev-guide.rust-lang.org/compiler-debugging.html#rustc_-test-attributes)
The one non-doc change (it's internal and not observable):
Be even more defensive in `query constness`'s impl (spiritual follow-up to #134122) (see self review comment).
Fixes#133494.
r\? **any**(compiler-errors, oli-obk)
ptr::copy: fix docs for the overlapping case
Fixes https://github.com/rust-lang/unsafe-code-guidelines/issues/549
As discussed in that issue, it doesn't make any sense for `copy` to read a byte via `src` after it was already written via `dst`. The entire point of this method is that is copies correctly even if they overlap, and that requires always reading any given location before writing it.
Cc `@rust-lang/opsem`
[macro_metavar_expr_concat] Fix#128346Fix#128346Fix#131393
The syntax is invalid in both issues so I guess that theoretically the compiler should have aborted early.
This PR tries to fix a local problem but let me know if there are better options.
cc `@petrochenkov` if you are interested
fix default-backtrace-ice test
when running `tests/ui/panics/default-backtrace-ice.rs locally it gave this error:
```
failures:
---- [ui] tests/ui/panics/default-backtrace-ice.rs stdout ----
Saved the actual stderr to "/home/jyn/src/rust3/build/x86_64-unknown-linux-gnu/test/ui/panics/default-backtrace-ice/default-backtrace-ice.stderr"
diff of stderr:
7
8 aborting due to `-Z treat-err-as-bug=1`
9 stack backtrace:
- (end_short_backtrace)
- (begin_short_backtrace)
- (end_short_backtrace)
- (begin_short_backtrace)
+ [... omitted 22 frames ...]
+
```
(note that you must *not* use --bless; we previously did not have an error annotation to verify it was a full backtrace instead of a short backtrace.)
this is a regression from setting RUST_BACKTRACE=1 by default in https://github.com/rust-lang/rust/pull/134743. we need to turn off the new behavior when running UI tests so that they reflect our dist compiler. normally that's done by checking `sess.unstable_opts.ui_testing`, but this happens extremely early in the compiler before we've expanded arg files. do an extremely simple hack that doesn't work in all cases - we don't need it to work in all cases, only when running UI tests.
cc https://github.com/rust-lang/rust/pull/129658#issuecomment-2561988081
r? `@jieyouxu`
when running `tests/ui/panics/default-backtrace-ice.rs locally it gave this error:
```
failures:
---- [ui] tests/ui/panics/default-backtrace-ice.rs stdout ----
Saved the actual stderr to "/home/jyn/src/rust3/build/x86_64-unknown-linux-gnu/test/ui/panics/default-backtrace-ice/default-backtrace-ice.stderr"
diff of stderr:
7
8 aborting due to `-Z treat-err-as-bug=1`
9 stack backtrace:
- (end_short_backtrace)
- (begin_short_backtrace)
- (end_short_backtrace)
- (begin_short_backtrace)
+ [... omitted 22 frames ...]
+
```
this is a regression from setting RUST_BACKTRACE=1 by default. we need to turn off the new behavior when running UI tests so that they reflect our dist compiler. normally that's done by checking `sess.unstable_opts.ui_testing`, but this happens extremely early in the compiler before we've expanded arg files. do an extremely simple hack that doesn't work in all cases - we don't need it to work in all cases, only when running UI tests.
Detect when a manual `Default` implementation isn't using the existing default field values and suggest using `..` instead:
```
error: `Default` impl doesn't use the declared default field values
--> $DIR/manual-default-impl-could-be-derived.rs:14:1
|
LL | / impl Default for A {
LL | | fn default() -> Self {
LL | | A {
LL | | y: 0,
| | - this field has a default value
... |
LL | | }
| |_^
|
= help: use the default values in the `impl` with `Struct { mandatory_field, .. }` to avoid them diverging over time
note: the lint level is defined here
--> $DIR/manual-default-impl-could-be-derived.rs:5:9
|
LL | #![deny(default_overrides_default_fields)]
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
```
Without a standard library, we cannot unwind, so it should be
panic=abort by default.
Additionally, it does not have std because while it is
Linux, it cannot use libc, which std uses today for Linux.
Update `#[coverage(..)]` attribute error messages to match the current implementation
The allowed positions for `#[coverage(..)]` attributes were expanded by #126721, but the corresponding error messages were never updated to reflect the new behaviour.
Part of #134749.
Default to short backtraces for dev builds of rustc itself
A dev build almost certainly means that whoever's built the compiler has the opportunity to rerun it to collect a more complete trace. So we don't need to default to a complete trace; we should hide irrelevant details by default.
A dev build almost certainly means that whoever's built the compiler
has the opportunity to rerun it to collect a more complete trace. So
we don't need to default to a complete trace; we should hide irrelevant
details by default.
Correctly note item kind in `NonConstFunctionCall` error message
Don't just call everything a "`fn`". This is more consistent with the error message we give for conditionally-const items, which do note the item's def kind.
r? fmease, this is a prerequisite for making those `~const PartialEq` error messages better. Re-roll if you're busy or don't want to review this.
Begin to implement type system layer of unsafe binders
Mostly TODOs, but there's a lot of match arms that are basically just noops so I wanted to split these out before I put up the MIR lowering/projection part of this logic.
r? oli-obk
Tracking:
- https://github.com/rust-lang/rust/issues/130516
Use `#[derive(Default)]` instead of manual `impl` when possible
While working on #134175 I noticed a few manual `Default` `impl`s that could be `derive`d instead. These likely predate the existence of the `#[default]` attribute for `enum`s.
Revert stabilization of the `#[coverage(..)]` attribute
Due to a process mixup, the PR to stabilize the `#[coverage(..)]` attribute (#130766) was merged while there are still outstanding concerns. The default action in that situation is to revert, and the feature is not sufficiently urgent or uncontroversial to justify special treatment, so this PR reverts that stabilization.
---
- A key point that came up in offline discussions is that unlike most user-facing features, this one never had a proper RFC, so parts of the normal stabilization process that implicitly rely on an RFC break down in this case.
- As the implementor and de-facto owner of the feature in its current form, I would like to think that I made good choices in designing and implementing it, but I don't feel comfortable proceeding to stabilization without further scrutiny.
- There hasn't been a clear opportunity for T-compiler to weigh in or express concerns prior to stabilization.
- The stabilization PR cites a T-lang FCP that occurred in the tracking issue, but due to the messy design and implementation history (and lack of a clear RFC), it's unclear what that FCP approval actually represents in this case.
- At the very least, we should not proceed without a clear statement from T-lang or the relevant members about the team's stance on this feature, especially in light of the other concerns listed here.
- The existing user-facing documentation doesn't clearly reflect which parts of the feature are stable commitments, and which parts are subject to change. And there doesn't appear to be a clear consensus anywhere about where that line is actually drawn, or whether the chosen boundary is acceptable to the relevant teams and individuals.
- For example, the [stabilization report comment](https://github.com/rust-lang/rust/issues/84605#issuecomment-2166514660) mentions that some aspects are subject to change, but that text isn't consistent with my earlier comments, and there doesn't appear to have been any explicit discussion or approval process.
- [The current reference text](https://github.com/rust-lang/reference/blob/4dfaa4f/src/attributes/coverage-instrumentation.md) doesn't mention this distinction at all, and instead simply describes the current implementation behaviour.
- When the implementation was changed to its current form, the associated user-facing error messages were not updated, so they still refer to the attribute only being allowed on functions and closures.
- On its own, this might have been reasonable to fix-forward in the absence of other concerns, but the fact that it never came up earlier highlights the breakdown in process that has occurred here.
---
Apologies to everyone who was excited for this stabilization to land, but unfortunately it simply isn't ready yet.
cleanup `TypeVerifier`
We should merge it with the `TypeChecker` as we no longer bail in cases where it encounters an error since #111863.
It's quite inconsistent whether a check lives in the verifier or the `TypeChecker`, so this feels like a quite impactful cleanup. I expect that for this we may want to change the `TypeChecker` to also be a MIR visitor 🤔 this is non-trivial so I didn't fully do it in this PR.
Best reviewed commit by commit.
r? `@compiler-errors` feel free to reassign however
Rollup of 6 pull requests
Successful merges:
- #130289 (docs: Permissions.readonly() also ignores root user special permissions)
- #134583 (docs: `transmute<&mut T, &mut MaybeUninit<T>>` is unsound when exposed to safe code)
- #134611 (Align `{i686,x86_64}-win7-windows-msvc` to their parent targets)
- #134629 (compiletest: Allow using a specific debugger when running debuginfo tests)
- #134642 (Implement `PointerLike` for `isize`, `NonNull`, `Cell`, `UnsafeCell`, and `SyncUnsafeCell`.)
- #134660 (Fix spacing of markdown code block fences in compiler rustdoc)
r? `@ghost`
`@rustbot` modify labels: rollup
Fix spacing of markdown code block fences in compiler rustdoc
Two place have misaligned open and close ```` ``` ````.
I noticed these because one of them disrupted syntax highlighting in my editor for the rest of the file as I was working on a different change.
<p align="center"><img src="https://github.com/user-attachments/assets/5de21d08-c30c-4e9c-8587-e83b988b9db5" width="500"></p>
Align `{i686,x86_64}-win7-windows-msvc` to their parent targets
There were some changes to `{i686,x86_64}-pc-windows-msvc`, include them in the backward compatibility targets as well.
CC `@roblabla`
Use `PtrMetadata` instead of `Len` in slice drop shims
I tried to do a bigger change in #134297 which didn't work, so here's the part I really wanted: Removing another use of `Len`, in favour of `PtrMetadata`.
Split into two commits where the first just adds a test, so you can look at the second commit to see how the drop shim for an array changes with this PR.
Reusing the same reviewer from the last one:
r? BoxyUwU
coroutine_clone: add comments
I was very surprised to learn that coroutines can be cloned. This has non-trivial semantic consequences that I do not think have been considered. Lucky enough, it's still unstable. Let's add some comments and pointers so we hopefully become aware when a MIR opt actually is in conflict with this.
Cc `@rust-lang/wg-mir-opt`
Explain why a type is not eligible for `impl PointerLike`.
The rules were baffling when I ran in to them trying to add some impls (to `std`, not my own code, as it happens), so I made the compiler explain them to me.
The logic of the successful cases is unchanged, but I did rearrange it to reverse the order of the primitive and `Adt` cases; this makes producing the errors easier. I'm still not very familiar with `rustc` internals, so let me know if there's a better way to do any of this.
This also adds test coverage for which impls are accepted or rejected, which I didn't see any of already.
The PR template tells me I should consider mentioning a tracking issue, but there isn't one for `pointer_like_trait`, so I'll mention `dyn_star`: #102425
Use E0665 for missing `#[default]` on enum and update doc
The docs for E0665 when doing `#[derive(Default]` on an `enum` previously didn't mention `#[default]` at all, or made a distinction between unit variants, that can be annotated, and tuple or struct variants, which cannot.
