| Age | Commit message (Collapse) | Author | Lines |
|---|
|
|
|
`-Zhigher-ranked-assumptions`: Consider WF of coroutine witness when proving outlives assumptions
### TL;DR
This PR introduces an unstable flag `-Zhigher-ranked-assumptions` which tests out a new algorithm for dealing with some of the higher-ranked outlives problems that come from auto trait bounds on coroutines. See:
* rust-lang/rust#110338
While it doesn't fix all of the issues, it certainly fixed many of them, so I'd like to get this landed so people can test the flag on their own code.
### Background
Consider, for example:
```rust
use std::future::Future;
trait Client {
type Connecting<'a>: Future + Send
where
Self: 'a;
fn connect(&self) -> Self::Connecting<'_>;
}
fn call_connect<C>(c: C) -> impl Future + Send
where
C: Client + Send + Sync,
{
async move { c.connect().await }
}
```
Due to the fact that we erase the lifetimes in a coroutine, we can think of the interior type of the async block as something like: `exists<'r, 's> { C, &'r C, C::Connecting<'s> }`. The first field is the `c` we capture, the second is the auto-ref that we perform on the call to `.connect()`, and the third is the resulting future we're awaiting at the first and only await point. Note that every region is uniquified differently in the interior types.
For the async block to be `Send`, we must prove that both of the interior types are `Send`. First, we have an `exists<'r, 's>` binder, which needs to be instantiated universally since we treat the regions in this binder as *unknown*[^exist]. This gives us two types: `{ &'!r C, C::Connecting<'!s> }`. Proving `&'!r C: Send` is easy due to a [`Send`](https://doc.rust-lang.org/nightly/std/marker/trait.Send.html#impl-Send-for-%26T) impl for references.
Proving `C::Connecting<'!s>: Send` can only be done via the item bound, which then requires `C: '!s` to hold (due to the `where Self: 'a` on the associated type definition). Unfortunately, we don't know that `C: '!s` since we stripped away any relationship between the interior type and the param `C`. This leads to a bogus borrow checker error today!
### Approach
Coroutine interiors are well-formed by virtue of them being borrow-checked, as long as their callers are invoking their parent functions in a well-formed way, then substitutions should also be well-formed. Therefore, in our example above, we should be able to deduce the assumption that `C: '!s` holds from the well-formedness of the interior type `C::Connecting<'!s>`.
This PR introduces the notion of *coroutine assumptions*, which are the outlives assumptions that we can assume hold due to the well-formedness of a coroutine's interior types. These are computed alongside the coroutine types in the `CoroutineWitnessTypes` struct. When we instantiate the binder when proving an auto trait for a coroutine, we instantiate the `CoroutineWitnessTypes` and stash these newly instantiated assumptions in the region storage in the `InferCtxt`. Later on in lexical region resolution or MIR borrowck, we use these registered assumptions to discharge any placeholder outlives obligations that we would otherwise not be able to prove.
### How well does it work?
I've added a ton of tests of different reported situations that users have shared on issues like rust-lang/rust#110338, and an (anecdotally) large number of those examples end up working straight out of the box! Some limitations are described below.
### How badly does it not work?
The behavior today is quite rudimentary, since we currently discharge the placeholder assumptions pretty early in region resolution. This manifests itself as some limitations on the code that we accept.
For example, `tests/ui/async-await/higher-ranked-auto-trait-11.rs` continues to fail. In that test, we must prove that a placeholder is equal to a universal for a param-env candidate to hold when proving an auto trait, e.g. `'!1 = 'a` is required to prove `T: Trait<'!1>` in a param-env that has `T: Trait<'a>`. Unfortunately, at that point in the MIR body, we only know that the placeholder is equal to some body-local existential NLL var `'?2`, which only gets equated to the universal `'a` when being stored into the return local later on in MIR borrowck.
This could be fixed by integrating these assumptions into the type outlives machinery in a more first-class way, and delaying things to the end of MIR typeck when we know the full relationship between existential and universal NLL vars. Doing this integration today is quite difficult today.
`tests/ui/async-await/higher-ranked-auto-trait-11.rs` fails because we don't compute the full transitive outlives relations between placeholders. In that test, we have in our region assumptions that some `'!1 = '!2` and `'!2 = '!3`, but we must prove `'!1 = '!3`.
This can be fixed by computing the set of coroutine outlives assumptions in a more transitive way, or as I mentioned above, integrating these assumptions into the type outlives machinery in a more first-class way, since it's already responsible for the transitive outlives assumptions of universals.
### Moving forward
I'm still quite happy with this implementation, and I'd like to land it for testing. I may work on overhauling both the way we compute these coroutine assumptions and also how we deal with the assumptions during (lexical/nll) region checking. But for now, I'd like to give users a chance to try out this new `-Zhigher-ranked-assumptions` flag to uncover more shortcomings.
