Auto merge of rust-lang#123272 - saethlin:reachable-mono-cleanup, r=<try>

Only collect mono items from reachable blocks

Fixes the wrong commented pointed out in: rust-lang#121421 (comment)
Moves the analysis to use the worklist strategy: rust-lang#121421 (comment)
Also fixes rust-lang#85836, using the same reachability analysis

Author: bors

compiler/rustc_codegen_ssa/src/mir/mod.rs

@@ -257,20 +257,20 @@ pub fn codegen_mir<'a, 'tcx, Bx: BuilderMethods<'a, 'tcx>>(
     // Apply debuginfo to the newly allocated locals.
     fx.debug_introduce_locals(&mut start_bx);
 
-    let reachable_blocks = mir.reachable_blocks_in_mono(cx.tcx(), instance);
-
     // The builders will be created separately for each basic block at `codegen_block`.
     // So drop the builder of `start_llbb` to avoid having two at the same time.
     drop(start_bx);
 
+    let reachable_blocks =
+        rustc_middle::mir::traversal::mono_reachable_bitset(mir, cx.tcx(), instance);
+
     // Codegen the body of each block using reverse postorder
     for (bb, _) in traversal::reverse_postorder(mir) {
         if reachable_blocks.contains(bb) {
             fx.codegen_block(bb);
         } else {
-            // This may have references to things we didn't monomorphize, so we
-            // don't actually codegen the body. We still create the block so
-            // terminators in other blocks can reference it without worry.
+            // We want to skip this block, because it's not reachable. But we still create
+            // the block so terminators in other blocks can reference it.
             fx.codegen_block_as_unreachable(bb);
         }
     }

compiler/rustc_middle/src/mir/mod.rs

@@ -30,7 +30,6 @@ pub use rustc_ast::Mutability;
 use rustc_data_structures::fx::FxHashMap;
 use rustc_data_structures::fx::FxHashSet;
 use rustc_data_structures::graph::dominators::Dominators;
-use rustc_data_structures::stack::ensure_sufficient_stack;
 use rustc_index::bit_set::BitSet;
 use rustc_index::{Idx, IndexSlice, IndexVec};
 use rustc_serialize::{Decodable, Encodable};
@@ -687,57 +686,6 @@ impl<'tcx> Body<'tcx> {
         self.injection_phase.is_some()
     }
 
-    /// Finds which basic blocks are actually reachable for a specific
-    /// monomorphization of this body.
-    ///
-    /// This is allowed to have false positives; just because this says a block
-    /// is reachable doesn't mean that's necessarily true. It's thus always
-    /// legal for this to return a filled set.
-    ///
-    /// Regardless, the [`BitSet::domain_size`] of the returned set will always
-    /// exactly match the number of blocks in the body so that `contains`
-    /// checks can be done without worrying about panicking.
-    ///
-    /// This is mostly useful because it lets us skip lowering the `false` side
-    /// of `if <T as Trait>::CONST`, as well as `intrinsics::debug_assertions`.
-    pub fn reachable_blocks_in_mono(
-        &self,
-        tcx: TyCtxt<'tcx>,
-        instance: Instance<'tcx>,
-    ) -> BitSet<BasicBlock> {
-        let mut set = BitSet::new_empty(self.basic_blocks.len());
-        self.reachable_blocks_in_mono_from(tcx, instance, &mut set, START_BLOCK);
-        set
-    }
-
-    fn reachable_blocks_in_mono_from(
-        &self,
-        tcx: TyCtxt<'tcx>,
-        instance: Instance<'tcx>,
-        set: &mut BitSet<BasicBlock>,
-        bb: BasicBlock,
-    ) {
-        if !set.insert(bb) {
-            return;
-        }
-
-        let data = &self.basic_blocks[bb];
-
-        if let Some((bits, targets)) = Self::try_const_mono_switchint(tcx, instance, data) {
-            let target = targets.target_for_value(bits);
-            ensure_sufficient_stack(|| {
-                self.reachable_blocks_in_mono_from(tcx, instance, set, target)
-            });
-            return;
-        }
-
-        for target in data.terminator().successors() {
-            ensure_sufficient_stack(|| {
-                self.reachable_blocks_in_mono_from(tcx, instance, set, target)
-            });
-        }
-    }
-
     /// If this basic block ends with a [`TerminatorKind::SwitchInt`] for which we can evaluate the
     /// dimscriminant in monomorphization, we return the discriminant bits and the
     /// [`SwitchTargets`], just so the caller doesn't also have to match on the terminator.