E0665 was not being emitted, we now use it for the same error it belonged to before.
```
error[E0665]: `#[derive(Default)]` on enum with no `#[default]`
--> $DIR/macros-nonfatal-errors.rs:42:10
|
LL | #[derive(Default)]
| ^^^^^^^
LL | / enum NoDeclaredDefault {
LL | | Foo,
LL | | Bar,
LL | | }
| |_- this enum needs a unit variant marked with `#[default]`
|
= note: this error originates in the derive macro `Default` (in Nightly builds, run with -Z macro-backtrace for more info)
help: make this unit variant default by placing `#[default]` on it
|
LL | #[default] Foo,
| ++++++++++
help: make this unit variant default by placing `#[default]` on it
|
LL | #[default] Bar,
| ++++++++++
```
update `rustc_index_macros` feature handling
It seems that cargo can't [conditionally propagate features](214587c89d/compiler/rustc_index/Cargo.toml (L20)) if they were enabled by default on the target crate, but disabled with `default-features = false` in the current/parent crate.
Fixes#118129
Foundations of location-sensitive polonius
I'd like to land the prototype I'm describing in the [polonius project goal](https://github.com/rust-lang/rust-project-goals/issues/118). It still is incomplete and naive and terrible but it's working "well enough" to consider landing.
I'd also like to make review easier by not opening a huge PR, but have a couple small-ish ones (the +/- line change summary of this PR looks big, but >80% is moving datalog to a single place).
This PR starts laying the foundation for that work:
- it refactors and collects 99% of the old datalog fact gen, which was spread around everywhere, into a single dedicated module. It's still present at 3 small places (one of which we should revert anyways) that are kinda deep within localized components and are not as easily extractable into the rest of fact gen, so it's fine for now.
- starts introducing the localized constraints, the building blocks of the naive way of implementing the location-sensitive analysis in-tree, which is roughly sketched out in https://smallcultfollowing.com/babysteps/blog/2023/09/22/polonius-part-1/ and https://smallcultfollowing.com/babysteps/blog/2023/09/29/polonius-part-2/ but with a different vibe than per-point environments described in these posts, just `r1@p: r2@q` constraints.
- sets up the skeleton of generating these localized constraints: converting NLL typeck constraints, and creating liveness constraints
- introduces the polonius dual to NLL MIR to help development and debugging. It doesn't do much currently but is a way to see these localized constraints: it's an NLL MIR dump + a dumb listing of the constraints, that can be dumped with `-Zdump-mir=polonius -Zpolonius=next`. Its current state is not intended to be a long-term thing, just for testing purposes -- I will replace its contents in the future with a different approach (an HTML+js file where we can more easily explore/filter/trace these constraints and loan reachability, have mermaid graphs of the usual graphviz dumps, etc).
I've started documenting the approach in this PR, I'll add more in the future. It's quite simple, and should be very clear when more constraints are introduced anyways.
r? `@matthewjasper`
Best reviewed per commit so that the datalog move is less bothersome to read, but if you'd prefer we separate that into a different PR, I can do that (and michael has offered to review these more mechanical changes if it'd help).
Use orphaned error code for the same error it belonged to before.
```
error[E0665]: `#[derive(Default)]` on enum with no `#[default]`
--> $DIR/macros-nonfatal-errors.rs:42:10
|
LL | #[derive(Default)]
| ^^^^^^^
LL | / enum NoDeclaredDefault {
LL | | Foo,
LL | | Bar,
LL | | }
| |_- this enum needs a unit variant marked with `#[default]`
|
= note: this error originates in the derive macro `Default` (in Nightly builds, run with -Z macro-backtrace for more info)
help: make this unit variant default by placing `#[default]` on it
|
LL | #[default] Foo,
| ~~~~~~~~~~~~~~
help: make this unit variant default by placing `#[default]` on it
|
LL | #[default] Bar,
| ~~~~~~~~~~~~~~
```
Provide a fallback in `best_blame_constraint` when `find_constraint_paths_between_regions` doesn't have a result. This code is due a rework to avoid the letf-over `unwrap()`, but avoids the ICE caused by the repro.
Fix#133252.
handle member constraints directly in the mir type checker
cleaner, faster, easier to change going forward :> fixes#109654
r? `@oli-obk` `@compiler-errors`
Fix parenthesization of chained comparisons by pretty-printer
Example:
```rust
macro_rules! repro {
() => {
1 < 2
};
}
fn main() {
let _ = repro!() == false;
}
```
Previously `-Zunpretty=expanded` would pretty-print this syntactically invalid output: `fn main() { let _ = 1 < 2 == false; }`
```console
error: comparison operators cannot be chained
--> <anon>:8:23
|
8 | fn main() { let _ = 1 < 2 == false; }
| ^ ^^
|
help: parenthesize the comparison
|
8 | fn main() { let _ = (1 < 2) == false; }
| + +
```
With the fix, it will print `fn main() { let _ = (1 < 2) == false; }`.
Making `-Zunpretty=expanded` consistently produce syntactically valid Rust output is important because that is what makes it possible for `cargo expand` to format and perform filtering on the expanded code.
## Review notes
According to `rg '\.fixity\(\)' compiler/` the `fixity` function is called only 3 places:
- 13170cd787/compiler/rustc_ast_pretty/src/pprust/state/expr.rs (L283-L287)
- 13170cd787/compiler/rustc_hir_pretty/src/lib.rs (L1295-L1299)
- 13170cd787/compiler/rustc_parse/src/parser/expr.rs (L282-L289)
The 2 pretty printers definitely want to treat comparisons using `Fixity::None`. That's the whole bug being fixed. Meanwhile, the parser's `Fixity::None` codepath is previously unreachable as indicated by the comment, so as long as `Fixity::None` here behaves exactly the way that `Fixity::Left` used to behave, you can tell that this PR definitely does not constitute any behavior change for the parser.
My guess for why comparison operators were set to `Fixity::Left` instead of `Fixity::None` is that it's a very old workaround for giving a good chained comparisons diagnostic (like what I pasted above). Nowadays that is handled by a different dedicated codepath.
Detect missing `.` in method chain in `let` bindings and statements
On parse errors where an ident is found where one wasn't expected, see if the next elements might have been meant as method call or field access.
```
error: expected one of `.`, `;`, `?`, `else`, or an operator, found `map`
--> $DIR/missing-dot-on-statement-expression.rs:7:29
|
LL | let _ = [1, 2, 3].iter()map(|x| x);
| ^^^ expected one of `.`, `;`, `?`, `else`, or an operator
|
help: you might have meant to write a method call
|
LL | let _ = [1, 2, 3].iter().map(|x| x);
| +
```
The rules were baffling when I ran in to them trying to add some impls,
so I made the compiler explain them to me.
The logic of the successful cases is unchanged, but I did rearrange it
to reverse the order of the primitive and `Adt` cases; this makes
producing the errors easier.
On parse errors where an ident is found where one wasn't expected, see if the next elements might have been meant as method call or field access.
```
error: expected one of `.`, `;`, `?`, `else`, or an operator, found `map`
--> $DIR/missing-dot-on-statement-expression.rs:7:29
|
LL | let _ = [1, 2, 3].iter()map(|x| x);
| ^^^ expected one of `.`, `;`, `?`, `else`, or an operator
|
help: you might have meant to write a method call
|
LL | let _ = [1, 2, 3].iter().map(|x| x);
| +
```
Also lint on option of function pointer comparisons
This PR is the first part of #134536, ie. the linting on `Option<{fn ptr}>` in the `unpredictable_function_pointer_comparisons` lint, which isn't part of the lang nomination that the second part is going trough, and so should be able to be approved independently.
Related to https://github.com/rust-lang/rust/issues/134527
r? `@compiler-errors`
Restrict `#[non_exaustive]` on structs with default field values
Do not allow users to apply `#[non_exaustive]` to a struct when they have also used default field values.
Arbitrary self types v2: no deshadow pre feature.
The arbitrary self types v2 work introduces a check for shadowed methods, whereby a method in some "outer" smart pointer type may called in preference to a method in the inner referent. This is bad if the outer pointer adds a method later, as it may change behavior, so we ensure we error in this circumstance.
It was intended that this new shadowing detection system only comes into play for users who enable the `arbitrary_self_types` feature (or of course everyone later if it's stabilized). It was believed that the new deshadowing code couldn't be reached without building the custom smart pointers that `arbitrary_self_types` enables, and therefore there was no risk of this code impacting existing users.
However, it turns out that cunning use of `Pin::get_ref` can cause this type of shadowing error to be emitted now. This commit adds a test for this case.
As we want this test to pass without arbitrary_self_types, but fail with it, I've split it into two files (one with run-pass and one without). If there's a better way I can amend it.
Part of #44874
r? ```@wesleywiser```
Precedence improvements: closures and jumps
This PR fixes some cases where rustc's pretty printers would redundantly parenthesize expressions that didn't need it.
<table>
<tr><th>Before</th><th>After</th></tr>
<tr><td><code>return (|x: i32| x)</code></td><td><code>return |x: i32| x</code></td></tr>
<tr><td><code>(|| -> &mut () { std::process::abort() }).clone()</code></td><td><code>|| -> &mut () { std::process::abort() }.clone()</code></td></tr>
<tr><td><code>(continue) + 1</code></td><td><code>continue + 1</code></td></tr>
</table>
Tested by `echo "fn main() { let _ = $AFTER; }" | rustc -Zunpretty=expanded /dev/stdin`.
The pretty-printer aims to render the syntax tree as it actually exists in rustc, as faithfully as possible, in Rust syntax. It can insert parentheses where forced by Rust's grammar in order to preserve the meaning of a macro-generated syntax tree, for example in the case of `a * $rhs` where $rhs is `b + c`. But for any expression parsed from source code, without a macro involved, there should never be a reason for inserting additional parentheses not present in the original.
For closures and jumps (return, break, continue, yield, do yeet, become) the unneeded parentheses came from the precedence of some of these expressions being misidentified. In the same order as the table above:
- Jumps and closures are supposed to have equal precedence. The [Rust Reference](https://doc.rust-lang.org/1.83.0/reference/expressions.html#expression-precedence) says so, and in Syn they do. There is no Rust syntax that would require making a precedence distinction between jumps and closures. But in rustc these were previously 2 distinct levels with the closure being lower, hence the parentheses around a closure inside a jump (but not a jump inside a closure).
- When a closure is written with an explicit return type, the grammar [requires](https://doc.rust-lang.org/1.83.0/reference/expressions/closure-expr.html) that the closure body consists of exactly one block expression, not any other arbitrary expression as usual for closures. Parsing of the closure body does not continue after the block expression. So in `|| { 0 }.clone()` the clone is inside the closure body and applies to `{ 0 }`, whereas in `|| -> _ { 0 }.clone()` the clone is outside and applies to the closure as a whole.
