//! Trait Resolution. See the [rustc dev guide] for more information on how this works. //! //! [rustc dev guide]: https://rustc-dev-guide.rust-lang.org/traits/resolution.html pub mod auto_trait; mod chalk_fulfill; pub mod codegen; mod coherence; pub mod const_evaluatable; mod engine; pub mod error_reporting; mod fulfill; pub mod misc; mod object_safety; mod on_unimplemented; pub mod outlives_bounds; mod project; pub mod query; pub(crate) mod relationships; mod select; mod specialize; mod structural_match; mod util; pub mod wf; use crate::errors::DumpVTableEntries; use crate::infer::outlives::env::OutlivesEnvironment; use crate::infer::{InferCtxt, TyCtxtInferExt}; use crate::traits::error_reporting::InferCtxtExt as _; use crate::traits::query::evaluate_obligation::InferCtxtExt as _; use rustc_errors::ErrorGuaranteed; use rustc_hir as hir; use rustc_hir::def_id::DefId; use rustc_hir::lang_items::LangItem; use rustc_infer::traits::TraitEngineExt as _; use rustc_middle::ty::fold::TypeFoldable; use rustc_middle::ty::subst::{InternalSubsts, SubstsRef}; use rustc_middle::ty::visit::TypeVisitable; use rustc_middle::ty::{ self, DefIdTree, GenericParamDefKind, Subst, ToPredicate, Ty, TyCtxt, TypeSuperVisitable, VtblEntry, }; use rustc_span::{sym, Span}; use smallvec::SmallVec; use std::fmt::Debug; use std::ops::ControlFlow; pub use self::FulfillmentErrorCode::*; pub use self::ImplSource::*; pub use self::ObligationCauseCode::*; pub use self::SelectionError::*; pub use self::coherence::{add_placeholder_note, orphan_check, overlapping_impls}; pub use self::coherence::{OrphanCheckErr, OverlapResult}; pub use self::engine::{ObligationCtxt, TraitEngineExt}; pub use self::fulfill::{FulfillmentContext, PendingPredicateObligation}; pub use self::object_safety::astconv_object_safety_violations; pub use self::object_safety::is_vtable_safe_method; pub use self::object_safety::MethodViolationCode; pub use self::object_safety::ObjectSafetyViolation; pub use self::on_unimplemented::{OnUnimplementedDirective, OnUnimplementedNote}; pub use self::project::{normalize, normalize_projection_type, normalize_to}; pub use self::select::{EvaluationCache, SelectionCache, SelectionContext}; pub use self::select::{EvaluationResult, IntercrateAmbiguityCause, OverflowError}; pub use self::specialize::specialization_graph::FutureCompatOverlapError; pub use self::specialize::specialization_graph::FutureCompatOverlapErrorKind; pub use self::specialize::{specialization_graph, translate_substs, OverlapError}; pub use self::structural_match::{ search_for_adt_const_param_violation, search_for_structural_match_violation, }; pub use self::util::{ elaborate_obligations, elaborate_predicates, elaborate_predicates_with_span, elaborate_trait_ref, elaborate_trait_refs, }; pub use self::util::{expand_trait_aliases, TraitAliasExpander}; pub use self::util::{ get_vtable_index_of_object_method, impl_item_is_final, predicate_for_trait_def, upcast_choices, }; pub use self::util::{ supertrait_def_ids, supertraits, transitive_bounds, transitive_bounds_that_define_assoc_type, SupertraitDefIds, Supertraits, }; pub use self::chalk_fulfill::FulfillmentContext as ChalkFulfillmentContext; pub use rustc_infer::traits::*; /// Whether to skip the leak check, as part of a future compatibility warning step. /// /// The "default" for skip-leak-check corresponds to the current /// behavior (do not skip the leak check) -- not the behavior we are /// transitioning into. #[derive(Copy, Clone, PartialEq, Eq, Debug, Default)] pub enum SkipLeakCheck { Yes, #[default] No, } impl SkipLeakCheck { fn is_yes(self) -> bool { self == SkipLeakCheck::Yes } } /// The mode that trait queries run in. #[derive(Copy, Clone, PartialEq, Eq, Debug)] pub enum TraitQueryMode { /// Standard/un-canonicalized queries get accurate /// spans etc. passed in and hence can do reasonable /// error reporting on their own. Standard, /// Canonicalized queries get dummy spans and hence /// must generally propagate errors to /// pre-canonicalization callsites. Canonical, } /// Creates predicate obligations from the generic bounds. pub fn predicates_for_generics<'tcx>( cause: impl Fn(usize, Span) -> ObligationCause<'tcx>, param_env: ty::ParamEnv<'tcx>, generic_bounds: ty::InstantiatedPredicates<'tcx>, ) -> impl Iterator> { let generic_bounds = generic_bounds; debug!