[^exist]: Instantiating this binder with infer regions would be incomplete, since we'd be asking for *some* instantiation of the interior types, not proving something for *all* instantiations of the interior types.
|
|
|
|
|
|
|
|
|
|
|
|
Fast path for WF goals in new solver
Hopefully self-explanatory.
|
|
Add note to `find_const_ty_from_env`
Add a note to `find_const_ty_from_env` to explain why it has an `unwrap` which "often" causes ICEs.
Also, uplift it into the new trait solver. This avoids needing to go through the interner to call this method which is otherwise an inherent method in the compiler. I can remove this part if desired.
r? `@boxyuwu`
|
|
Remove some glob imports from the type system
Namely, remove the glob imports for `BoundRegionConversionTime`, `RegionVariableOrigin`, `SubregionOrigin`, `TyOrConstInferVar`, `RegionResolutionError`, `SelectionError`, `ProjectionCandidate`, `ProjectionCandidateSet`, and some more specific scoped globs (like `Inserted` in the impl overlap graph construction.
These glob imports are IMO very low value, since they're not used nearly as often as other globs (like `TyKind`).
|
|
`evaluate_goal` avoid unnecessary step
based on rust-lang/rust#142617.
This does not mess with the debug logging for the trait solver and is a very nice cleanup for rust-lang/rust#142735. E.g. for
```rust
#[derive(Clone)]
struct Wrapper<T>(T);
#[derive(Clone)]
struct Nested; // using a separate type to avoid the fast paths
fn is_clone<T: Clone>() {}
fn main() {
is_clone::<Wrapper<Nested>>();
}
```
We get the following proof tree with `RUSTC_LOG=rustc_type_ir::search_graph=debug,rustc_next_trait_solver=debug`
```
rustc_next_trait_solver::solve::eval_ctxt::evaluate_root_goal goal=Goal { param_env: ParamEnv { caller_bounds: [] }, predicate: Binder { value: TraitPredicate(<Wrapper<Nested> as std::clone::Clone>, polarity:Positive), bound_vars: [] } }, generate_proof_tree=No, span=src/main.rs:7:5: 7:34 (#0), stalled_on=None
rustc_type_ir::search_graph::evaluate_goal input=CanonicalQueryInput { canonical: Canonical { value: QueryInput { goal: Goal { param_env: ParamEnv { caller_bounds: [] }, predicate: Binder { value: TraitPredicate(<Wrapper<Nested> as std::clone::Clone>, polarity:Positive), bound_vars: [] } }, predefined_opaques_in_body: PredefinedOpaques(PredefinedOpaquesData { opaque_types: [] }) }, max_universe: U0, variables: [] }, typing_mode: Analysis { defining_opaque_types_and_generators: [] } }, step_kind_from_parent=Unknown
rustc_next_trait_solver::solve::eval_ctxt::probe::enter source=Impl(DefId(0:10 ~ main[21d2]::{impl#0}))
rustc_next_trait_solver::solve::eval_ctxt::add_goal source=ImplWhereBound, goal=Goal { param_env: ParamEnv { caller_bounds: [] }, predicate: Binder { value: TraitPredicate(<_ as std::marker::Sized>, polarity:Positive), bound_vars: [] } }
rustc_next_trait_solver::solve::eval_ctxt::add_goal source=ImplWhereBound, goal=Goal { param_env: ParamEnv { caller_bounds: [] }, predicate: Binder { value: TraitPredicate(<_ as std::clone::Clone>, polarity:Positive), bound_vars: [] } }
rustc_type_ir::search_graph::evaluate_goal input=CanonicalQueryInput { canonical: Canonical { value: QueryInput { goal: Goal { param_env: ParamEnv { caller_bounds: [] }, predicate: Binder { value: TraitPredicate(<Nested as std::clone::Clone>, polarity:Positive), bound_vars: [] } }, predefined_opaques_in_body: PredefinedOpaques(PredefinedOpaquesData { opaque_types: [] }) }, max_universe: U0, variables: [] }, typing_mode: Analysis { defining_opaque_types_and_generators: [] } }, step_kind_from_parent=Unknown
0ms DEBUG rustc_type_ir::search_graph global cache hit, required_depth=0
0ms DEBUG rustc_type_ir::search_graph return=Ok(Canonical { value: Response { certainty: Yes, var_values: CanonicalVarValues { var_values: [] }, external_constraints: ExternalConstraints(ExternalConstraintsData { region_constraints: [], opaque_types: [], normalization_nested_goals: NestedNormalizationGoals([]) }) }, max_universe: U0, variables: [] })
rustc_next_trait_solver::solve::eval_ctxt::probe::enter source=BuiltinImpl(Misc)
rustc_next_trait_solver::solve::trait_goals::merge_trait_candidates candidates=[Candidate { source: Impl(DefId(0:10 ~ main[21d2]::{impl#0})), result: Canonical { value: Response { certainty: Yes, var_values: CanonicalVarValues { var_values: [] }, external_constraints: ExternalConstraints(ExternalConstraintsData { region_constraints: [], opaque_types: [], normalization_nested_goals: NestedNormalizationGoals([]) }) }, max_universe: U0, variables: [] } }]
0ms DEBUG rustc_next_trait_solver::solve::trait_goals return=Ok((Canonical { value: Response { certainty: Yes, var_values: CanonicalVarValues { var_values: [] }, external_constraints: ExternalConstraints(ExternalConstraintsData { region_constraints: [], opaque_types: [], normalization_nested_goals: NestedNormalizationGoals([]) }) }, max_universe: U0, variables: [] }, Some(Misc)))
0ms DEBUG rustc_type_ir::search_graph insert global cache, evaluation_result=EvaluationResult { encountered_overflow: false, required_depth: 1, heads: CycleHeads { heads: {} }, nested_goals: NestedGoals { nested_goals: {} }, result: Ok(Canonical { value: Response { certainty: Yes, var_values: CanonicalVarValues { var_values: [] }, external_constraints: ExternalConstraints(ExternalConstraintsData { region_constraints: [], opaque_types: [], normalization_nested_goals: NestedNormalizationGoals([]) }) }, max_universe: U0, variables: [] }) }
0ms DEBUG rustc_type_ir::search_graph return=Ok(Canonical { value: Response { certainty: Yes, var_values: CanonicalVarValues { var_values: [] }, external_constraints: ExternalConstraints(ExternalConstraintsData { region_constraints: [], opaque_types: [], normalization_nested_goals: NestedNormalizationGoals([]) }) }, max_universe: U0, variables: [] })
```
|
|
|
|
|
|
|
|
|
|
|
|
Extend the fast path for `Sized` traits to include constness and
`MetaSized`.