- Continue never needs parentheses. It was previously marked as having the lowest possible precedence but it should have been the highest, next to paths and loops and function calls, not next to jumps.
unimplement `PointerLike` for trait objects
Values of type `dyn* PointerLike` or `dyn PointerLike` are not pointer-like so these types should not implement `PointerLike`.
After https://github.com/rust-lang/rust/pull/133226, `PointerLike` allows user implementations, so we can't just mark it with `#[rustc_deny_explicit_impl(implement_via_object = false)]`. Instead, this PR splits the `#[rustc_deny_explicit_impl(implement_via_object = ...)]` attribute into two separate attributes `#[rustc_deny_explicit_impl]` and `#[rustc_do_not_implement_via_object]` so that we opt out of the automatic `impl PointerLike for dyn PointerLike` and still allow user implementations.
For traits that are marked with `#[do_not_implement_via_object]` but not `#[rustc_deny_explicit_impl]` I've also made it possible to add a manual `impl Trait for dyn Trait`. There is no immediate need for this, but it was one line to implement and seems nice to have.
fixes https://github.com/rust-lang/rust/issues/134545
fixes https://github.com/rust-lang/rust/issues/134543
r? `@compiler-errors`
Remove some dead code around import library generation
This was missed when replacing the usage of LLVM for generating import libraries.
``@bors`` rollup
Reduce the amount of explicit FatalError.raise()
Instead use dcx.abort_if_error() or guar.raise_fatal() instead. These guarantee that an error actually happened previously and thus we don't silently abort.
This commit splits the `#[rustc_deny_explicit_impl(implement_via_object = ...)]` attribute
into two attributes `#[rustc_deny_explicit_impl]` and `#[rustc_do_not_implement_via_object]`.
This allows us to have special traits that can have user-defined impls but do not have the
automatic trait impl for trait objects (`impl Trait for dyn Trait`).
Instead use dcx.abort_if_error() or guar.raise_fatal() instead. These
guarantee that an error actually happened previously and thus we don't
silently abort.
coverage: Rename `basic_coverage_blocks` to just `graph`
During coverage instrumentation, this variable always holds the current function's coverage graph, which is a simplified view of its MIR control-flow graph. The new name is clearer in context, and also shorter.
---
This is purely a rename, so there is no functional change.
Improve dependency_format a bit
* Make `DependencyList` an `IndexVec` rather than emulating one using a `Vec` (which was off-by-one as LOCAL_CRATE was intentionally skipped)
* Update some comments for the fact that we now use `#[global_allocator]` rather than `extern crate alloc_system;`/`extern crate alloc_jemalloc;` for specifying which allocator to use. We still use a similar mechanism for the panic runtime, so refer to the panic runtime in those comments instead.
* An unrelated refactor to `create_and_enter_global_ctxt` I forgot to include in https://github.com/rust-lang/rust/pull/134302. This refactor is too small to be worth it's own PR.
Fix logical error with what text is considered whitespace.
There appears to be a logical issue around what counts as leading white-space. There is code which does a subtraction assuming that no errors will be reported inside the leading whitespace. However we compute the length of that whitespace with std::char::is_whitespace and not rustc_lexer::is_whitespace. The former will include a no-break space while later will excluded it. We can only safely make the assumption that no errors will be reported in whitespace if it is all "Rust Standard" whitespace. Indeed an error does occur in unicode whitespace if it contains a no-break space. In that case the subtraction will cause a ICE (for a compiler in debug mode) as described in https://github.com/rust-lang/rust/issues/132918.
The arbitrary self types v2 work introduces a check for shadowed
methods, whereby a method in some "outer" smart pointer type may called
in preference to a method in the inner referent. This is bad if the
outer pointer adds a method later, as it may change behavior, so we
ensure we error in this circumstance.
It was intended that this new shadowing detection system only comes into
play for users who enable the `arbitrary_self_types` feature (or of
course everyone later if it's stabilized). It was believed that the
new deshadowing code couldn't be reached without building the custom
smart pointers that `arbitrary_self_types` enables, and therefore there
was no risk of this code impacting existing users.
However, it turns out that cunning use of `Pin::get_ref` can cause
this type of shadowing error to be emitted now. This commit adds a test
for this case.
During coverage instrumentation, this variable always holds the coverage graph,
which is a simplified view of the MIR control-flow graph. The new name is
clearer in context, and also shorter.
Use `MixedBitSet`s in const qualif
These analyses' domains should be very homogeneous, having compressed bitmaps on huge cfgs should make a difference (and doesn’t have an impact on the smaller / regular cfgs in our benchmarks).
This is a >40% walltime reduction on [this stress test](https://github.com/Manishearth/icu4x_compile_sample) extracted from a real world ICU case, and a 10x or so max-rss reduction.
cc `@oli-obk` `@RalfJung`
Should help with (or fix) issue #134404.
Speed up `Parser::expected_tokens`
The constant pushing/clearing of `Parser::expected_tokens` during parsing is slow. This PR speeds it up greatly.
r? `@estebank`
Make sure we handle `backwards_incompatible_lint` drops appropriately in drop elaboration
In #131326, a new kind of scheduled drop (`drop_kind: DropKind::Value` + `backwards_incompatible_lint: true`) was added so that we could insert a new kind of no-op MIR statement (`backward incompatible drop`) for linting purposes.
These drops were intended to have *no side-effects*, but drop elaboration code forgot to handle these drops specially and they were handled otherwise as normal drops in most of the code. This ends up being **unsound** since we insert more than one drop call for some values, which means that `Drop::drop` could be called more than once.
This PR fixes this by splitting out the `DropKind::ForLint` and adjusting the code. I'm not totally certain if all of the places I've adjusted are either reachable or correct, but I'm pretty certain that it's *more* correct than it was previously.
cc `@dingxiangfei2009`
r? nikomatsakis
Fixes#134482
Remove a duplicated check that doesn't do anything anymore.
fixes#134005
This code didn't actually `lub` the type of the previous expressions, but just the current type over and over again. Changing it to using the actual expression type does not change anything either, so may as well remove the entire loop.
coverage: Store coverage source regions as `Span` until codegen (take 2)
This is an attempt to re-land #133418:
> Historically, coverage spans were converted into line/column coordinates during the MIR instrumentation pass.
> This PR moves that conversion step into codegen, so that coverage spans spend most of their time stored as Span instead.
> In addition to being conceptually nicer, this also reduces the size of coverage mappings in MIR, because Span is smaller than 4x u32.
That PR was reverted by #133608, because in some circumstances not covered by our test suite we were emitting coverage metadata that was causing `llvm-cov` to exit with an error (#133606).
---
The implementation here is *mostly* the same, but adapted for subsequent changes in the relevant code (e.g. #134163).
I believe that the changes in #134163 should be sufficient to prevent the problem that required the original PR to be reverted. But I haven't been able to reproduce the original breakage in a regression test, and the `llvm-cov` error message is extremely unhelpful, so I can't completely rule out the possibility of this breaking again.
r? jieyouxu (reviewer of the original PR)
Some destructor/drop related tweaks
Two random tweaks I got from investigating some stuff around drops in edition 2024:
1. Use the `TypingEnv` of the mir builder, rather than making it over again.
2. Rename the `id` field from `Scope` to `local_id`, to reflect that it's a local id, and remove the `item_local_id()` accessor which just returned the id field.
Forbid overwriting types in typeck
While trying to figure out some type setting logic in https://github.com/rust-lang/rust/pull/134248 I realized that we sometimes set a type twice. While hopefully that would have been the same type, we didn't ensure that at all and just silently accepted it. So now we reject setting it twice, unless errors are happening, then we don't care.
Best reviewed commit by commit.
No behaviour change is intended.
Variants::Single: do not use invalid VariantIdx for uninhabited enums
~~Stacked on top of https://github.com/rust-lang/rust/pull/133681, only the last commit is new.~~
Currently, `Variants::Single` for an empty enum contains a `VariantIdx` of 0; looking that up in the enum variant list will ICE. That's quite confusing. So let's fix that by adding a new `Variants::Empty` case for types that have 0 variants.
try-job: i686-msvc
cleanup region handling: add `LateParamRegionKind`
The second commit is to enable a split between `BoundRegionKind` and `LateParamRegionKind`, by avoiding `BoundRegionKind` where it isn't necessary.
The third comment then adds `LateParamRegionKind` to avoid having the same late-param region for separate bound regions. This fixes#124021.
r? `@compiler-errors`
In codegen, a used function with `FunctionCoverageInfo` but no mappings has
historically indicated a bug. However, that will no longer be the case after
moving some fallible span-processing steps into codegen.
Currently it relies on having the right integer for every variant, and
if you add a variant you need to adjust the integers for all subsequent
variants, which is a pain.
This commit introduces a match guard formulation that takes advantage of
the enum-to-integer conversion to avoid specifying the integer for each
variant. And it does this via a macro to avoid lots of boilerplate.
The parser pushes a `TokenType` to `Parser::expected_token_types` on
every call to the various `check`/`eat` methods, and clears it on every
call to `bump`. Some of those `TokenType` values are full tokens that
require cloning and dropping. This is a *lot* of work for something
that is only used in error messages and it accounts for a significant
fraction of parsing execution time.
This commit overhauls `TokenType` so that `Parser::expected_token_types`
can be implemented as a bitset. This requires changing `TokenType` to a
C-style parameterless enum, and adding `TokenTypeSet` which uses a
`u128` for the bits. (The new `TokenType` has 105 variants.)
The new types `ExpTokenPair` and `ExpKeywordPair` are now arguments to
the `check`/`eat` methods. This is for maximum speed. The elements in
the pairs are always statically known; e.g. a
`token::BinOp(token::Star)` is always paired with a `TokenType::Star`.
So we now compute `TokenType`s in advance and pass them in to
`check`/`eat` rather than the current approach of constructing them on
insertion into `expected_token_types`.
Values of these pair types can be produced by the new `exp!` macro,
which is used at every `check`/`eat` call site. The macro is for
convenience, allowing any pair to be generated from a single identifier.
The ident/keyword filtering in `expected_one_of_not_found` is no longer
necessary. It was there to account for some sloppiness in
`TokenKind`/`TokenType` comparisons.
The existing `TokenType` is moved to a new file `token_type.rs`, and all
its new infrastructure is added to that file. There is more boilerplate
code than I would like, but I can't see how to make it shorter.
This is a naming convention used in a handful of spots in the parser for
delimiters. It confused me when I first saw it a long time ago, and I've
never liked it. A web search says "Bra-ket notation" exists in linear
algebra but the terminology has zero prior use in a programming context,
as far as I can tell.