("predicates_for_generics(generic_bounds={:?})", generic_bounds); std::iter::zip(generic_bounds.predicates, generic_bounds.spans).enumerate().map( move |(idx, (predicate, span))| Obligation { cause: cause(idx, span), recursion_depth: 0, param_env: param_env, predicate, }, ) } /// Determines whether the type `ty` is known to meet `bound` and /// returns true if so. Returns false if `ty` either does not meet /// `bound` or is not known to meet bound (note that this is /// conservative towards *no impl*, which is the opposite of the /// `evaluate` methods). pub fn type_known_to_meet_bound_modulo_regions<'a, 'tcx>( infcx: &InferCtxt<'a, 'tcx>, param_env: ty::ParamEnv<'tcx>, ty: Ty<'tcx>, def_id: DefId, span: Span, ) -> bool { debug!( "type_known_to_meet_bound_modulo_regions(ty={:?}, bound={:?})", ty, infcx.tcx.def_path_str(def_id) ); let trait_ref = ty::Binder::dummy(ty::TraitRef { def_id, substs: infcx.tcx.mk_substs_trait(ty, &[]) }); let obligation = Obligation { param_env, cause: ObligationCause::misc(span, hir::CRATE_HIR_ID), recursion_depth: 0, predicate: trait_ref.without_const().to_predicate(infcx.tcx), }; let result = infcx.predicate_must_hold_modulo_regions(&obligation); debug!( "type_known_to_meet_ty={:?} bound={} => {:?}", ty, infcx.tcx.def_path_str(def_id), result ); if result && ty.has_infer_types_or_consts() { // Because of inference "guessing", selection can sometimes claim // to succeed while the success requires a guess. To ensure // this function's result remains infallible, we must confirm // that guess. While imperfect, I believe this is sound. // We can use a dummy node-id here because we won't pay any mind // to region obligations that arise (there shouldn't really be any // anyhow). let cause = ObligationCause::misc(span, hir::CRATE_HIR_ID); // The handling of regions in this area of the code is terrible, // see issue #29149. We should be able to improve on this with // NLL. let errors = fully_solve_bound(infcx, cause, param_env, ty, def_id); // Note: we only assume something is `Copy` if we can // *definitively* show that it implements `Copy`. Otherwise, // assume it is move; linear is always ok. match &errors[..] { [] => { debug!( "type_known_to_meet_bound_modulo_regions: ty={:?} bound={} success", ty, infcx.tcx.def_path_str(def_id) ); true } errors => { debug!( ?ty, bound = %infcx.tcx.def_path_str(def_id), ?errors, "type_known_to_meet_bound_modulo_regions" ); false } } } else { result } } #[instrument(level = "debug", skip(tcx, elaborated_env))] fn do_normalize_predicates<'tcx>( tcx: TyCtxt<'tcx>, cause: ObligationCause<'tcx>, elaborated_env: ty::ParamEnv<'tcx>, predicates: Vec>, ) -> Result>, ErrorGuaranteed> { let span = cause.span; // FIXME. We should really... do something with these region // obligations. But this call just continues the older // behavior (i.e., doesn't cause any new bugs), and it would // take some further refactoring to actually solve them. In // particular, we would have to handle implied bounds // properly, and that code is currently largely confined to // regionck (though I made some efforts to extract it // out). -nmatsakis // // @arielby: In any case, these obligations are checked // by wfcheck anyway, so I'm not sure we have to check // them here too, and we will remove this function when // we move over to lazy normalization *anyway*. tcx.infer_ctxt().ignoring_regions().enter(|infcx| { let predicates = match fully_normalize(&infcx, cause, elaborated_env, predicates) { Ok(predicates) => predicates, Err(errors) => { let reported = infcx.report_fulfillment_errors(&errors, None, false); return Err(reported); } }; debug!("do_normalize_predictes: normalized predicates = {:?