recursively evaluate the constants in everything that is 'mentioned'

compiler/rustc_data_structures/src/sync.rs

@@ -20,6 +20,7 @@
 //! | ----------------------- | ------------------- | ------------------------------- |
 //! | `Lrc<T>`                | `rc::Rc<T>`         | `sync::Arc<T>`                  |
 //! |` Weak<T>`               | `rc::Weak<T>`       | `sync::Weak<T>`                 |
+//! | `LRef<'a, T>` [^2]      | `&'a mut T`         | `&'a T`                         |
 //! |                         |                     |                                 |
 //! | `AtomicBool`            | `Cell<bool>`        | `atomic::AtomicBool`            |
 //! | `AtomicU32`             | `Cell<u32>`         | `atomic::AtomicU32`             |
@@ -38,7 +39,7 @@
 //! of a `RefCell`. This is appropriate when interior mutability is not
 //! required.
 //!
-//! [^2] `MTLockRef` is a typedef.
+//! [^2] `MTRef`, `MTLockRef` are type aliases.
 
 pub use crate::marker::*;
 use std::collections::HashMap;
@@ -208,7 +209,7 @@ cfg_match! {
 
         use std::cell::RefCell as InnerRwLock;
 
-        pub type MTLockRef<'a, T> = &'a mut MTLock<T>;
+        pub type LRef<'a, T> = &'a mut T;
 
         #[derive(Debug, Default)]
         pub struct MTLock<T>(T);
@@ -274,7 +275,7 @@ cfg_match! {
         pub use std::sync::Arc as Lrc;
         pub use std::sync::Weak as Weak;
 
-        pub type MTLockRef<'a, T> = &'a MTLock<T>;
+        pub type LRef<'a, T> = &'a T;
 
         #[derive(Debug, Default)]
         pub struct MTLock<T>(Lock<T>);
@@ -314,6 +315,8 @@ cfg_match! {
     }
 }
 
+pub type MTLockRef<'a, T> = LRef<'a, MTLock<T>>;
+
 #[derive(Default)]
 #[cfg_attr(parallel_compiler, repr(align(64)))]
 pub struct CacheAligned<T>(pub T);

compiler/rustc_middle/src/mir/mod.rs

@@ -20,6 +20,7 @@ use rustc_hir::def_id::{DefId, CRATE_DEF_ID};
 use rustc_hir::{self, CoroutineDesugaring, CoroutineKind, ImplicitSelfKind};
 use rustc_hir::{self as hir, HirId};
 use rustc_session::Session;
+use rustc_span::source_map::Spanned;
 use rustc_target::abi::{FieldIdx, VariantIdx};
 
 use polonius_engine::Atom;
@@ -44,6 +45,7 @@ use std::ops::{Index, IndexMut};
 use std::{iter, mem};
 
 pub use self::query::*;
+use self::visit::TyContext;
 pub use basic_blocks::BasicBlocks;
 
 mod basic_blocks;
@@ -312,6 +314,21 @@ impl<'tcx> CoroutineInfo<'tcx> {
     }
 }
 
+/// Some item that needs to monomorphize successfully for a MIR body to be considered well-formed.
+#[derive(Copy, Clone, PartialEq, Eq, Debug, Hash, HashStable, TyEncodable, TyDecodable)]
+#[derive(TypeFoldable, TypeVisitable)]
+pub enum MentionedItem<'tcx> {
+    /// A function that gets called. We don't necessarily know its precise type yet, since it can be
+    /// hidden behind a generic.
+    Fn(Ty<'tcx>),
+    /// A type that has its drop shim called.
+    Drop(Ty<'tcx>),
+    /// Unsizing casts might require vtables, so we have to record them.
+    UnsizeCast { source_ty: Ty<'tcx>, target_ty: Ty<'tcx> },
+    /// A closure that is coerced to a function pointer.
+    Closure(Ty<'tcx>),
+}
+
 /// The lowered representation of a single function.
 #[derive(Clone, TyEncodable, TyDecodable, Debug, HashStable, TypeFoldable, TypeVisitable)]
 pub struct Body<'tcx> {
@@ -375,8 +392,24 @@ pub struct Body<'tcx> {
 