This commit changes it to `open`/`close`, which is consistent with the
rest of the compiler.
The most significant is `check_keyword`: it now only pushes to
`expected_token_types` if the keyword check fails, which matches how all
the other `check` methods work.
The remainder are just tweaks to make these methods more consistent with
each other.
Point at lint name instead of whole attr for gated lints
```
warning: unknown lint: `test_unstable_lint`
--> $DIR/warn-unknown-unstable-lint-inline.rs:4:10
|
LL | #![allow(test_unstable_lint, another_unstable_lint)]
| ^^^^^^^^^^^^^^^^^^
|
= note: the `test_unstable_lint` lint is unstable
= help: add `#![feature(test_unstable_lint)]` to the crate attributes to enable
= note: this compiler was built on YYYY-MM-DD; consider upgrading it if it is out of date
note: the lint level is defined here
--> $DIR/warn-unknown-unstable-lint-inline.rs:3:9
|
LL | #![warn(unknown_lints)]
| ^^^^^^^^^^^^^
warning: unknown lint: `test_unstable_lint`
--> $DIR/warn-unknown-unstable-lint-inline.rs:4:29
|
LL | #![allow(test_unstable_lint, another_unstable_lint)]
| ^^^^^^^^^^^^^^^^^^^^^
|
= note: the `another_unstable_lint` lint is unstable
= help: add `#![feature(another_unstable_lint)]` to the crate attributes to enable
= note: this compiler was built on YYYY-MM-DD; consider upgrading it if it is out of date
```
This is particularly relevant when there are multiple lints in the same `warn` attribute. Pointing at the smaller span makes it clearer which one the warning is complaining about.
-Znext-solver: modify candidate preference rules
This implements the design proposed in the FCP in #132325 and matches the old solver behavior. I hope the inline comments are all sufficiently clear, I personally think this is a fairly clear improvement over the existing approach using `fn discard_impls_shadowed_by_env`. This fixes https://github.com/rust-lang/trait-system-refactor-initiative/issues/96.
This also fixes#133639 which encounters an ICE in negative coherence when evaluating the where-clause. Given the features required to trigger this ICE 🤷
r? ``@compiler-errors``
Stabilize `#[diagnostic::do_not_recommend]`
This PR seeks to stabilize the `#[diagnostic::do_not_recommend]`attribute.
This attribute was first proposed as `#[do_not_recommend`] attribute in RFC 2397 (https://github.com/rust-lang/rfcs/pull/2397). It gives the crate authors the ability to not suggest to the compiler to not show certain traits in its error messages.
With the presence of the `#[diagnostic]` tool attribute namespace it was decided to move the attribute there, as that lowers the amount of guarantees the compiler needs to give about the exact way this influences error messages. It turns the attribute into a hint which can be ignored. In addition to the original proposed functionality this attribute now also hides the marked trait in help messages ("This trait is implemented by: ").
The attribute does not accept any argument and can only be placed on trait implementations. If it is placed somewhere else a lint warning is emitted and the attribute is otherwise ignored. If an argument is detected a lint warning is emitted and the argument is ignored. This follows the rules outlined by the diagnostic namespace.
This attribute allows crates like diesel to improve their error messages drastically. The most common example here is the following error message:
```
error[E0277]: the trait bound `&str: Expression` is not satisfied
--> /home/weiznich/Documents/rust/rust/tests/ui/diagnostic_namespace/do_not_recommend.rs:53:15
|
LL | SelectInt.check("bar");
| ^^^^^ the trait `Expression` is not implemented for `&str`, which is required by `&str: AsExpression<Integer>`
|
= help: the following other types implement trait `Expression`:
Bound<T>
SelectInt
note: required for `&str` to implement `AsExpression<Integer>`
--> /home/weiznich/Documents/rust/rust/tests/ui/diagnostic_namespace/do_not_recommend.rs:26:13
|
LL | impl<T, ST> AsExpression<ST> for T
| ^^^^^^^^^^^^^^^^ ^
LL | where
LL | T: Expression<SqlType = ST>,
| ------------------------ unsatisfied trait bound introduced here
```
By applying the new attribute to the wild card trait implementation of
`AsExpression` for `T: Expression` the error message becomes:
```
error[E0277]: the trait bound `&str: AsExpression<Integer>` is not satisfied
--> $DIR/as_expression.rs:55:15
|
LL | SelectInt.check("bar");
| ^^^^^ the trait `AsExpression<Integer>` is not implemented for `&str`
|
= help: the trait `AsExpression<Text>` is implemented for `&str`
= help: for that trait implementation, expected `Text`, found `Integer`
```
which makes it much easier for users to understand that they are facing a type mismatch.
Other explored example usages include:
* This standard library error message: https://github.com/rust-lang/rust/pull/128008
* That bevy derived example:
e1f3068995/tests/ui/diagnostic_namespace/do_not_recommend/supress_suggestions_in_help.rs (No
more tuple pyramids)
Fixes#51992
r? ``@compiler-errors``
This PR also adds a few more tests, makes sure that all the tests are run for the old and new trait solver and adds a check that the attribute does not contain arguments.
```
warning: unknown lint: `test_unstable_lint`
--> $DIR/warn-unknown-unstable-lint-inline.rs:4:10
|
LL | #![allow(test_unstable_lint, another_unstable_lint)]
| ^^^^^^^^^^^^^^^^^^
|
= note: the `test_unstable_lint` lint is unstable
= help: add `#![feature(test_unstable_lint)]` to the crate attributes to enable
= note: this compiler was built on YYYY-MM-DD; consider upgrading it if it is out of date
note: the lint level is defined here
--> $DIR/warn-unknown-unstable-lint-inline.rs:3:9
|
LL | #![warn(unknown_lints)]
| ^^^^^^^^^^^^^
warning: unknown lint: `test_unstable_lint`
--> $DIR/warn-unknown-unstable-lint-inline.rs:4:29
|
LL | #![allow(test_unstable_lint, another_unstable_lint)]
| ^^^^^^^^^^^^^^^^^^^^^
|
= note: the `test_unstable_lint` lint is unstable
= help: add `#![feature(test_unstable_lint)]` to the crate attributes to enable
= note: this compiler was built on YYYY-MM-DD; consider upgrading it if it is out of date
note: the lint level is defined here
--> $DIR/warn-unknown-unstable-lint-inline.rs:3:9
|
LL | #![warn(unknown_lints)]
| ^^^^^^^^^^^^^
```
This is particularly relevant when there are multiple lints in the same `warn` attribute. Pointing at the smaller span makes it clearer which one the warning is complaining about.
Use field init shorthand where possible
Field init shorthand allows writing initializers like `tcx: tcx` as
`tcx`. The compiler already uses it extensively. Fix the last few places
where it isn't yet used.
EDIT: this PR also updates `rustfmt.toml` to set
`use_field_init_shorthand = true`.
Merge some patterns together
just something I noticed while browsing code. No change in functionality, deduplicates the 100% equal match arms by creating one big or pattern
Do not do if ! else, use unnegated cond and swap the branches instead
I'm tidying up my ergonomic ref counting PR and I'm going to make some small, simple and unrelated changes outside that PR, so the main PR sticks more straight to the point.
Clarify the match ergonomics 2024 migration lint's output
This makes a few changes:
- Rather than using the whole pattern as a span for the lint, this collects spans for each problematic default binding mode reset and labels them with why they're problems.
- The lint's suggestions are now verbose-styled, so that it's clear what's being suggested vs. what's problematic.
- The wording is now less technical, and the hard error version of this diagnostic now links to the same reference material as the lint (currently an unwritten page of the edition guide).
I'm not totally confident in the wording or formatting, so I'd appreciate feedback on that in particular. I tried to draw a connection with word choice between the labels and the suggestion, but it might be imprecise, unclear, or cluttered. If so, it might be worth making the labels more terse and adding notes that explain them, but that's harder to read in a way too.
cc ```@Nadrieril``` ```@Jules-Bertholet```
Closes#133854. For reference, the error from that issue becomes:
```
error: pattern uses features incompatible with edition 2024
--> $DIR/remove-me.rs:6:25
|
LL | map.iter().filter(|(&(_x, _y), &_c)| false);
| ^ ^ cannot implicitly match against multiple layers of reference
| |
| cannot implicitly match against multiple layers of reference
|
help: make the implied reference pattern explicit
|
LL | map.iter().filter(|&(&(_x, _y), &_c)| false);
| +
```
Overhaul keyword handling
The compiler's list of keywords has some problems.
- It contains several items that aren't keywords.
- The order isn't quite right in a couple of places.
- Some of the names of predicates relating to keywords are confusing.
- rustdoc and rustfmt have their own (incorrect) versions of the keyword list.
- `AllKeywords` is unnecessarily complex.
r? ```@jieyouxu```
`gen` is an edition-specific keyword used in unstable Rust, and so
belongs with `try` (as `is_unused_keyword_conditional` indicates).
Also, the cases in `is_unused_keyword_conditional` should be in
alphabetical order, to match the keyword list.
These changes don't affect the behaviour of any of the `Symbol::is_*`
functions.
`rustc_symbol` is the source of truth for keywords.
rustdoc has its own implicit definition of keywords, via the
`is_doc_keyword`. It (presumably) intends to include all keywords, but
it omits `yeet`.
rustfmt has its own explicit list of Rust keywords. It also (presumably)
intends to include all keywords, but it omits `await`, `builtin`, `gen`,
`macro_rules`, `raw`, `reuse`, `safe`, and `yeet`. Also, it does linear
searches through this list, which is inefficient.
This commit fixes all of the above problems by introducing a new
predicate `is_any_keyword` in rustc and using it in rustdoc and rustfmt.
It documents that it's not the right predicate in most cases.
This commit seeks to stabilize the `#[diagnostic::do_not_recommend]`
attribute.
This attribute was first proposed as `#[do_not_recommend`] attribute in
RFC 2397 (https://github.com/rust-lang/rfcs/pull/2397). It gives the
crate authors the ability to not suggest to the compiler to not show
certain traits in it's error messages. With the presence of the
`#[diagnostic]` tool attribute namespace it was decided to move the
attribute there, as that lowers the amount of guarantees the compiler
needs to give about the exact way this influences error messages. It
turns the attribute into a hint which can be ignored. In addition to the
original proposed functionality this attribute now also hides the marked
trait in help messages ("This trait is implemented by: ").
The attribute does not accept any argument and can only be placed on
trait implementations. If it is placed somewhere else a lint warning is
emitted and the attribute is otherwise ignored. If an argument is
detected a lint warning is emitted and the argument is ignored. This
follows the rules outlined by the diagnostic namespace.