}", predicates); // We can use the `elaborated_env` here; the region code only // cares about declarations like `'a: 'b`. let outlives_env = OutlivesEnvironment::new(elaborated_env); // FIXME: It's very weird that we ignore region obligations but apparently // still need to use `resolve_regions` as we need the resolved regions in // the normalized predicates. let errors = infcx.resolve_regions(&outlives_env); if !errors.is_empty() { tcx.sess.delay_span_bug( span, format!( "failed region resolution while normalizing {elaborated_env:?}: {errors:?}" ), ); } match infcx.fully_resolve(predicates) { Ok(predicates) => Ok(predicates), Err(fixup_err) => { // If we encounter a fixup error, it means that some type // variable wound up unconstrained. I actually don't know // if this can happen, and I certainly don't expect it to // happen often, but if it did happen it probably // represents a legitimate failure due to some kind of // unconstrained variable. // // @lcnr: Let's still ICE here for now. I want a test case // for that. span_bug!( span, "inference variables in normalized parameter environment: {}", fixup_err ); } } }) } // FIXME: this is gonna need to be removed ... /// Normalizes the parameter environment, reporting errors if they occur. #[instrument(level = "debug", skip(tcx))] pub fn normalize_param_env_or_error<'tcx>( tcx: TyCtxt<'tcx>, unnormalized_env: ty::ParamEnv<'tcx>, cause: ObligationCause<'tcx>, ) -> ty::ParamEnv<'tcx> { // I'm not wild about reporting errors here; I'd prefer to // have the errors get reported at a defined place (e.g., // during typeck). Instead I have all parameter // environments, in effect, going through this function // and hence potentially reporting errors. This ensures of // course that we never forget to normalize (the // alternative seemed like it would involve a lot of // manual invocations of this fn -- and then we'd have to // deal with the errors at each of those sites). // // In any case, in practice, typeck constructs all the // parameter environments once for every fn as it goes, // and errors will get reported then; so outside of type inference we // can be sure that no errors should occur. let mut predicates: Vec<_> = util::elaborate_predicates(tcx, unnormalized_env.caller_bounds().into_iter()) .map(|obligation| obligation.predicate) .collect(); debug!("normalize_param_env_or_error: elaborated-predicates={:?}", predicates); let elaborated_env = ty::ParamEnv::new( tcx.intern_predicates(&predicates), unnormalized_env.reveal(), unnormalized_env.constness(), ); // HACK: we are trying to normalize the param-env inside *itself*. The problem is that // normalization expects its param-env to be already normalized, which means we have // a circularity. // // The way we handle this is by normalizing the param-env inside an unnormalized version // of the param-env, which means that if the param-env contains unnormalized projections, // we'll have some normalization failures. This is unfortunate. // // Lazy normalization would basically handle this by treating just the // normalizing-a-trait-ref-requires-itself cycles as evaluation failures. // // Inferred outlives bounds can create a lot of `TypeOutlives` predicates for associated // types, so to make the situation less bad, we normalize all the predicates *but* // the `TypeOutlives` predicates first inside the unnormalized parameter environment, and // then we normalize the `TypeOutlives` bounds inside the normalized parameter environment. // // This works fairly well because trait matching does not actually care about param-env // TypeOutlives predicates - these are normally used by regionck. let outlives_predicates: Vec<_> = predicates .drain_filter(|predicate| { matches!(predicate.kind().skip_binder(), ty::PredicateKind::TypeOutlives(..)) }) .collect(); debug!( "normalize_param_env_or_error: predicates=(non-outlives={:?}, outlives={:?})", predicates, outlives_predicates ); let Ok(non_outlives_predicates) = do_normalize_predicates( tcx, cause.clone(), elaborated_env, predicates, ) else { // An unnormalized env is better than nothing. debug!