     /// Constants that are required to evaluate successfully for this MIR to be well-formed.
     /// We hold in this field all the constants we are not able to evaluate yet.
+    ///
+    /// This is soundness-critical, we make a guarantee that all consts syntactically mentioned in a
+    /// function have successfully evaluated if the function ever gets executed at runtime.
     pub required_consts: Vec<ConstOperand<'tcx>>,
 
+    /// Further items that were mentioned in this function and hence *may* become monomorphized,
+    /// depending on optimizations. We use this to avoid optimization-dependent compile errors: the
+    /// collector recursively traverses all "mentioned" items and evaluates all their
+    /// `required_consts`.
+    ///
+    /// This is *not* soundness-critical and the contents of this list are *not* a stable guarantee.
+    /// All that's relevant is that this set is optimization-level-independent, and that it includes
+    /// everything that the collector would consider "used". (For example, we currently compute this
+    /// set after drop elaboration, so some drop calls that can never be reached are not considered
+    /// "mentioned".) See the documentation of `CollectionMode` in
+    /// `compiler/rustc_monomorphize/src/collector.rs` for more context.
+    pub mentioned_items: Vec<Spanned<MentionedItem<'tcx>>>,
+
     /// Does this body use generic parameters. This is used for the `ConstEvaluatable` check.
     ///
     /// Note that this does not actually mean that this body is not computable right now.
@@ -453,6 +486,7 @@ impl<'tcx> Body<'tcx> {
             var_debug_info,
             span,
             required_consts: Vec::new(),
+            mentioned_items: Vec::new(),
             is_polymorphic: false,
             injection_phase: None,
             tainted_by_errors,
@@ -482,6 +516,7 @@ impl<'tcx> Body<'tcx> {
             spread_arg: None,
             span: DUMMY_SP,
             required_consts: Vec::new(),
+            mentioned_items: Vec::new(),
             var_debug_info: Vec::new(),
             is_polymorphic: false,
             injection_phase: None,
@@ -576,6 +611,17 @@ impl<'tcx> Body<'tcx> {
         }
     }
 
+    pub fn span_for_ty_context(&self, ty_context: TyContext) -> Span {
+        match ty_context {
+            TyContext::UserTy(span) => span,
+            TyContext::ReturnTy(source_info)
+            | TyContext::LocalDecl { source_info, .. }
+            | TyContext::YieldTy(source_info)
+            | TyContext::ResumeTy(source_info) => source_info.span,
+            TyContext::Location(loc) => self.source_info(loc).span,
+        }
+    }
+
     /// Returns the return type; it always return first element from `local_decls` array.
     #[inline]
     pub fn return_ty(&self) -> Ty<'tcx> {

compiler/rustc_mir_build/src/build/custom/mod.rs

@@ -56,6 +56,7 @@ pub(super) fn build_custom_mir<'tcx>(
         var_debug_info: Vec::new(),
         span,
         required_consts: Vec::new(),
+        mentioned_items: Vec::new(),
         is_polymorphic: false,
         tainted_by_errors: None,
         injection_phase: None,

compiler/rustc_mir_transform/src/inline.rs

@@ -565,7 +565,8 @@ impl<'tcx> Inliner<'tcx> {
         mut callee_body: Body<'tcx>,
     ) {
         let terminator = caller_body[callsite.block].terminator.take().unwrap();
-        let TerminatorKind::Call { args, destination, unwind, target, .. } = terminator.kind else {
+        let TerminatorKind::Call { func, args, destination, unwind, target, .. } = terminator.kind
+        else {
             bug!("unexpected terminator kind {:?}", terminator.kind);
         };
 