|
|
|
|
Filter out universals and lifetimes from `stalled_vars`
lol
r? lcnr
|
|
Treat normalizing consts like normalizing types in deeply normalize
...so that we don't end up putting a top-level normalizes-to goal in the fulfillment context, which ICEs. This basically just models the normalize-const code off of the normalize-ty code above it, which uses an alias-relate goal instead.
Fixes rust-lang/rust#140571
r? lcnr
|
|
Clear nested candidates in select if certainty is yes
Proving these goals is redundant.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
add additional `TypeFlags` fast paths
Some crates, e.g. `diesel`, have items with a lot of where-clauses (more than 150). In these cases checking the `TypeFlags` of the whole `param_env` can be very beneficial.
This adds `fn fold_clauses` to mirror the existing `fn visit_clauses` and then uses this in folders which fold `ParamEnv`s.
Split out from rust-lang/rust#141451, depends on rust-lang/rust#141442.
r? `@compiler-errors`
|
|
|
|
Rollup of 10 pull requests
Successful merges:
- rust-lang/rust#140898 (minor improvements on running miri)
- rust-lang/rust#141392 (Avoid obligation construction dance with query region constraints)
- rust-lang/rust#141431 (Emit dummy open drop for unsafe binder)
- rust-lang/rust#141433 (Properly analyze captures from unsafe binders)
- rust-lang/rust#141439 (Deduplicate dyn compatibility violations due to coercion)
- rust-lang/rust#141449 (further deduplicate ast visitor code)
- rust-lang/rust#141513 (interpret: add allocation parameters to `AllocBytes`)
- rust-lang/rust#141516 (speed up charsearcher for ascii chars)
- rust-lang/rust#141526 (add a dedicated section for compiler environment variables in the unstable book)
- rust-lang/rust#141550 (Fix `unused_braces` lint suggestion when encountering attributes)
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Rollup of 8 pull requests
Successful merges:
- #141355 (ci: improve citool job db errors)
- #141359 (Fix `FnOnce` impl for `AsyncFn`/`AsyncFnMut` self-borrowing closures in new solver)
- #141362 (Normalize aliases to correct kind of error term)
- #141377 (Remove unnecessary `is_empty` checks)
- #141381 (try_cast_aligned: avoid bare int-to-ptr casts)
- #141382 (ci: convert distcheck to free runner)
- #141389 (ci: prepare aws access keys for migration)
- #141390 (Don't allow `poly_select` in new solver)
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
|
|
|
|
|
|
Do not discard constraints on overflow if there was candidate ambiguity
Fixes https://github.com/rust-lang/trait-system-refactor-initiative/issues/201.
There's a pretty chunky justification in the test.
r? lcnr
|
|
|
|
with ambiguity
|
|
|
|
This commit does the following:
- Replaces use of rustc_type_ir by rustc_middle in rustc_infer.
- The DelayedMap type is exposed by rustc_middle so everything can be
accessed through rustc_middle in a coherent manner.
- API-layer traits, like InferCtxtLike, Interner or inherent::* must be
accessed via rustc_type_ir, not rustc_middle::ty. For this reason
these are not reexported by rustc_middle::ty.
- Replaces use of ty::Interner by rustc_type_ir::Interner in
rustc_trait_selection
|
|
Minor tweaks to make some normalization (adjacent) code less confusing
r? lcnr
sorry for double ping lol
|
|
This commit does the following:
- Replaces use of rustc_type_ir by rustc_middle
- Removes the rustc_type_ir dependency
- The DelayedSet type is exposed by rustc_middle so everything can be
accessed through rustc_middle in a coherent manner.
|
|
|