("normalize_param_env_or_error: errored resolving non-outlives predicates"); return elaborated_env; }; debug!("normalize_param_env_or_error: non-outlives predicates={:?}", non_outlives_predicates); // Not sure whether it is better to include the unnormalized TypeOutlives predicates // here. I believe they should not matter, because we are ignoring TypeOutlives param-env // predicates here anyway. Keeping them here anyway because it seems safer. let outlives_env: Vec<_> = non_outlives_predicates.iter().chain(&outlives_predicates).cloned().collect(); let outlives_env = ty::ParamEnv::new( tcx.intern_predicates(&outlives_env), unnormalized_env.reveal(), unnormalized_env.constness(), ); let Ok(outlives_predicates) = do_normalize_predicates( tcx, cause, outlives_env, outlives_predicates, ) else { // An unnormalized env is better than nothing. debug!("normalize_param_env_or_error: errored resolving outlives predicates"); return elaborated_env; }; debug!("normalize_param_env_or_error: outlives predicates={:?}", outlives_predicates); let mut predicates = non_outlives_predicates; predicates.extend(outlives_predicates); debug!("normalize_param_env_or_error: final predicates={:?}", predicates); ty::ParamEnv::new( tcx.intern_predicates(&predicates), unnormalized_env.reveal(), unnormalized_env.constness(), ) } /// Normalize a type and process all resulting obligations, returning any errors pub fn fully_normalize<'a, 'tcx, T>( infcx: &InferCtxt<'a, 'tcx>, cause: ObligationCause<'tcx>, param_env: ty::ParamEnv<'tcx>, value: T, ) -> Result>> where T: TypeFoldable<'tcx>, { debug!("fully_normalize_with_fulfillcx(value={:?})", value); let selcx = &mut SelectionContext::new(infcx); let Normalized { value: normalized_value, obligations } = project::normalize(selcx, param_env, cause, value); debug!( "fully_normalize: normalized_value={:?} obligations={:?}", normalized_value, obligations ); let mut fulfill_cx = FulfillmentContext::new(); for obligation in obligations { fulfill_cx.register_predicate_obligation(infcx, obligation); } debug!("fully_normalize: select_all_or_error start"); let errors = fulfill_cx.select_all_or_error(infcx); if !errors.is_empty() { return Err(errors); } debug!("fully_normalize: select_all_or_error complete"); let resolved_value = infcx.resolve_vars_if_possible(normalized_value); debug!("fully_normalize: resolved_value={:?}", resolved_value); Ok(resolved_value) } /// Process an obligation (and any nested obligations that come from it) to /// completion, returning any errors pub fn fully_solve_obligation<'a, 'tcx>( infcx: &InferCtxt<'a, 'tcx>, obligation: PredicateObligation<'tcx>, ) -> Vec> { let mut engine = >::new(infcx.tcx); engine.register_predicate_obligation(infcx, obligation); engine.select_all_or_error(infcx) } /// Process a set of obligations (and any nested obligations that come from them) /// to completion pub fn fully_solve_obligations<'a, 'tcx>( infcx: &InferCtxt<'a, 'tcx>, obligations: impl IntoIterator>, ) -> Vec> { let mut engine = >::new(infcx.tcx); engine.register_predicate_obligations(infcx, obligations); engine.select_all_or_error(infcx) } /// Process a bound (and any nested obligations that come from it) to completion. /// This is a convenience function for traits that have no generic arguments, such /// as auto traits, and builtin traits like Copy or Sized. pub fn fully_solve_bound<'a, 'tcx>( infcx: &InferCtxt<'a, 'tcx>, cause: ObligationCause<'tcx>, param_env: ty::ParamEnv<'tcx>, ty: Ty<'tcx>, bound: DefId, ) -> Vec> { let mut engine = >::new(infcx.tcx); engine.register_bound(infcx, param_env, ty, bound, cause); engine.select_all_or_error(infcx) } /// Normalizes the predicates and checks whether they hold in an empty environment. If this /// returns true, then either normalize encountered an error or one of the predicates did not /// hold. Used when creating vtables to check for unsatisfiable methods. pub fn impossible_predicates<'tcx>( tcx: TyCtxt<'tcx>, predicates: Vec>, ) -> bool { debug!