This attribute allows crates like diesel to improve their error messages
drastically. The most common example here is the following error
message:
```
error[E0277]: the trait bound `&str: Expression` is not satisfied
--> /home/weiznich/Documents/rust/rust/tests/ui/diagnostic_namespace/do_not_recommend.rs:53:15
|
LL | SelectInt.check("bar");
| ^^^^^ the trait `Expression` is not implemented for `&str`, which is required by `&str: AsExpression<Integer>`
|
= help: the following other types implement trait `Expression`:
Bound<T>
SelectInt
note: required for `&str` to implement `AsExpression<Integer>`
--> /home/weiznich/Documents/rust/rust/tests/ui/diagnostic_namespace/do_not_recommend.rs:26:13
|
LL | impl<T, ST> AsExpression<ST> for T
| ^^^^^^^^^^^^^^^^ ^
LL | where
LL | T: Expression<SqlType = ST>,
| ------------------------ unsatisfied trait bound introduced here
```
By applying the new attribute to the wild card trait implementation of
`AsExpression` for `T: Expression` the error message becomes:
```
error[E0277]: the trait bound `&str: AsExpression<Integer>` is not satisfied
--> $DIR/as_expression.rs:55:15
|
LL | SelectInt.check("bar");
| ^^^^^ the trait `AsExpression<Integer>` is not implemented for `&str`
|
= help: the trait `AsExpression<Text>` is implemented for `&str`
= help: for that trait implementation, expected `Text`, found `Integer`
```
which makes it much easier for users to understand that they are facing
a type mismatch.
Other explored example usages included
* This standard library error message: https://github.com/rust-lang/rust/pull/128008
* That bevy derived example:
e1f3068995/tests/ui/diagnostic_namespace/do_not_recommend/supress_suggestions_in_help.rs (No
more tuple pyramids)
Fixes#51992
Re-export more `rustc_span::symbol` things from `rustc_span`.
`rustc_span::symbol` defines some things that are re-exported from `rustc_span`, such as `Symbol` and `sym`. But it doesn't re-export some closely related things such as `Ident` and `kw`. So you can do `use rustc_span::{Symbol, sym}` but you have to do `use rustc_span::symbol::{Ident, kw}`, which is inconsistent for no good reason.
This commit re-exports `Ident`, `kw`, and `MacroRulesNormalizedIdent`, and changes many `rustc_span::symbol::` qualifiers to `rustc_span::`. This is a 300+ net line of code reduction, mostly because many files with two `use rustc_span` items can be reduced to one.
r? `@jieyouxu`
`rustc_span::symbol` defines some things that are re-exported from
`rustc_span`, such as `Symbol` and `sym`. But it doesn't re-export some
closely related things such as `Ident` and `kw`. So you can do `use
rustc_span::{Symbol, sym}` but you have to do `use
rustc_span::symbol::{Ident, kw}`, which is inconsistent for no good
reason.
This commit re-exports `Ident`, `kw`, and `MacroRulesNormalizedIdent`,
and changes many `rustc_span::symbol::` qualifiers in `compiler/` to
`rustc_span::`. This is a 200+ net line of code reduction, mostly
because many files with two `use rustc_span` items can be reduced to
one.
- Move it to `rustc_parse`, which is the only crate that uses it. This
lets us remove all the `pub` markers from it.
- Change `next_ref` and `look_ahead` to `get` and `bump`, which work
better for the `rustc_parse` uses.
- This requires adding a `TokenStream::get` method, which is simple.
- In `TokenCursor`, we currently duplicate the
`DelimSpan`/`DelimSpacing`/`Delimiter` from the surrounding
`TokenTree::Delimited` in the stack. This isn't necessary so long as
we don't prematurely move past the `Delimited`, and is a small perf
win on a very hot code path.
- In `parse_token_tree`, we clone the relevant `TokenTree::Delimited`
instead of constructing an identical one from pieces.
Rollup of 7 pull requests
Successful merges:
- #133265 (Add a range argument to vec.extract_if)
- #133801 (Promote powerpc64le-unknown-linux-musl to tier 2 with host tools)
- #134323 (coverage: Dismantle `map_data.rs` by moving its responsibilities elsewhere)
- #134378 (An octuple of polonius fact generation cleanups)
- #134408 (Regression test for RPIT inheriting lifetime from projection)
- #134423 (bootstrap: use specific-purpose ui test path for `test_valid` self-test)
- #134426 (Fix typo in uint_macros.rs)
Failed merges:
- #133103 (Pass FnAbi to find_mir_or_eval_fn)
r? `@ghost`
`@rustbot` modify labels: rollup
Currently there are two ways to peek at a `TokenStreamIter`.
- Wrap it in a `Peekable` and use that traits `peek` method.
- Use `TokenStreamIter`'s inherent `peek` method.
Some code uses one, some use the other. This commit converts all places
to the inherent method. This eliminates mixing of `TokenStreamIter` and
`Peekable<TokenStreamIter>` and some use of `impl Iterator` and `dyn
Iterator`.
Because `TokenStreamIter` is a much better name for a `TokenStream`
iterator. Also rename the `TokenStream::trees` method as
`TokenStream::iter`, and some local variables.
Field init shorthand allows writing initializers like `tcx: tcx` as
`tcx`. The compiler already uses it extensively. Fix the last few places
where it isn't yet used.
An octuple of polonius fact generation cleanups
This PR is extracted from https://github.com/rust-lang/rust/pull/134268 for easier review and contains its first 8 commits. They have already been reviewed by ````@jackh726```` over there.
r? ````@jackh726````
coverage: Dismantle `map_data.rs` by moving its responsibilities elsewhere
This is a series of incremental changes that combine to let us get rid of `coverageinfo/map_data.rs`, by moving all of its responsibilities into more appropriate places.
Some of the notable consequences are:
- We once again build the per-CGU file table on the fly while preparing individual covfun records, instead of building the whole table up-front. The up-front approach was introduced by #117042 to work around various other problems in generating the covmap/covfun records, but subsequent cleanups have made that approach no longer necessary.
- Expression conversion and mapping-region conversion are now performed directly in `mapgen::covfun`, which should make future changes easier.
- We no longer insert unused function instances into the same map that is also used to track used function instances. This helps to decouple the handling of used vs unused functions.
---
There should be no meaningful change to compiler output. The file table is no longer sorted, because reordering it would invalidate the file indices stored in individual covfun records, but the table order should still be deterministic (albeit arbitrary).
There are some subsequent cleanups that I intend to investigate, but this is enough change for one PR.
Promote powerpc64le-unknown-linux-musl to tier 2 with host tools
MCP: https://github.com/rust-lang/compiler-team/issues/803
I'm using crosstool-ng for building a toolchain because GCC 9 from `musl-toolchain.sh` has float ABI issues (?) and can't compile LLVM, and writing a crosstool-ng config for a target feels less hacky than yet another target specific shell script. I also defined a kernel version, since there wasn't one specified before. If a lower version is desired, just let me know. I also tried to match the rust configure args with the loongarch64 musl tier 2 target.
The resulting compiler works fine, built with `DEPLOY=1 ./src/ci/docker/run.sh dist-powerpc64le-linux` and tested on Alpine Linux in a VM and on a bare metal POWER8 machine:
```
qemu-ppc64le:/tmp/rust-nightly-powerpc64le-unknown-linux-musl$ ash install.sh
install: creating uninstall script at /usr/local/lib/rustlib/uninstall.sh
install: installing component 'rustc'
install: installing component 'rust-std-powerpc64le-unknown-linux-musl'
install: installing component 'cargo'
install: installing component 'rustfmt-preview'
install: installing component 'rls-preview'
install: installing component 'rust-analyzer-preview'
install: installing component 'llvm-tools-preview'
install: installing component 'clippy-preview'
install: installing component 'miri-preview'
install: installing component 'rust-analysis-powerpc64le-unknown-linux-musl'
install: installing component 'llvm-bitcode-linker-preview'
install: WARNING: failed to run ldconfig. this may happen when not installing as root. run with --verbose to see the error
rust installed.
qemu-ppc64le:~$ echo 'fn main() { println!("hello world"); }' > test.rs
qemu-ppc64le:~$ rustc test.rs
qemu-ppc64le:~$ ./test
hello world
qemu-ppc64le:~$ file test
test: ELF 64-bit LSB executable, 64-bit PowerPC or cisco 7500, OpenPOWER ELF V2 ABI, version 1 (SYSV), statically linked, BuildID[sha1]=596ee6abf9add487ebc54fb71c2076fb6faea013, with debug_info, not stripped
```
try-job: dist-powerpc64le-linux
Split up attribute parsing code and move data types to `rustc_attr_data_structures`
This change renames `rustc_attr` to `rustc_attr_parsing`, and splits up the parsing code. At the same time, all the data types used move to `rustc_attr_data_structures`. This is in preparation of also having a third crate: `rustc_attr_validation`
I initially envisioned this as two separate PRs, but I think doing it in one go reduces the number of ways others would have to rebase their changes on this. However, I can still split them.
r? `@oli-obk` (we already discussed how this is a first step in a larger plan)
For a more detailed plan on how attributes are going to change, see https://github.com/rust-lang/rust/issues/131229
Edit: this looks like a giant PR, but the changes are actually rather trivial. Each commit is reviewable on its own, and mostly moves code around. No new logic is added.
Fix some comments related to upvars handling
I'm tidying up my ergonomic ref counting PR and I'm going to make some small, simple and unrelated changes outside that PR, so the main PR sticks more straight to the point.
rustc_borrowck: Suggest changing `&raw const` to `&raw mut` if applicable
Closes#127562
For reference, here is the diff compared to the original error reported in that issue before #134244 stopped suggesting the invalid syntax:
```
diff --git a/tests/ui/borrowck/no-invalid-mut-suggestion-for-raw-pointer-issue-127562.stderr b/tests/ui/borrowck/no-invalid-mut-suggestion-for-raw-pointer-issue-127562.stderr
index 0da5d15cf7f..dbe834b6b78 100644
--- a/tests/ui/borrowck/no-invalid-mut-suggestion-for-raw-pointer-issue-127562.stderr
+++ b/tests/ui/borrowck/no-invalid-mut-suggestion-for-raw-pointer-issue-127562.stderr
``@@`` -6,8 +6,8 ``@@`` LL | unsafe { *ptr = 3; }
|
help: consider changing this to be a mutable pointer
|
-LL | let ptr = &mut raw const val;
- | +++
+LL | let ptr = &raw mut val;
+ | ~~~
error: aborting due to 1 previous error
```
Use links to edition guide for edition migrations
This switches the migration lints for the 2024 edition to point to the edition guide documentation instead of the tracking issues. I expect the documentation should be easier to understand for a user, compared to most of the issues which don't have any direct information, and can be a bit confusing to navigate, or have outdated information.