compiler/rustc_middle/src/mir/traversal.rs

@@ -279,3 +279,87 @@ pub fn reverse_postorder<'a, 'tcx>(
 {
     body.basic_blocks.reverse_postorder().iter().map(|&bb| (bb, &body.basic_blocks[bb]))
 }
+
+#[derive(Clone)]
+pub struct MonoReachable<'a, 'tcx> {
+    body: &'a Body<'tcx>,
+    tcx: TyCtxt<'tcx>,
+    instance: Instance<'tcx>,
+    visited: BitSet<BasicBlock>,
+    worklist: Vec<BasicBlock>,
+}
+
+impl<'a, 'tcx> MonoReachable<'a, 'tcx> {
+    pub fn new(
+        body: &'a Body<'tcx>,
+        tcx: TyCtxt<'tcx>,
+        instance: Instance<'tcx>,
+    ) -> MonoReachable<'a, 'tcx> {
+        MonoReachable {
+            body,
+            tcx,
+            instance,
+            visited: BitSet::new_empty(body.basic_blocks.len()),
+            worklist: vec![START_BLOCK],
+        }
+    }
+}
+
+/// Finds which basic blocks are actually reachable for a monomorphized [`Instance`].
+///
+/// This is allowed to have false positives; just because this says a block
+/// is reachable doesn't mean that's necessarily true. It's thus always
+/// legal for this to return a filled set.
+///
+/// Regardless, the [`BitSet::domain_size`] of the returned set will always
+/// exactly match the number of blocks in the body so that `contains`
+/// checks can be done without worrying about panicking.
+///
+/// This is mostly useful because it lets us skip lowering the `false` side
+/// of `if <T as Trait>::CONST`, as well as [`NullOp::UbChecks`].
+///
+/// [`NullOp::UbChecks`]: rustc_middle::mir::NullOp::UbChecks
+pub fn mono_reachable<'a, 'tcx>(
+    body: &'a Body<'tcx>,
+    tcx: TyCtxt<'tcx>,
+    instance: Instance<'tcx>,
+) -> MonoReachable<'a, 'tcx> {
+    MonoReachable::new(body, tcx, instance)
+}
+
+pub fn mono_reachable_bitset<'a, 'tcx>(
+    body: &'a Body<'tcx>,
+    tcx: TyCtxt<'tcx>,
+    instance: Instance<'tcx>,
+) -> BitSet<BasicBlock> {
+    let mut it = MonoReachable::new(body, tcx, instance);
+    while let Some(_) = it.next() {}
+    it.visited
+}
+
+impl<'a, 'tcx> Iterator for MonoReachable<'a, 'tcx> {
+    type Item = (BasicBlock, &'a BasicBlockData<'tcx>);
+
+    fn next(&mut self) -> Option<(BasicBlock, &'a BasicBlockData<'tcx>)> {
+        while let Some(idx) = self.worklist.pop() {
+            if !self.visited.insert(idx) {
+                continue;
+            }
+
+            let data = &self.body[idx];
+
+            if let Some((bits, targets)) =
+                Body::try_const_mono_switchint(self.tcx, self.instance, data)
+            {
+                let target = targets.target_for_value(bits);
+                self.worklist.push(target);
+            } else {
+                self.worklist.extend(data.terminator().successors());
+            }
+
+            return Some((idx, data));
+        }
+
+        None
+    }
+}

compiler/rustc_monomorphize/src/collector.rs

@@ -1414,15 +1414,16 @@ fn collect_items_of_instance<'tcx>(
     };
 
     if mode == CollectionMode::UsedItems {
-        // Visit everything. Here we rely on the visitor also visiting `required_consts`, so that we
-        // evaluate them and abort compilation if any of them errors.
-        collector.visit_body(body);
-    } else {
-        // We only need to evaluate all constants, but can ignore the rest of the MIR.
-        for const_op in &body.required_consts {
-            if let Some(val) = collector.eval_constant(const_op) {
-                collect_const_value(tcx, val, mentioned_items);
-            }
+        for (bb, data) in body.basic_blocks.iter_enumerated() {
+            collector.visit_basic_block_data(bb, data)
+        }
+    }
+
+    // Always visit all `required_consts`, so that we evaluate them and abort compilation if any of
+    // them errors.
+    for const_op in &body.required_consts {
+        if let Some(val) = collector.eval_constant(const_op) {
+            collect_const_value(tcx, val, mentioned_items);
         }
     }