("impossible_predicates(predicates={:?})", predicates); let result = tcx.infer_ctxt().enter(|infcx| { let param_env = ty::ParamEnv::reveal_all(); let ocx = ObligationCtxt::new(&infcx); let predicates = ocx.normalize(ObligationCause::dummy(), param_env, predicates); for predicate in predicates { let obligation = Obligation::new(ObligationCause::dummy(), param_env, predicate); ocx.register_obligation(obligation); } let errors = ocx.select_all_or_error(); // Clean up after ourselves let _ = infcx.inner.borrow_mut().opaque_type_storage.take_opaque_types(); !errors.is_empty() }); debug!("impossible_predicates = {:?}", result); result } fn subst_and_check_impossible_predicates<'tcx>( tcx: TyCtxt<'tcx>, key: (DefId, SubstsRef<'tcx>), ) -> bool { debug!("subst_and_check_impossible_predicates(key={:?})", key); let mut predicates = tcx.predicates_of(key.0).instantiate(tcx, key.1).predicates; // Specifically check trait fulfillment to avoid an error when trying to resolve // associated items. if let Some(trait_def_id) = tcx.trait_of_item(key.0) { let trait_ref = ty::TraitRef::from_method(tcx, trait_def_id, key.1); predicates.push(ty::Binder::dummy(trait_ref).to_poly_trait_predicate().to_predicate(tcx)); } predicates.retain(|predicate| !predicate.needs_subst()); let result = impossible_predicates(tcx, predicates); debug!("subst_and_check_impossible_predicates(key={:?}) = {:?}", key, result); result } /// Checks whether a trait's method is impossible to call on a given impl. /// /// This only considers predicates that reference the impl's generics, and not /// those that reference the method's generics. fn is_impossible_method<'tcx>( tcx: TyCtxt<'tcx>, (impl_def_id, trait_item_def_id): (DefId, DefId), ) -> bool { struct ReferencesOnlyParentGenerics<'tcx> { tcx: TyCtxt<'tcx>, generics: &'tcx ty::Generics, trait_item_def_id: DefId, } impl<'tcx> ty::TypeVisitor<'tcx> for ReferencesOnlyParentGenerics<'tcx> { type BreakTy = (); fn visit_ty(&mut self, t: Ty<'tcx>) -> ControlFlow { // If this is a parameter from the trait item's own generics, then bail if let ty::Param(param) = t.kind() && let param_def_id = self.generics.type_param(param, self.tcx).def_id && self.tcx.parent(param_def_id) == self.trait_item_def_id { return ControlFlow::BREAK; } t.super_visit_with(self) } fn visit_region(&mut self, r: ty::Region<'tcx>) -> ControlFlow { if let ty::ReEarlyBound(param) = r.kind() && let param_def_id = self.generics.region_param(¶m, self.tcx).def_id && self.tcx.parent(param_def_id) == self.trait_item_def_id { return ControlFlow::BREAK; } r.super_visit_with(self) } fn visit_const(&mut self, ct: ty::Const<'tcx>) -> ControlFlow { if let ty::ConstKind::Param(param) = ct.kind() && let param_def_id = self.generics.const_param(¶m, self.tcx).def_id && self.tcx.parent(param_def_id) == self.trait_item_def_id { return ControlFlow::BREAK; } ct.super_visit_with(self) } } let generics = tcx.generics_of(trait_item_def_id); let predicates = tcx.predicates_of(trait_item_def_id); let impl_trait_ref = tcx.impl_trait_ref(impl_def_id).expect("expected impl to correspond to trait"); let param_env = tcx.param_env(impl_def_id); let mut visitor = ReferencesOnlyParentGenerics { tcx, generics, trait_item_def_id }; let predicates_for_trait = predicates.predicates.iter().filter_map(|(pred, span)| { if pred.visit_with(&mut visitor).is_continue() { Some(Obligation::new( ObligationCause::dummy_with_span(*span), param_env, ty::EarlyBinder(*pred).subst(tcx, impl_trait_ref.substs), )) } else { None } }); tcx.infer_ctxt().ignoring_regions().enter(|ref infcx| { for obligation in predicates_for_trait { // Ignore overflow error, to be conservative. if let Ok(result) = infcx.evaluate_obligation(&obligation) && !result.may_apply() { return true; } } false }) } #[derive(Clone, Debug)] enum VtblSegment<'tcx> { MetadataDSA, TraitOwnEntries { trait_ref: ty::PolyTraitRef<'tcx>, emit_vptr: bool }, } /// Prepare the segments for a vtable fn