Rename `rustc_mir_build::build` to `builder`
GitHub's “Go to file” feature silently ignores all files in this module, presumably because they are in a directory named `build`, which is mistaken for a build-output directory. That makes it meaningfully harder to view those files and their history via GitHub.
This PR sidesteps that issue by renaming `build` to `builder`, which in context has basically the same meaning, but can't be mistaken for a build-output directory.
---
Extracted from #133404, after #133592 changed the .gitignore rule from `build/` to `/build`. The problem of GitHub ignoring these files still exists even after that change, which suggests that GitHub's behaviour is a hard-coded heuristic that isn't influenced by the repository's git settings.
Currently this PR doesn't include the tidy rule forbidding `build` as a module name, but that could be added if people think it's a good idea.
Handle fndef rendering together with signature rendering
Pulled out of https://github.com/rust-lang/rust/pull/134353
Changes some highlighting in type mismatch errors around fndefs
Remove queries from the driver interface
All uses of driver queries in the public api of rustc_driver have been removed in https://github.com/rust-lang/rust/pull/134130 already. This removes driver queries from rustc_interface and does a couple of cleanups around TyCtxt construction and entering enabled by this removal.
Finishes the removal of driver queries started with https://github.com/rust-lang/rust/pull/126834.
There is a logical issue around what counts as leading white-space.
There is code which does a subtraction assuming that no errors will be reported
inside the leading whitespace. However we compute the length of
that whitespace with std::char::is_whitespace and not
rustc_lexer::is_whitespace. The former will include a no-break space while
later will excluded it. We can only safely make the assumption that no errors
will be reported in whitespace if it is all "Rust Standard" whitespace.
Indeed an error does occur in unicode whitespace if it contains a no-break
space.
In #129533 the main hash function changed and the order of `-Z
input-stats` output changed, which showed that it is dependent on the
hash function, even though it is sorted. That's because entries with the
same cumulative size are ordered in a way that depends on the hash
function.
This commit fixes that by using the entry label as the secondary
ordering key.
This patch dismantles what was left of `FunctionCoverage` in `map_data.rs`,
replaces `function_coverage_map` with a set, and overhauls how we prepare
covfun records for unused functions.
`CheckAttrVisitor::check_doc_keyword` checks `#[doc(keyword = "..")]`
attributes to ensure they are on an empty module, and that the value is
a non-empty identifier.
The `rustc::existing_doc_keyword` lint checks these attributes to ensure
that the value is the name of a keyword.
It's silly to have two different checking mechanisms for these
attributes. This commit does the following.
- Changes `check_doc_keyword` to check that the value is the name of a
keyword (avoiding the need for the identifier check, which removes a
dependency on `rustc_lexer`).
- Removes the lint.
- Updates tests accordingly.
There is one hack: the `SelfTy` FIXME case used to used to be handled by
disabling the lint, but now is handled with a special case in
`is_doc_keyword`. That hack will go away if/when the FIXME is fixed.
Co-Authored-By: Guillaume Gomez <guillaume1.gomez@gmail.com>
The checks in `is_eligible_for_coverage` include `is_fn_like`, but will also
exclude various function-like things that cannot possibly have coverage
instrumentation.
Since the previos commit renamed `assignment_rhs_span` to just
`rhs_span` there is no need for a variable just to shorten the
expression on the next line. Inline the variable.
Some trait method vs impl method signature difference diagnostic cleanups
Just some things I noticed while debugging a weird diagnostic in https://github.com/rust-lang/rust/pull/134353
best reviewed commit by commit
Check for array lengths that aren't actually `usize`
I wish typeck wouldn't give us `ty::Array`s that have this problem in the first place, but we can check for it.
Fixes#134352
cc ``@matthiaskrgr``
reject unsound toggling of RISCV target features
~~Stacked on top of https://github.com/rust-lang/rust/pull/133417, only the last commit is new.~~
Works towards https://github.com/rust-lang/rust/issues/132618 (but more [remains to be done](https://github.com/rust-lang/rust/pull/134337#issuecomment-2544228958))
Part of https://github.com/rust-lang/rust/issues/116344
Cc ``@beetrees`` I hope I got everything. I didn't do anything about "The f and zfinx features are incompatible" and that's not an ABI thing (right?) and I am not sure how to handle it with these ABI checks.
r? ``@workingjubilee``
Ideally we'd also reject target specs that disable the `f` feature but set an ABI that requires `f`... but I don't want to duplicate this logic. I have some ideas for how maybe the entire float ABI check logic should be different, now that we have some examples of what these ABI checks look like, but that will be a future PR.
Keep track of patterns that could have introduced a binding, but didn't
When we recover from a pattern parse error, or a pattern uses `..`, we keep track of that and affect resolution error for missing bindings that could have been provided by that pattern. We differentiate between `..` and parse recovery. We silence resolution errors likely caused by the pattern parse error.
```
error[E0425]: cannot find value `title` in this scope
--> $DIR/struct-pattern-with-missing-fields-resolve-error.rs:18:30
|
LL | if let Website { url, .. } = website {
| ------------------- this pattern doesn't include `title`, which is available in `Website`
LL | println!("[{}]({})", title, url);
| ^^^^^ not found in this scope
```
Fix#74863.
rustc_mir_build: Clarify that 'mirrored' does not mean 'flipped' or 'reversed'
My intuition for 'mirrored' is that it means 'flipped' or 'reversed'. Clarify that that is not what is meant to 'mirror' the THIR from the HIR.
Update spelling of "referring"
I noticed that `referring` was spelled incorrectly in the output of `unexpected 'cfg' condition name` warnings; it looks like it was also incorrectly spelled in a doc comment. I've update both instances.
Fix `trimmed_def_paths` ICE in the function ptr comparison lint
This PR fixes an ICE with `trimmed_def_paths` ICE in the function ptr comparison lint, specifically when pretty-printing user types but then not using the resulting pretty-printing.
Fixes#134345
r? `@saethlin`
Make sure to use normalized ty for unevaluated const in default struct value
This cleans up the way that we construct the `mir::Const::Unevaluated` for default struct values. We were previously using `from_unevaluated`, which doesn't normalize the type, and is really only used for inline assembly. Other codepaths (such as `ExprKind::NamedConst`) use the type from the body.
Also, let's stop using `literal_operand`, which also is really not meant for calls other than for literal comparisons in pattern lowering.
Also move all of the tests to a separate subdirectory so they don't need to have the same prefix on all the test files.
Fixes#134298
r? estebank or reassign
This allows compile-time configuration based on this.
In the near future we should do this across all RISCV targets, probably,
but this cfg is essential for building software usable on these targets.
It is possible to avoid the clone as suggested in the comment. It would
require introducing an enum with two variants
`CloneBeforeModifying(&Domain)` and `Modifiable(&mut Domain)`. But it's
not worth the effort, because this code path just isn't very hot. E.g.
when compiling a large benchmark like `cargo-0.60.0` it's only hit a few
thousand times.
Switches to the idiom used elsewhere of calling `Analysis::bottom_value`
to initialize a `state` value outside a loop, and then using
`clone_from` to update it within the loop. This is simpler and has no
impact on performance.
Current `SwitchInt` handling has complicated control flow.
- The dataflow engine calls `Analysis::apply_switch_int_edge_effects`,
passing in an "applier" that impls `SwitchIntEdgeEffects`.
- `apply_switch_int_edge_effects` possibly calls `apply` on the applier,
passing it a closure.
- The `apply` method calls the closure on each `SwitchInt` edge.
- The closure operates on the edge.
I.e. control flow goes from the engine, to the analysis, to the applier
(which came from the engine), to the closure (which came from the
analysis). It took me a while to work this out.
This commit changes to a simpler structure that maintains the important
characteristics.
- The dataflow engine calls `Analysis::get_switch_int_data`.
- `get_switch_int_data` returns an `Option<Self::SwitchIntData>` value.
- If that returned value was `Some`, the dataflow engine calls
`Analysis::apply_switch_int_edge_effect` on each edge, passing the
`Self::SwitchIntData` value.
- `Analysis::apply_switch_int_edge_effect` operates on the edge.
I.e. control flow goes from the engine, to the analysis, to the
engine, to the analysis.
Added:
- The `Analysis::SwitchIntData` assoc type and the
`Analysis::get_switch_int_data` method. Both only need to be
defined by analyses that look at `SwitchInt` terminators.
- The `MaybePlacesSwitchIntData` struct, which has three fields.
Changes:
- `Analysis::apply_switch_int_edge_effects` becomes
`Analysis::apply_switch_int_edge_effect`, which is a little simpler
because it's dealing with a single edge instead of all edges.
Removed:
- The `SwitchIntEdgeEffects` trait, and its two impls:
`BackwardSwitchIntEdgeEffectsApplier` (which has six fields) and
`ForwardSwitchIntEdgeEffectsApplier` structs (which has four fields).
- The closure.
The new structure is more concise and simpler.
Hir attributes
This PR needs some explanation, it's somewhat large.
- This is step one as described in https://github.com/rust-lang/compiler-team/issues/796. I've added a new `hir::Attribute` which is a lowered version of `ast::Attribute`. Right now, this has few concrete effects, however every place that after this PR parses a `hir::Attribute` should later get a pre-parsed attribute as described in https://github.com/rust-lang/compiler-team/issues/796 and transitively https://github.com/rust-lang/rust/issues/131229.
- an extension trait `AttributeExt` is added, which is implemented for both `ast::Attribute` and `hir::Atribute`. This makes `hir::Attributes` mostly compatible with code that used to parse `ast::Attribute`. All its methods are also added as inherent methods to avoid having to import the trait everywhere in the compiler.
- Incremental can not not hash `ast::Attribute` at all.
Pass `TyCtxt` to early diagostics decoration
This PR pass a `TyCtxt` to the early diagnostics decoration code so that diagnostics code that take advantage of (a very limited but still useful) `TyCtxt` in their note, help, suggestions, ...
This is particulary useful for #133221 which wants to get the crate name of a `DefId`, which is possible with `tcx.crate_name(...)`.
I highly recommend reviewing this PR commit by commit.
r? `@jieyouxu`
reject aarch64 target feature toggling that would change the float ABI
~~Stacked on top of https://github.com/rust-lang/rust/pull/133099. Only the last two commits are new.~~
The first new commit lays the groundwork for separately controlling whether a feature may be enabled or disabled. The second commit uses that to make it illegal to *disable* the `neon` feature (which is only possible via `-Ctarget-feature`, and so the new check just adds a warning). Enabling the `neon` feature remains allowed on targets that don't disable `neon` or `fp-armv8`, which is all our built-in targets. This way, the entire PR is not a breaking change.
Fixes https://github.com/rust-lang/rust/issues/131058 for hardfloat targets (together with https://github.com/rust-lang/rust/pull/133102 which fixed it for softfloat targets).
Part of https://github.com/rust-lang/rust/issues/116344.