@@ -717,6 +718,24 @@ impl<'tcx> Inliner<'tcx> {
                 Const::Val(..) | Const::Unevaluated(..) => true,
             },
         ));
+        // Now that we incorporated the callee's `required_consts`, we can remove the callee from
+        // `mentioned_items` -- but we have to take their `mentioned_items` in return. This does
+        // some extra work here to save the monomorphization collector work later. It helps a lot,
+        // since monomorphization can avoid a lot of work when the "mentioned items" are similar to
+        // the actually used items. By doing this we can entirely avoid visiting the callee!
+        // We need to reconstruct the `required_item` for the callee so that we can find and
+        // remove it.
+        let callee_item = MentionedItem::Fn(func.ty(caller_body, self.tcx));
+        if let Some(idx) =
+            caller_body.mentioned_items.iter().position(|item| item.node == callee_item)
+        {
+            // We found the callee, so remove it and add its items instead.
+            caller_body.mentioned_items.remove(idx);
+            caller_body.mentioned_items.extend(callee_body.mentioned_items);
+        } else {
+            // If we can't find the callee, there's no point in adding its items.
+            // Probably it already got removed by being inlined elsewhere in the same function.
+        }
     }
 
     fn make_call_args(

compiler/rustc_mir_transform/src/lib.rs

@@ -89,6 +89,7 @@ mod lint;
 mod lower_intrinsics;
 mod lower_slice_len;
 mod match_branches;
+mod mentioned_items;
 mod multiple_return_terminators;
 mod normalize_array_len;
 mod nrvo;
@@ -566,6 +567,10 @@ fn run_optimization_passes<'tcx>(tcx: TyCtxt<'tcx>, body: &mut Body<'tcx>) {
         tcx,
         body,
         &[
+            // Before doing anything, remember which items are being mentioned so that the set of items
+            // visited does not depend on the optimization level.
+            &mentioned_items::MentionedItems,
+            // Add some UB checks before any UB gets optimized away.
             &check_alignment::CheckAlignment,
             // Before inlining: trim down MIR with passes to reduce inlining work.
 