prepare_vtable_segments<'tcx, T>( tcx: TyCtxt<'tcx>, trait_ref: ty::PolyTraitRef<'tcx>, mut segment_visitor: impl FnMut(VtblSegment<'tcx>) -> ControlFlow, ) -> Option { // The following constraints holds for the final arrangement. // 1. The whole virtual table of the first direct super trait is included as the // the prefix. If this trait doesn't have any super traits, then this step // consists of the dsa metadata. // 2. Then comes the proper pointer metadata(vptr) and all own methods for all // other super traits except those already included as part of the first // direct super trait virtual table. // 3. finally, the own methods of this trait. // This has the advantage that trait upcasting to the first direct super trait on each level // is zero cost, and to another trait includes only replacing the pointer with one level indirection, // while not using too much extra memory. // For a single inheritance relationship like this, // D --> C --> B --> A // The resulting vtable will consists of these segments: // DSA, A, B, C, D // For a multiple inheritance relationship like this, // D --> C --> A // \-> B // The resulting vtable will consists of these segments: // DSA, A, B, B-vptr, C, D // For a diamond inheritance relationship like this, // D --> B --> A // \-> C -/ // The resulting vtable will consists of these segments: // DSA, A, B, C, C-vptr, D // For a more complex inheritance relationship like this: // O --> G --> C --> A // \ \ \-> B // | |-> F --> D // | \-> E // |-> N --> J --> H // \ \-> I // |-> M --> K // \-> L // The resulting vtable will consists of these segments: // DSA, A, B, B-vptr, C, D, D-vptr, E, E-vptr, F, F-vptr, G, // H, H-vptr, I, I-vptr, J, J-vptr, K, K-vptr, L, L-vptr, M, M-vptr, // N, N-vptr, O // emit dsa segment first. if let ControlFlow::Break(v) = (segment_visitor)(VtblSegment::MetadataDSA) { return Some(v); } let mut emit_vptr_on_new_entry = false; let mut visited = util::PredicateSet::new(tcx); let predicate = trait_ref.without_const().to_predicate(tcx); let mut stack: SmallVec<[(ty::PolyTraitRef<'tcx>, _, _); 5]> = smallvec![(trait_ref, emit_vptr_on_new_entry, None)]; visited.insert(predicate); // the main traversal loop: // basically we want to cut the inheritance directed graph into a few non-overlapping slices of nodes // that each node is emitted after all its descendents have been emitted. // so we convert the directed graph into a tree by skipping all previously visited nodes using a visited set. // this is done on the fly. // Each loop run emits a slice - it starts by find a "childless" unvisited node, backtracking upwards, and it // stops after it finds a node that has a next-sibling node. // This next-sibling node will used as the starting point of next slice. // Example: // For a diamond inheritance relationship like this, // D#1 --> B#0 --> A#0 // \-> C#1 -/ // Starting point 0 stack [D] // Loop run #0: Stack after diving in is [D B A], A is "childless" // after this point, all newly visited nodes won't have a vtable that equals to a prefix of this one. // Loop run #0: Emitting the slice [B A] (in reverse order), B has a next-sibling node, so this slice stops here. // Loop run #0: Stack after exiting out is [D C], C is the next starting point. // Loop run #1: Stack after diving in is [D C], C is "childless", since its child A is skipped(already emitted). // Loop run #1: Emitting the slice [D C] (in reverse order). No one has a next-sibling node. // Loop run #1: Stack after exiting out is []. Now the function exits. loop { // dive deeper into the stack, recording the path 'diving_in: loop { if let Some((inner_most_trait_ref, _, _)) = stack.last() { let inner_most_trait_ref = *inner_most_trait_ref; let mut direct_super_traits_iter = tcx .super_predicates_of(inner_most_trait_ref.def_id()) .predicates .into_iter() .filter_map(move |(pred, _)| { pred.subst_supertrait(tcx, &inner_most_trait_ref).to_opt_poly_trait_pred() }); 'diving_in_skip_visited_traits: loop { if let Some(next_super_trait) = direct_super_traits_iter.next() { if