instantiate higher ranked goals in candidate selection again
This reverts #119820 as that PR has a significant impact and breaks code which *feels like it should work*. The impact ended up being larger than we expected during the FCP and we've ended up with some ideas for how we can work around this issue in the next solver. This has been discussed in the previous high bandwidth t-types meeting: https://rust-lang.zulipchat.com/#narrow/stream/326132-t-types.2Fmeetings/topic/2024-07-09.20high.20bandwidth.20meeting.
We'll therefore keep this inconsistency between the two solvers for now and will have to deal with it before stabilizating the use of the new solver outside of coherence: https://github.com/rust-lang/trait-system-refactor-initiative/issues/120.
fixes#125194 after a beta-backport.
The pattern which is more widely used than expected and feels like it should work, especially without deep knowledge of the type system is
```rust
trait Trait<'a> {}
impl<'a, T> Trait<'a> for T {}
fn trait_bound<T: for<'a> Trait<'a>>() {}
// A function with a where-bound which is more restrictive than the impl.
fn function1<T: Trait<'static>>() {
// stable: ok
// with #119820: error as we prefer the where-bound over the impl
// with this PR: back to ok
trait_bound::<T>();
}
```
r? `@rust-lang/types`
Fixup failing fuchsia tests
The Fuchsia platform passes all tests with these changes. Two tests are ignored because they rely on Fuchsia not returning a status code upon a process aborting. See #102032 and #58590 for more details on that topic.
Many formatting changes are also included in this PR.
r? tmandry
r? erickt
Update `f16`/`f128` FIXMEs that needed `(NEG_)INFINITY`
Just a small fix to the pattern matching tests now that we can. Also contains a small unrelated comment tweak.
Automatically taint when reporting errors from ItemCtxt
This isn't very robust yet, as you need to use `itemctxt.dcx()` instead of `tcx.dcx()` for it to take effect, but it's at least more convenient than sprinkling `set_tainted_by_errors` calls in individual places.
based on https://github.com/rust-lang/rust/pull/127357
r? `@fmease`
Both test-panic-abort-nocapture.rs and test-panic-abort.rs assert the
stderr output of the test. On Fuchsia, if a test fails an assertion,
this output will contain a line noting the process returned the code
-1028 (ZX_TASK_RETCODE_EXCEPTION_KILL). But the asserted stderr output
lacks this note. Presumably this is because other platforms implement
-Cpanic=abort by killing the process instead of returned a status
code.
Infer async closure signature from (old-style) two-part `Fn` + `Future` bounds
When an async closure is passed to a function that has a "two-part" `Fn` and `Future` trait bound, like:
```rust
use std::future::Future;
fn not_exactly_an_async_closure(_f: F)
where
F: FnOnce(String) -> Fut,
Fut: Future<Output = ()>,
{}
```
The we want to be able to extract the signature to guide inference in the async closure, like:
```rust
not_exactly_an_async_closure(async |string| {
for x in string.split('\n') { ... }
//~^ We need to know that the type of `string` is `String` to call methods on it.
})
```
Closure signature inference will see two bounds: `<?F as FnOnce<Args>>::Output = ?Fut`, `<?Fut as Future>::Output = String`. We need to extract the signature by looking through both projections.
### Why?
I expect the ecosystem's move onto `async Fn` trait bounds (which are not affected by this PR, and already do signature inference fine) to be slow. In the mean time, I don't see major overhead to supporting this "old–style" of trait bounds that were used to model async closures.
r? oli-obk
Fixes#127468Fixes#127425
Mark format! with must_use hint
Uses unstable feature https://github.com/rust-lang/rust/issues/94745
Part of #126475
First contribution to rust, please let me know if the blessing of tests is correct
Thanks `@bjorn3` for the help
Make casts of pointers to trait objects stricter
This is an attempt to `fix` https://github.com/rust-lang/rust/issues/120222 and https://github.com/rust-lang/rust/issues/120217.
This is done by adding restrictions on casting pointers to trait objects.
Before this PR the rules were as follows:
> When casting `*const X<dyn A>` -> `*const Y<dyn B>`, principal traits in `A` and `B` must refer to the same trait definition (or no trait).
With this PR the rules are changed to
> When casting `*const X<dyn Src>` -> `*const Y<dyn Dst>`
> - if `Dst` has a principal trait `DstP`,
> - `Src` must have a principal trait `SrcP`
> - `dyn SrcP` and `dyn DstP` must be the same type (modulo the trait object lifetime, `dyn T+'a` -> `dyn T+'b` is allowed)
> - Auto traits in `Dst` must be a subset of auto traits in `Src`
> - Not adhering to this is currently a FCW (warn-by-default + `FutureReleaseErrorReportInDeps`), instead of an error
> - if `Src` has a principal trait `Dst` must as well
> - this restriction will be removed in a follow up PR
This ensures that
1. Principal trait's generic arguments match (no `*const dyn Tr<A>` -> `*const dyn Tr<B>` casts, which are a problem for [#120222](https://github.com/rust-lang/rust/issues/120222))
2. Principal trait's lifetime arguments match (no `*const dyn Tr<'a>` -> `*const dyn Tr<'b>` casts, which are a problem for [#120217](https://github.com/rust-lang/rust/issues/120217))
3. No auto traits can be _added_ (this is a problem for arbitrary self types, see [this comment](https://github.com/rust-lang/rust/pull/120248#discussion_r1463835350))
Some notes:
- We only care about the metadata/last field, so you can still cast `*const dyn T` to `*const WithHeader<dyn T>`, etc
- The lifetime of the trait object itself (`dyn A + 'lt`) is not checked, so you can still cast `*mut FnOnce() + '_` to `*mut FnOnce() + 'static`, etc
- This feels fishy, but I couldn't come up with a reason it must be checked
The diagnostics are currently not great, to say the least, but as far as I can tell this correctly fixes the issues.