- integrate it within existing fact generation instead of being called
in typeck
- simplify access fact extraction
- also remove single use fact emit functions in root fact generation
Modifies the index instruction from `gep [0 x %Type]` to `gep %Type`
Fixes#133979.
This PR modifies the index instruction from `gep [0 x %Type]` to `gep %Type`, which is the same with pointer offset calculation.
This will help LLVM calculate various formats of GEP instructions. According to [[RFC] Replacing getelementptr with ptradd](https://discourse.llvm.org/t/rfc-replacing-getelementptr-with-ptradd/68699), we ultimately aim to canonicalize everything to `gep i8`. Based on the results from https://github.com/rust-lang/rust/pull/134117#issuecomment-2531717076, I think we still need to investigate some missing optimizations, so this PR is just a small step forward.
r? compiler
A couple of polonius fact generation cleanups
This PR is extracted from #134268 for easier review and contains its first two commits. They have already been reviewed by `@jackh726.`
r? `@jackh726`
Don't make a def id for `impl_trait_in_bindings`
The def collector is awkward, so for now just wrap let statements in a new `ImplTraitContext::InBinding` which tells `visit_ty` not to make a def id for the type. This will not generalize to other ITIB cases, like if we allow them in turbofishes (e.g. `foo::<impl Fn()>(|| {})`).
Fixes#134307
r? oli-obk
Add some convenience helper methods on `hir::Safety`
Makes a lot of call sites simpler and should make any refactorings needed for https://github.com/rust-lang/rust/pull/134090#issuecomment-2541332415 simpler, as fewer sites have to be touched in case we end up storing some information in the variants of `hir::Safety`
don't show the full linker args unless `--verbose` is passed
the linker arguments can be *very* long, especially for crates with many dependencies. often they are not useful. omit them unless the user specifically requests them.
split out from https://github.com/rust-lang/rust/pull/119286. fixes https://github.com/rust-lang/rust/issues/109979.
r? `@bjorn3`
try-build: i686-mingw
Remove support for specializing ToString outside the standard library
This is the only trait specializable outside of the standard library. Before stabilizing specialization we will probably want to remove support for this. It was originally made specializable to allow a more efficient ToString in libproc_macro back when this way the only way to get any data out of a TokenStream. We now support getting individual tokens, so proc macros no longer need to call it as often.
the linker arguments can be *very* long, especially for crates with many dependencies. some parts of them are not very useful. unless specifically requested:
- omit object files specific to the current invocation
- fold rlib files into a single braced argument (in shell expansion format)
this shortens the output significantly without removing too much information.
Rollup of 7 pull requests
Successful merges:
- #132939 (Suggest using deref in patterns)
- #133293 (Updates Solaris target information, adds Solaris maintainer)
- #133392 (Fix ICE when multiple supertrait substitutions need assoc but only one is provided)
- #133986 (Add documentation for anonymous pipe module)
- #134022 (Doc: Extend for tuples to be stabilized in 1.85.0)
- #134259 (Clean up `infer_return_ty_for_fn_sig`)
- #134264 (Arbitrary self types v2: Weak & NonNull diagnostics)
r? `@ghost`
`@rustbot` modify labels: rollup
Arbitrary self types v2: Weak & NonNull diagnostics
This builds on top of #134262 which is more urgent to review and merge first. I'll likely rebase this PR once that lands.
This is the first part of the diagnostic enhancements planned for Arbitrary Self Types v2.
Various types can be used as method receivers, such as `Rc<>`, `Box<>` and `Arc<>`. The arbitrary self types v2 work allows further types to be made method receivers by implementing the Receiver trait.
With that in mind, it may come as a surprise to people when certain common types do not implement Receiver and thus cannot be used as a method receiver.
The RFC for arbitrary self types v2 therefore proposes emitting specific
lint hints for these cases:
* `NonNull`
* `Weak`
* Raw pointers
The code already emits a hint for this third case, in that it advises folks that the `arbitrary_self_types_pointers` feature may meet their need. This PR adds diagnostic hints for the `Weak` and `NonNull` cases.
Tracking issue #44874
r? `@wesleywiser`
Clean up `infer_return_ty_for_fn_sig`
The code for lowering fn signatures from HIR currently is structured to prefer the recovery path (where users write `-> _`) over the good path. This PR pulls the recovery code out into a separate fn.
Review w/o whitespace
Fix ICE when multiple supertrait substitutions need assoc but only one is provided
Dyn traits must have all of their associated types constrained either by:
1. writing them in the dyn trait itself as an associated type bound, like `dyn Iterator<Item = u32>`,
2. A supertrait bound, like `trait ConstrainedIterator: Iterator<Item = u32> {}`, then you may write `dyn ConstrainedIterator` which doesn't need to mention `Item`.
However, the object type lowering code did not consider the fact that there may be multiple supertraits with different substitutions, so it just used the associated type's *def id* as a key for keeping track of which associated types are missing:
1fc691e6dd/compiler/rustc_hir_analysis/src/hir_ty_lowering/dyn_compatibility.rs (L131)
This means that we can have missing associated types when there are mutliple supertraits with different substitutions and only one of them is constrained, like:
```rust
trait Sup<T> {
type Assoc: Default;
}
impl<T: Default> Sup<T> for () {
type Assoc = T;
}
impl<T: Default, U: Default> Dyn<T, U> for () {}
trait Dyn<A, B>: Sup<A, Assoc = A> + Sup<B> {}
```
The above example allows you to name `<dyn Dyn<i32, u32> as Sup<u32>>::Assoc` even though it is not possible to project since it's neither constrained by a manually written projection bound or a supertrait bound. This successfully type-checks, but leads to a codegen ICE since we are not able to project the associated type.
This PR fixes the validation for checking that a dyn trait mentions all of its associated type bounds. This is theoretically a breaking change, since you could technically use that `dyn Dyn<A, B>` type mentionedin the example above without actually *projecting* to the bad associated type, but I don't expect it to ever be relevant to a user since it's almost certainly a bug. This is corroborated with the crater results[^crater], which show no failures[^unknown].
Crater: https://github.com/rust-lang/rust/pull/133392#issuecomment-2508769703Fixes#133388
[^crater]: I cratered this originally with #133397, which is a PR that is stacked on top, then re-ran crater with just the failures from that PR.
[^unknown]: If you look at the crater results, it shows all of the passes as "unknown". I believe this is a crater bug, since looking at the results manually shows them as passes.
Suggest using deref in patterns
Fixes#132784
This changes the following code:
```rs
use std::sync::Arc;
fn main() {
let mut x = Arc::new(Some(1));
match x {
Some(_) => {}
None => {}
}
}
```
to output
```rs
error[E0308]: mismatched types
--> src/main.rs:5:9
|
LL | match x {
| - this expression has type `Arc<Option<{integer}>>`
...
LL | Some(_) => {}
| ^^^^^^^ expected `Arc<Option<{integer}>>`, found `Option<_>`
|
= note: expected struct `Arc<Option<{integer}>>`
found enum `Option<_>`
help: consider dereferencing to access the inner value using the Deref trait
|
LL | match *x {
| ~~
```
instead of
```rs
error[E0308]: mismatched types
--> src/main.rs:5:9
|
4 | match x {
| - this expression has type `Arc<Option<{integer}>>`
5 | Some(_) => {}
| ^^^^^^^ expected `Arc<Option<{integer}>>`, found `Option<_>`
|
= note: expected struct `Arc<Option<{integer}>>`
found enum `Option<_>`
```
This makes it more obvious that a Deref is available, and gives a suggestion on how to use it in order to fix the issue at hand.
Bounds-check with PtrMetadata instead of Len in MIR
Rather than emitting `Len(*_n)` in array index bounds checks, emit `PtrMetadata(copy _n)` instead -- with some asterisks for arrays and `&mut` that need it to be done slightly differently.
We're getting pretty close to removing `Len` entirely, actually. I think just one more PR after this (for slice drop shims).
r? mir
Various types can be used as method receivers, such as Rc<>, Box<> and
Arc<>. The arbitrary self types v2 work allows further types to be made
method receivers by implementing the Receiver trait.
With that in mind, it may come as a surprise to people when certain
common types do not implement Receiver and thus cannot be used as a
method receiver.
The RFC for arbitrary self types v2 therefore proposes emitting specific
lint hints for these cases:
* NonNull
* Weak
* Raw pointers
The code already emits a hint for this third case, in that it advises
folks that the `arbitrary_self_types_pointers` feature may meet their
need. This PR adds diagnostic hints for the Weak and NonNull cases.
Rollup of 6 pull requests
Successful merges:
- #133221 (Add external macros specific diagnostics for check-cfg)
- #133386 (Update linux_musl base to dynamically link the crt by default)
- #134191 (Make some types and methods related to Polonius + Miri public)
- #134227 (Update wasi-sdk used to build WASI targets)
- #134279 ((Re-)return adjustment target if adjust kind is never-to-any)
- #134295 (Encode coroutine-closures in SMIR)
r? `@ghost`
`@rustbot` modify labels: rollup
- use consistent names
- inline single use functions
- dedupe and simplify some paths
- fix fact generation timer activity: it was missing the walk and
extraction process
(Re-)return adjustment target if adjust kind is never-to-any
This PR fixes#134162 where we ICE'd on
```rs
fn main() {
struct X;
let _ = [X] == [panic!(); 2];
}
```
In https://github.com/rust-lang/rust/pull/121208#discussion_r1494187622, there was a change
```diff
- if let Some(adjustments) = self.typeck_results.borrow().adjustments().get(expr.hir_id) {
- let reported = self.dcx().span_delayed_bug(
- expr.span,
- "expression with never type wound up being adjusted",
- );
- return if let [Adjustment { kind: Adjust::NeverToAny, target }] = &adjustments[..] {
- target.to_owned()
- } else {
- Ty::new_error(self.tcx(), reported)
- };
- }
+ if let Some(_) = self.typeck_results.borrow().adjustments().get(expr.hir_id) {
+ self.dcx()
+ .span_bug(expr.span, "expression with never type wound up being adjusted");
+ }
```
It turned out returning the adjustment target if the adjustment kind is `NeverToAny` is necessary, as otherwise we will go through a series of `delay_bug`s and eventually find that we constructed a `TyKind::Error` without having actually emitted an error.
This PR addresses that by re-returning the adjustment target if the adjustment kind is `NeverToAny`, partially reverting this change from #121208.
This PR has two commits:
1. The first commit adds a regression test for #134162, which will ICE (on stable 1.83.0, beta and nightly 2024-12-13).
2. The second commit is the partial revert, which will fix the ICE.
cc `@nnethercote` FYI as this is related to #121208 changes. The changes from #121208 exposed that we lacked test coverage for the code pattern reported in #134162.