visited.insert(next_super_trait.to_predicate(tcx)) { // We're throwing away potential constness of super traits here. // FIXME: handle ~const super traits let next_super_trait = next_super_trait.map_bound(|t| t.trait_ref); stack.push(( next_super_trait, emit_vptr_on_new_entry, Some(direct_super_traits_iter), )); break 'diving_in_skip_visited_traits; } else { continue 'diving_in_skip_visited_traits; } } else { break 'diving_in; } } } } // Other than the left-most path, vptr should be emitted for each trait. emit_vptr_on_new_entry = true; // emit innermost item, move to next sibling and stop there if possible, otherwise jump to outer level. 'exiting_out: loop { if let Some((inner_most_trait_ref, emit_vptr, siblings_opt)) = stack.last_mut() { if let ControlFlow::Break(v) = (segment_visitor)(VtblSegment::TraitOwnEntries { trait_ref: *inner_most_trait_ref, emit_vptr: *emit_vptr, }) { return Some(v); } 'exiting_out_skip_visited_traits: loop { if let Some(siblings) = siblings_opt { if let Some(next_inner_most_trait_ref) = siblings.next() { if visited.insert(next_inner_most_trait_ref.to_predicate(tcx)) { // We're throwing away potential constness of super traits here. // FIXME: handle ~const super traits let next_inner_most_trait_ref = next_inner_most_trait_ref.map_bound(|t| t.trait_ref); *inner_most_trait_ref = next_inner_most_trait_ref; *emit_vptr = emit_vptr_on_new_entry; break 'exiting_out; } else { continue 'exiting_out_skip_visited_traits; } } } stack.pop(); continue 'exiting_out; } } // all done return None; } } } fn dump_vtable_entries<'tcx>( tcx: TyCtxt<'tcx>, sp: Span, trait_ref: ty::PolyTraitRef<'tcx>, entries: &[VtblEntry<'tcx>], ) { tcx.sess.emit_err(DumpVTableEntries { span: sp, trait_ref, entries: format!("{:#?}", entries), }); } fn own_existential_vtable_entries<'tcx>( tcx: TyCtxt<'tcx>, trait_ref: ty::PolyExistentialTraitRef<'tcx>, ) -> &'tcx [DefId] { let trait_methods = tcx .associated_items(trait_ref.def_id()) .in_definition_order() .filter(|item| item.kind == ty::AssocKind::Fn); // Now list each method's DefId (for within its trait). let own_entries = trait_methods.filter_map(move |trait_method| { debug!("own_existential_vtable_entry: trait_method={:?}", trait_method); let def_id = trait_method.def_id; // Some methods cannot be called on an object; skip those. if !is_vtable_safe_method(tcx, trait_ref.def_id(), &trait_method) { debug!("own_existential_vtable_entry: not vtable safe"); return None; } Some(def_id) }); tcx.arena.alloc_from_iter(own_entries.into_iter()) } /// Given a trait `trait_ref`, iterates the vtable entries /// that come from `trait_ref`, including its supertraits. fn vtable_entries<'tcx>( tcx: TyCtxt<'tcx>, trait_ref: ty::PolyTraitRef<'tcx>, ) -> &'tcx [VtblEntry<'tcx>] { debug!("vtable_entries({:?})", trait_ref); let mut entries = vec![]; let vtable_segment_callback = |segment| -> ControlFlow<()> { match segment { VtblSegment::MetadataDSA => { entries.extend(TyCtxt::COMMON_VTABLE_ENTRIES); } VtblSegment::TraitOwnEntries { trait_ref, emit_vptr } => { let existential_trait_ref = trait_ref .map_bound(|trait_ref| ty::ExistentialTraitRef::erase_self_ty(tcx, trait_ref)); // Lookup the shape of vtable for the trait. let own_existential_entries = tcx.own_existential_vtable_entries(existential_trait_ref); let own_entries = own_existential_entries.iter().copied().map(|def_id| { debug!("vtable_entries: trait_method={:?}", def_id); // The method may have some early-bound lifetimes; add regions for those. let substs = trait_ref.map_bound(|trait_ref| { InternalSubsts::for_item(tcx, def_id, |param, _| match param.kind { GenericParamDefKind::Lifetime => tcx.lifetimes.re_erased.into(), GenericParamDefKind::Type { .. } | GenericParamDefKind::Const { .. } => { trait_ref.substs[param.index as usize] } }) }); // The trait type may have higher-ranked lifetimes in it; // erase them if they appear, so that we get the type // at some particular call site. let substs = tcx .normalize_erasing_late_bound_regions(ty::ParamEnv::reveal_all(), substs); // It's possible that the method relies on where-clauses that // do not hold for this particular set of type parameters. // Note that this method could then never be called, so we // do not want to try and codegen it, in that case (see #23435). let predicates = tcx.predicates_of(def_id).instantiate_own(tcx, substs); if impossible_predicates(tcx, predicates.predicates) { debug!