cc `@oli-obk` `@compiler-errors` `@lcnr`
Rollup of 10 pull requests
Successful merges:
- #126841 ([`macro_metavar_expr_concat`] Add support for literals)
- #126881 (Make `NEVER_TYPE_FALLBACK_FLOWING_INTO_UNSAFE` a deny-by-default lint in edition 2024)
- #126921 (Give VaList its own home)
- #127367 (Run alloc sync tests)
- #127431 (Use field ident spans directly instead of the full field span in diagnostics on local fields)
- #127437 (Uplift trait ref is knowable into `rustc_next_trait_solver`)
- #127439 (Uplift elaboration into `rustc_type_ir`)
- #127451 (Improve `run-make/output-type-permutations` code and improve `filename_not_in_denylist` API)
- #127452 (Fix intrinsic const parameter counting with `effects`)
- #127459 (rustdoc-json: add type/trait alias tests)
r? `@ghost`
`@rustbot` modify labels: rollup
Use field ident spans directly instead of the full field span in diagnostics on local fields
This improves diagnostics and avoids having to store the `DefId`s of fields
Make `NEVER_TYPE_FALLBACK_FLOWING_INTO_UNSAFE` a deny-by-default lint in edition 2024
I don't actually really care about this, but ``@traviscross`` asked me to do this, because lang team briefly discussed this before.
(TC here:)
Specifically, our original FCPed plan included this step:
- Add a lint against fallback affecting a generic that is passed to an `unsafe` function.
- Perhaps make this lint `deny-by-default` or a hard error in Rust 2024.
That is, we had left as an open question strengthening this in Rust 2024, and had marked it as an open question on the tracking issue. We're nominating here to address the open question. (Closing the remaining open question helps us to fully mark this off for Rust 2024.)
r? ``@compiler-errors``
Tracking:
- https://github.com/rust-lang/rust/issues/123748
[`macro_metavar_expr_concat`] Add support for literals
Adds support for things like `${concat($variable, 123)}` or `${concat("hello", "_world")}` .
cc #124225
Support tail calls in mir via `TerminatorKind::TailCall`
This is one of the interesting bits in tail call implementation — MIR support.
This adds a new `TerminatorKind` which represents a tail call:
```rust
TailCall {
func: Operand<'tcx>,
args: Vec<Operand<'tcx>>,
fn_span: Span,
},
```
*Structurally* this is very similar to a normal `Call` but is missing a few fields:
- `destination` — tail calls don't write to destination, instead they pass caller's destination to the callee (such that eventual `return` will write to the caller of the function that used tail call)
- `target` — similarly to `destination` tail calls pass the caller's return address to the callee, so there is nothing to do
- `unwind` — I _think_ this is applicable too, although it's a bit confusing
- `call_source` — `become` forbids operators and is not created as a lowering of something else; tail calls always come from HIR (at least for now)
It might be helpful to read the interpreter implementation to understand what `TailCall` means exactly, although I've tried documenting it too.
-----
There are a few `FIXME`-questions still left, ideally we'd be able to answer them during review ':)
-----
r? `@oli-obk`
cc `@scottmcm` `@DrMeepster` `@JakobDegen`
Make `can_eq` process obligations (almost) everywhere
Move `can_eq` to an extension trait on `InferCtxt` in `rustc_trait_selection`, and change it so that it processes obligations. This should strengthen it to be more accurate in some cases, but is most important for the new trait solver which delays relating aliases to `AliasRelate` goals. Without this, we always basically just return true when passing aliases to `can_eq`, which can lead to weird errors, for example #127149.
I'm not actually certain if we should *have* `can_eq` be called on the good path. In cases where we need `can_eq`, we probably should just be using a regular probe.
Fixes#127149
r? lcnr
Emit a wrap expr span_bug only if context is not tainted
Fixes#127332
The ICE occurs because of this `span_bug`: 51917e2e69/compiler/rustc_hir_typeck/src/expr_use_visitor.rs (L732-L738)
which is triggered by the fact that we're trying to use an `enum` in a `with` expression instead of a `struct`.
The issue originates in commit 814bfe9335 from PR #127202. As per the title of that commit the ICEing code should not be reachable any more, but looks like it still is.
This PR changes the code so that the `span_bug` will be emitted only if the context is not tainted by a previous error.
Don't try to label `ObligationCauseCode::CompareImplItem` for an RPITIT, since it has no name
The old (current) trait solver has a limitation that when a where clause in param-env must be normalized using the same where clause, then we get spurious errors in `normalize_param_env_or_error`. I don't think there's an issue tracking it, but it's the root cause for many of the "fixed-by-next-solver" labeled issues.
Specifically, these errors may occur when checking predicate entailment of the GAT that comes out of desugaring RPITITs. Since we use `ObligationCauseCode::CompareImplItem` for these predicates, we try calling `item_name` on an RPITIT which fails, since the RPITIT has no name.
We simply suppress this logic when we're reporting a predicate entailment error for an RPITIT. RPITITs should never have predicate entailment errors, *by construction*, but they may due to this bug in the old solver.
Addresses the ICE in #127331, though doesn't fix the underlying issue (which is fundamental to the old solver).
r? types
