diff options
Diffstat (limited to 'compiler/rustc_middle/src/ty/outlives.rs')
| -rw-r--r-- | compiler/rustc_middle/src/ty/outlives.rs | 206 |
1 files changed, 206 insertions, 0 deletions
diff --git a/compiler/rustc_middle/src/ty/outlives.rs b/compiler/rustc_middle/src/ty/outlives.rs new file mode 100644 index 00000000000..1a8693b8df7 --- /dev/null +++ b/compiler/rustc_middle/src/ty/outlives.rs @@ -0,0 +1,206 @@ +// The outlines relation `T: 'a` or `'a: 'b`. This code frequently +// refers to rules defined in RFC 1214 (`OutlivesFooBar`), so see that +// RFC for reference. + +use crate::ty::subst::{GenericArg, GenericArgKind}; +use crate::ty::{self, Ty, TyCtxt, TypeFoldable}; +use smallvec::SmallVec; + +#[derive(Debug)] +pub enum Component<'tcx> { + Region(ty::Region<'tcx>), + Param(ty::ParamTy), + UnresolvedInferenceVariable(ty::InferTy), + + // Projections like `T::Foo` are tricky because a constraint like + // `T::Foo: 'a` can be satisfied in so many ways. There may be a + // where-clause that says `T::Foo: 'a`, or the defining trait may + // include a bound like `type Foo: 'static`, or -- in the most + // conservative way -- we can prove that `T: 'a` (more generally, + // that all components in the projection outlive `'a`). This code + // is not in a position to judge which is the best technique, so + // we just product the projection as a component and leave it to + // the consumer to decide (but see `EscapingProjection` below). + Projection(ty::ProjectionTy<'tcx>), + + // In the case where a projection has escaping regions -- meaning + // regions bound within the type itself -- we always use + // the most conservative rule, which requires that all components + // outlive the bound. So for example if we had a type like this: + // + // for<'a> Trait1< <T as Trait2<'a,'b>>::Foo > + // ~~~~~~~~~~~~~~~~~~~~~~~~~ + // + // then the inner projection (underlined) has an escaping region + // `'a`. We consider that outer trait `'c` to meet a bound if `'b` + // outlives `'b: 'c`, and we don't consider whether the trait + // declares that `Foo: 'static` etc. Therefore, we just return the + // free components of such a projection (in this case, `'b`). + // + // However, in the future, we may want to get smarter, and + // actually return a "higher-ranked projection" here. Therefore, + // we mark that these components are part of an escaping + // projection, so that implied bounds code can avoid relying on + // them. This gives us room to improve the regionck reasoning in + // the future without breaking backwards compat. + EscapingProjection(Vec<Component<'tcx>>), +} + +impl<'tcx> TyCtxt<'tcx> { + /// Push onto `out` all the things that must outlive `'a` for the condition + /// `ty0: 'a` to hold. Note that `ty0` must be a **fully resolved type**. + pub fn push_outlives_components(self, ty0: Ty<'tcx>, out: &mut SmallVec<[Component<'tcx>; 4]>) { + compute_components(self, ty0, out); + debug!("components({:?}) = {:?}", ty0, out); + } +} + +fn compute_components(tcx: TyCtxt<'tcx>, ty: Ty<'tcx>, out: &mut SmallVec<[Component<'tcx>; 4]>) { + // Descend through the types, looking for the various "base" + // components and collecting them into `out`. This is not written + // with `collect()` because of the need to sometimes skip subtrees + // in the `subtys` iterator (e.g., when encountering a + // projection). + match ty.kind { + ty::FnDef(_, substs) => { + // HACK(eddyb) ignore lifetimes found shallowly in `substs`. + // This is inconsistent with `ty::Adt` (including all substs) + // and with `ty::Closure` (ignoring all substs other than + // upvars, of which a `ty::FnDef` doesn't have any), but + // consistent with previous (accidental) behavior. + // See https://github.com/rust-lang/rust/issues/70917 + // for further background and discussion. + for child in substs { + match child.unpack() { + GenericArgKind::Type(ty) => { + compute_components(tcx, ty, out); + } + GenericArgKind::Lifetime(_) => {} + GenericArgKind::Const(_) => { + compute_components_recursive(tcx, child, out); + } + } + } + } + + ty::Array(element, _) => { + // Don't