compiler/rustc_mir_transform/src/mentioned_items.rs

@@ -0,0 +1,117 @@
+use rustc_middle::mir::visit::Visitor;
+use rustc_middle::mir::{self, Location, MentionedItem, MirPass};
+use rustc_middle::ty::{self, adjustment::PointerCoercion, TyCtxt};
+use rustc_session::Session;
+use rustc_span::source_map::Spanned;
+
+pub struct MentionedItems;
+
+struct MentionedItemsVisitor<'a, 'tcx> {
+    tcx: TyCtxt<'tcx>,
+    body: &'a mir::Body<'tcx>,
+    mentioned_items: &'a mut Vec<Spanned<MentionedItem<'tcx>>>,
+}
+
+impl<'tcx> MirPass<'tcx> for MentionedItems {
+    fn is_enabled(&self, _sess: &Session) -> bool {
+        // If this pass is skipped the collector assume that nothing got mentioned! We could
+        // potentially skip it in opt-level 0 if we are sure that opt-level will never *remove* uses
+        // of anything, but that still seems fragile. Furthermore, even debug builds use level 1, so
+        // special-casing level 0 is just not worth it.
+        true
+    }
+
+    fn run_pass(&self, tcx: TyCtxt<'tcx>, body: &mut mir::Body<'tcx>) {
+        debug_assert!(body.mentioned_items.is_empty());
+        let mut mentioned_items = Vec::new();
+        MentionedItemsVisitor { tcx, body, mentioned_items: &mut mentioned_items }.visit_body(body);
+        body.mentioned_items = mentioned_items;
+    }
+}
+
+// This visitor is carefully in sync with the one in `rustc_monomorphize::collector`. We are
+// visiting the exact same places but then instead of monomorphizing and creating `MonoItems`, we
+// have to remain generic and just recording the relevant information in `mentioned_items`, where it
+// will then be monomorphized later during "mentioned items" collection.
+impl<'tcx> Visitor<'tcx> for MentionedItemsVisitor<'_, 'tcx> {
+    fn visit_terminator(&mut self, terminator: &mir::Terminator<'tcx>, location: Location) {
+        self.super_terminator(terminator, location);
+        let span = || self.body.source_info(location).span;
+        match &terminator.kind {
+            mir::TerminatorKind::Call { func, .. } => {
+                let callee_ty = func.ty(self.body, self.tcx);
+                self.mentioned_items
+                    .push(Spanned { node: MentionedItem::Fn(callee_ty), span: span() });
+            }
+            mir::TerminatorKind::Drop { place, .. } => {
+                let ty = place.ty(self.body, self.tcx).ty;
+                self.mentioned_items.push(Spanned { node: MentionedItem::Drop(ty), span: span() });
+            }
+            mir::TerminatorKind::InlineAsm { ref operands, .. } => {
+                for op in operands {
+                    match *op {
+                        mir::InlineAsmOperand::SymFn { ref value } => {
+                            self.mentioned_items.push(Spanned {
+                                node: MentionedItem::Fn(value.const_.ty()),
+                                span: span(),
+                            });
+                        }
+                        _ => {}
+                    }
+                }
+            }
+            _ => {}
+        }
+    }
+
+    fn visit_rvalue(&mut self, rvalue: &mir::Rvalue<'tcx>, location: Location) {
+        self.super_rvalue(rvalue, location);
+        let span = || self.body.source_info(location).span;
+        match *rvalue {
+            // We need to detect unsizing casts that required vtables.
+            mir::Rvalue::Cast(
+                mir::CastKind::PointerCoercion(PointerCoercion::Unsize),
+                ref operand,
+                target_ty,
+            )
+            | mir::Rvalue::Cast(mir::CastKind::DynStar, ref operand, target_ty) => {
+                // This isn't monomorphized yet so we can't tell what the actual types are -- just
+                // add everything that may involve a vtable.
+                let source_ty = operand.ty(self.body, self.tcx);
+                let may_involve_vtable = match (
+                    source_ty.builtin_deref(true).map(|t| t.ty.kind()),
+                    target_ty.builtin_deref(true).map(|t| t.ty.kind()),
+                ) {
+                    (Some(ty::Array(..)), Some(ty::Str | ty::Slice(..))) => false, // &str/&[T] unsizing
+                    _ => true,
+                };
+                if may_involve_vtable {
+                    self.mentioned_items.push(Spanned {
+                        node: MentionedItem::UnsizeCast { source_ty, target_ty },
+                        span: span(),
+                    });
+                }
+            }
+            // Similarly, record closures that are turned into function pointers.
+            mir::Rvalue::Cast(
+                mir::CastKind::PointerCoercion(PointerCoercion::ClosureFnPointer(_)),
+                ref operand,
+                _,
+            ) => {
+                let source_ty = operand.ty(self.body, self.tcx);
+                self.mentioned_items
+                    .push(Spanned { node: MentionedItem::Closure(source_ty), span: span() });
+            }
+            // And finally, function pointer reification casts.
+            mir::Rvalue::Cast(
+                mir::CastKind::PointerCoercion(PointerCoercion::ReifyFnPointer),
+                ref operand,
+                _,
+            ) => {
+                let fn_ty = operand.ty(self.body, self.tcx);
+                self.mentioned_items.push(Spanned { node: MentionedItem::Fn(fn_ty), span: span() });
+            }
+            _ => {}
+        }
+    }
+}

compiler/rustc_mir_transform/src/shim.rs

@@ -17,7 +17,7 @@ use std::iter;
 
 use crate::{
     abort_unwinding_calls, add_call_guards, add_moves_for_packed_drops, deref_separator,
-    pass_manager as pm, remove_noop_landing_pads, simplify,
+    mentioned_items, pass_manager as pm, remove_noop_landing_pads, simplify,
 };
 use rustc_middle::mir::patch::MirPatch;
 use rustc_mir_dataflow::elaborate_drops::{self, DropElaborator, DropFlagMode, DropStyle};
@@ -147,6 +147,7 @@ fn make_shim<'tcx>(tcx: TyCtxt<'tcx>, instance: ty::InstanceDef<'tcx>) -> Body<'
                     tcx,
                     &mut body,
                     &[
+                        &mentioned_items::MentionedItems,
                         &abort_unwinding_calls::AbortUnwindingCalls,
                         &add_call_guards::CriticalCallEdges,
                     ],
@@ -178,6 +179,7 @@ fn make_shim<'tcx>(tcx: TyCtxt<'tcx>, instance: ty::InstanceDef<'tcx>) -> Body<'
         tcx,
         &mut result,
         &[
+            &mentioned_items::MentionedItems,
             &add_moves_for_packed_drops::AddMovesForPackedDrops,
             &deref_separator::Derefer,
             &remove_noop_landing_pads::RemoveNoopLandingPads,