("vtable_entries: predicates do not hold"); return VtblEntry::Vacant; } let instance = ty::Instance::resolve_for_vtable( tcx, ty::ParamEnv::reveal_all(), def_id, substs, ) .expect("resolution failed during building vtable representation"); VtblEntry::Method(instance) }); entries.extend(own_entries); if emit_vptr { entries.push(VtblEntry::TraitVPtr(trait_ref)); } } } ControlFlow::Continue(()) }; let _ = prepare_vtable_segments(tcx, trait_ref, vtable_segment_callback); if tcx.has_attr(trait_ref.def_id(), sym::rustc_dump_vtable) { let sp = tcx.def_span(trait_ref.def_id()); dump_vtable_entries(tcx, sp, trait_ref, &entries); } tcx.arena.alloc_from_iter(entries.into_iter()) } /// Find slot base for trait methods within vtable entries of another trait fn vtable_trait_first_method_offset<'tcx>( tcx: TyCtxt<'tcx>, key: ( ty::PolyTraitRef<'tcx>, // trait_to_be_found ty::PolyTraitRef<'tcx>, // trait_owning_vtable ), ) -> usize { let (trait_to_be_found, trait_owning_vtable) = key; // #90177 let trait_to_be_found_erased = tcx.erase_regions(trait_to_be_found); let vtable_segment_callback = { let mut vtable_base = 0; move |segment| { match segment { VtblSegment::MetadataDSA => { vtable_base += TyCtxt::COMMON_VTABLE_ENTRIES.len(); } VtblSegment::TraitOwnEntries { trait_ref, emit_vptr } => { if tcx.erase_regions(trait_ref) == trait_to_be_found_erased { return ControlFlow::Break(vtable_base); } vtable_base += util::count_own_vtable_entries(tcx, trait_ref); if emit_vptr { vtable_base += 1; } } } ControlFlow::Continue(()) } }; if let Some(vtable_base) = prepare_vtable_segments(tcx, trait_owning_vtable, vtable_segment_callback) { vtable_base } else { bug!("Failed to find info for expected trait in vtable"); } } /// Find slot offset for trait vptr within vtable entries of another trait pub fn vtable_trait_upcasting_coercion_new_vptr_slot<'tcx>( tcx: TyCtxt<'tcx>, key: ( Ty<'tcx>, // trait object type whose trait owning vtable Ty<'tcx>, // trait object for supertrait ), ) -> Option { let (source, target) = key; assert!(matches!(&source.kind(), &ty::Dynamic(..)) && !source.needs_infer()); assert!(matches!(&target.kind(), &ty::Dynamic(..)) && !target.needs_infer()); // this has been typecked-before, so diagnostics is not really needed. let unsize_trait_did = tcx.require_lang_item(LangItem::Unsize, None); let trait_ref = ty::TraitRef { def_id: unsize_trait_did, substs: tcx.mk_substs_trait(source, &[target.into()]), }; let obligation = Obligation::new( ObligationCause::dummy(), ty::ParamEnv::reveal_all(), ty::Binder::dummy(ty::TraitPredicate { trait_ref, constness: ty::BoundConstness::NotConst, polarity: ty::ImplPolarity::Positive, }), ); let implsrc = tcx.infer_ctxt().enter(|infcx| { let mut selcx = SelectionContext::new(&infcx); selcx.select(&obligation).unwrap() }); let Some(ImplSource::TraitUpcasting(implsrc_traitcasting)) = implsrc else { bug!(); }; implsrc_traitcasting.vtable_vptr_slot } pub fn provide(providers: &mut ty::query::Providers) { object_safety::provide(providers); structural_match::provide(providers); *providers = ty::query::Providers { specialization_graph_of: specialize::specialization_graph_provider, specializes: specialize::specializes, codegen_select_candidate: codegen::codegen_select_candidate, own_existential_vtable_entries, vtable_entries, vtable_trait_upcasting_coercion_new_vptr_slot, subst_and_check_impossible_predicates, is_impossible_method, try_unify_abstract_consts: |tcx, param_env_and| { let (param_env, (a, b)) = param_env_and.into_parts(); const_evaluatable::try_unify_abstract_consts(tcx, (a, b), param_env) }, ..*providers }; }