look into the len const as it doesn't affect regions + compute_components(tcx, element, out); + } + + ty::Closure(_, ref substs) => { + for upvar_ty in substs.as_closure().upvar_tys() { + compute_components(tcx, upvar_ty, out); + } + } + + ty::Generator(_, ref substs, _) => { + // Same as the closure case + for upvar_ty in substs.as_generator().upvar_tys() { + compute_components(tcx, upvar_ty, out); + } + + // We ignore regions in the generator interior as we don't + // want these to affect region inference + } + + // All regions are bound inside a witness + ty::GeneratorWitness(..) => (), + + // OutlivesTypeParameterEnv -- the actual checking that `X:'a` + // is implied by the environment is done in regionck. + ty::Param(p) => { + out.push(Component::Param(p)); + } + + // For projections, we prefer to generate an obligation like + // `<P0 as Trait<P1...Pn>>::Foo: 'a`, because this gives the + // regionck more ways to prove that it holds. However, + // regionck is not (at least currently) prepared to deal with + // higher-ranked regions that may appear in the + // trait-ref. Therefore, if we see any higher-ranke regions, + // we simply fallback to the most restrictive rule, which + // requires that `Pi: 'a` for all `i`. + ty::Projection(ref data) => { + if !data.has_escaping_bound_vars() { + // best case: no escaping regions, so push the + // projection and skip the subtree (thus generating no + // constraints for Pi). This defers the choice between + // the rules OutlivesProjectionEnv, + // OutlivesProjectionTraitDef, and + // OutlivesProjectionComponents to regionck. + out.push(Component::Projection(*data)); + } else { + // fallback case: hard code + // OutlivesProjectionComponents. Continue walking + // through and constrain Pi. + let mut subcomponents = smallvec![]; + compute_components_recursive(tcx, ty.into(), &mut subcomponents); + out.push(Component::EscapingProjection(subcomponents.into_iter().collect())); + } + } + + // We assume that inference variables are fully resolved. + // So, if we encounter an inference variable, just record + // the unresolved variable as a component. + ty::Infer(infer_ty) => { + out.push(Component::UnresolvedInferenceVariable(infer_ty)); + } + + // Most types do not introduce any region binders, nor + // involve any other subtle cases, and so the WF relation + // simply constraints any regions referenced directly by + // the type and then visits the types that are lexically + // contained within. (The comments refer to relevant rules + // from RFC1214.) + ty::Bool | // OutlivesScalar + ty::Char | // OutlivesScalar + ty::Int(..) | // OutlivesScalar + ty::Uint(..) | // OutlivesScalar + ty::Float(..) | // OutlivesScalar + ty::Never | // ... + ty::Adt(..) | // OutlivesNominalType + ty::Opaque(..) | // OutlivesNominalType (ish) + ty::Foreign(..) | // OutlivesNominalType + ty::Str | // OutlivesScalar (ish) + ty::Slice(..) | // ... + ty::RawPtr(..) | // ... + ty::Ref(..) | // OutlivesReference + ty::Tuple(..) | // ... + ty::FnPtr(_) | // OutlivesFunction (*) + ty::Dynamic(..) | // OutlivesObject, OutlivesFragment (*) + ty::Placeholder(..) | + ty::Bound(..) | + ty::Error(_) => { + // (*) Function pointers and trait objects are both binders. + // In the RFC, this means we would add the bound regions to + // the "bound regions list". In our representation, no such + // list is maintained explicitly, because bound regions + // themselves can be readily identified. + compute_components_recursive(tcx, ty.into(), out); + } + } +} + +fn compute_components_recursive( + tcx: TyCtxt<'tcx>, + parent: GenericArg<'tcx>, + out: &mut SmallVec<[Component<'tcx>; 4]>, +) { + for child in parent.walk_shallow() { + match child.unpack() { + GenericArgKind::Type(ty) => { + compute_components(tcx, ty, out); + } + GenericArgKind::Lifetime(lt) => { + // Ignore late-bound regions. + if !lt.is_late_bound() { + out.push(Component::Region(lt)); + } + } + GenericArgKind::Const(_) => { + compute_components_recursive(tcx, child, out); + } + } + } +} |