Make some types and methods related to Polonius + Miri public
We have a tool, [Aquascope](https://github.com/cognitive-engineering-lab/aquascope/), which uses Polonius and Miri to visualize the compile-time and run-time semantics of a Rust program. Changes in the last few months to both APIs have hidden away details we depend upon. This PR re-exposes some of those details, specifically:
**Polonius:**
- `BorrowSet` and `BorrowData` are added to `rustc_borrowck::consumers`, and their fields are made `pub` instead of `pub(crate)`. We need this to interpret the `BorrowIndex`es generated by Polonius.
- `BorrowSet::build` is now `pub`. We need this because the borrowck API doesn't provide access to the `BorrowSet` constructed during checking.
- `PoloniusRegionVid` is added to `rustc_borrowck::consumers`. We need this because it's also contained in the Polonius facts.
**Miri:**
- `InterpCx::local_to_op` is now a special case of `local_at_frame_to_op`, which allows querying locals in any frame. We need this because we walk the whole stack at each step to collect the state of memory.
- `InterpCx::layout_of_local` is now `pub`. We need this because we need to know the layout of every local at each step.
If these changes go against some design goal for keeping certain types private, please let me know so we can hash out a better solution. Additionally, if there's a better way to document that it's important that certain types stay public, also let me know. For example, `BorrowSet` was previously public but was hidden in 6676cec, breaking our build.
cc ```@RalfJung``` ```@nnethercote``` ```@gavinleroy```
Update linux_musl base to dynamically link the crt by default
However, don't change the behavior of any existing targets at this time. For targets that used the old default, explicitly set `crt_static_default = true`.
This makes it easier for new targets to use the correct defaults while leaving the changing of individual targets to future PRs.
Related to https://github.com/rust-lang/compiler-team/issues/422
Add external macros specific diagnostics for check-cfg
This PR adds specific check-cfg diagnostics for unexpected cfg in external macros.
As well as hiding the some of the Cargo specific help/suggestions as they distraction for external macros and are generally not the right solution.
Follow-up to #132577
`@rustbot` label +L-unexpected_cfgs
r? compiler
Rollup of 6 pull requests
Successful merges:
- #132150 (Fix powerpc64 big-endian FreeBSD ABI)
- #133942 (Clarify how to use `black_box()`)
- #134081 (Try to evaluate constants in legacy mangling)
- #134192 (Remove `Lexer`'s dependency on `Parser`.)
- #134208 (coverage: Tidy up creation of covmap and covfun records)
- #134211 (On Neutrino QNX, reduce the need to set archiver via environment variables)
r? `@ghost`
`@rustbot` modify labels: rollup
(Re-)Implement `impl_trait_in_bindings`
This reimplements the `impl_trait_in_bindings` feature for local bindings.
"`impl Trait` in bindings" serve as a form of *trait* ascription, where the type basically functions as an infer var but additionally registering the `impl Trait`'s trait bounds for the infer type. These trait bounds can be used to enforce that predicates hold, and can guide inference (e.g. for closure signature inference):
```rust
let _: impl Fn(&u8) -> &u8 = |x| x;
```
They are implemented as an additional set of bounds that are registered when the type is lowered during typeck, and then these bounds are tied to a given `CanonicalUserTypeAscription` for borrowck. We enforce these `CanonicalUserTypeAscription` bounds during borrowck to make sure that the `impl Trait` types are sensitive to lifetimes:
```rust
trait Static: 'static {}
impl<T> Static for T where T: 'static {}
let local = 1;
let x: impl Static = &local;
//~^ ERROR `local` does not live long enough
```
r? oli-obk
cc #63065
---
Why can't we just use TAIT inference or something? Well, TAITs in bodies have the problem that they cannot reference lifetimes local to a body. For example:
```rust
type TAIT = impl Display;
let local = 0;
let x: TAIT = &local;
//~^ ERROR `local` does not live long enough
```
That's because TAITs requires us to do *opaque type inference* which is pretty strict, since we need to remap all of the lifetimes of the hidden type to universal regions. This is simply not possible here.
---
I consider this part of the "impl trait everywhere" experiment. I'm not certain if this needs yet another lang team experiment.
Without doing so, we'll run into a series of delayed bugs then find that
we have a `TyKind::Error` constructed yet fail to emit an error.
This partially reverts a change in
<https://github.com/rust-lang/rust/pull/121208> related to never type
adjustments in expr typecheck errors.
Rollup of 8 pull requests
Successful merges:
- #134252 (Fix `Path::is_absolute` on Hermit)
- #134254 (Fix building `std` for Hermit after `c_char` change)
- #134255 (Update includes in `/library/core/src/error.rs`.)
- #134261 (Document the symbol Visibility enum)
- #134262 (Arbitrary self types v2: adjust diagnostic.)
- #134265 (Rename `ty_def_id` so people will stop using it by accident)
- #134271 (Arbitrary self types v2: better feature gate test)
- #134274 (Add check-pass test for `&raw`)
r? `@ghost`
`@rustbot` modify labels: rollup
coverage: Tidy up creation of covmap and covfun records
This is a small follow-up to #134163 that mostly just inlines and renames some variables, and adds a few comments.
It also slightly defers the creation of the LLVM value that holds the filename table, to just before the value is needed.
---
try-job: x86_64-mingw-2
try-job: dist-x86_64-linux
Remove `Lexer`'s dependency on `Parser`.
Lexing precedes parsing, as you'd expect: `Lexer` creates a `TokenStream` and `Parser` then parses that `TokenStream`.
But, in a horrendous violation of layering abstractions and common sense, `Lexer` depends on `Parser`! The `Lexer::unclosed_delim_err` method does some error recovery that relies on creating a `Parser` to do some post-processing of the `TokenStream` that the `Lexer` just created.
This commit just removes `unclosed_delim_err`. This change removes `Lexer`'s dependency on `Parser`, and also means that `lex_token_tree`'s return value can have a more typical form.
The cost is slightly worse error messages in two obscure cases, as shown in these tests:
- tests/ui/parser/brace-in-let-chain.rs: there is slightly less explanation in this case involving an extra `{`.
- tests/ui/parser/diff-markers/unclosed-delims{,-in-macro}.rs: the diff marker detection is no longer supported (because that detection is implemented in the parser).
In my opinion this cost is outweighed by the magnitude of the code cleanup.
r? ```````@chenyukang```````
Try to evaluate constants in legacy mangling
Best reviewed commit by commit.
It seems kind of odd to treat literals differently from unevaluated free constants. So let's evaluate those constants and only fall back to `_` rendering if that fails to result in an integral constant
Rename `ty_def_id` so people will stop using it by accident
This function is just for cycle detection, but people keep using it because they think it's the right way of getting the def id from a `Ty` (and I can't blame them necessarily).
Arbitrary self types v2: adjust diagnostic.
The recently landed PR #132961 to adjust arbitrary self types was a bit overenthusiastic, advising folks to use the new Receiver trait even before it's been stabilized. Revert to the older wording of the lint in such cases.
Tracking issue #44874
r? ``@wesleywiser``
A bunch of cleanups (part 2)
Just like https://github.com/rust-lang/rust/pull/133567 these were all found while looking at the respective code, but are not blocking any other changes I want to make in the short term.
rustc_borrowck: Stop suggesting the invalid syntax `&mut raw const`
A legitimate suggestion would be to change from
&raw const val
to
&raw mut val
But until we have figured out how to make that happen we should at least
stop suggesting invalid syntax.
I recommend review commit-by-commit.
Part of #127562
Tweak multispan rendering to reduce output length
Consider comments and bare delimiters the same as an "empty line" for purposes of hiding rendered code output of long multispans. This results in more aggressive shortening of rendered output without losing too much context, specially in `*.stderr` tests that have "hidden" comments. We do that check not only on the first 4 lines of the multispan, but now also on the previous to last line as well.
Rollup of 7 pull requests
Successful merges:
- #133900 (Advent of `tests/ui` (misc cleanups and improvements) [1/N])
- #133937 (Keep track of parse errors in `mod`s and don't emit resolve errors for paths involving them)
- #133938 (`rustc_mir_dataflow` cleanups, including some renamings)
- #134058 (interpret: reduce usage of TypingEnv::fully_monomorphized)
- #134130 (Stop using driver queries in the public API)
- #134140 (Add AST support for unsafe binders)
- #134229 (Fix typos in docs on provenance)
r? `@ghost`
`@rustbot` modify labels: rollup
When we recover from a pattern parse error, or a pattern uses `..`, we keep track of that and affect resolution error for missing bindings that could have been provided by that pattern. We differentiate between `..` and parse recovery. We silence resolution errors likely caused by the pattern parse error.
```
error[E0425]: cannot find value `title` in this scope
--> $DIR/struct-pattern-with-missing-fields-resolve-error.rs:19:30
|
LL | println!("[{}]({})", title, url);
| ^^^^^ not found in this scope
|
note: `Website` has a field `title` which could have been included in this pattern, but it wasn't
--> $DIR/struct-pattern-with-missing-fields-resolve-error.rs:17:12
|
LL | / struct Website {
LL | | url: String,
LL | | title: Option<String> ,
| | ----- defined here
LL | | }
| |_-
...
LL | if let Website { url, .. } = website {
| ^^^^^^^^^^^^^^^^^^^ this pattern doesn't include `title`, which is available in `Website`
```
Fix#74863.
forbid toggling x87 and fpregs on hard-float targets
Part of https://github.com/rust-lang/rust/issues/116344, follow-up to https://github.com/rust-lang/rust/pull/129884:
The `x87` target feature on x86 and the `fpregs` target feature on ARM must not be disabled on a hardfloat target, as that would change the float ABI. However, *enabling* `fpregs` on ARM is [explicitly requested](https://github.com/rust-lang/rust/issues/130988) as it seems to be useful. Therefore, we need to refine the distinction of "forbidden" target features and "allowed" target features: all (un)stable target features can determine on a per-target basis whether they should be allowed to be toggled or not. `fpregs` then checks whether the current target has the `soft-float` feature, and if yes, `fpregs` is permitted -- otherwise, it is not. (Same for `x87` on x86).
Also fixes https://github.com/rust-lang/rust/issues/132351. Since `fpregs` and `x87` can be enabled on some builds and disabled on others, it would make sense that one can query it via `cfg`. Therefore, I made them behave in `cfg` like any other unstable target feature.
The first commit prepares the infrastructure, but does not change behavior. The second commit then wires up `fpregs` and `x87` with that new infrastructure.
r? `@workingjubilee`
Add AST support for unsafe binders
I'm splitting up #130514 into pieces. It's impossible for me to keep up with a huge PR like that. I'll land type system support for this next, probably w/o MIR lowering, which will come later.
r? `@oli-obk`
cc `@BoxyUwU` and `@lcnr` who also may want to look at this, though this PR doesn't do too much yet
