Add EndRegion statement kind to MIR.

 * Emit `EndRegion` for every code-extent for which we observe a
   borrow. To do this, we needed to thread source info back through
   to `fn in_scope`, which makes this commit a bit more painful than
   one might have expected.

 * There is `end_region` emission in `Builder::pop_scope` and in
   `Builder::exit_scope`; the first handles falling out of a scope
   normally, the second handles e.g. `break`.

 * Remove `EndRegion` statements during the erase_regions mir
   transformation.

 * Preallocate the terminator block, and throw an `Unreachable` marker
   on it from the outset. Then overwrite that Terminator as necessary
   on demand.

 * Instead of marking the scope as needs_cleanup after seeing a
   borrow, just treat every scope in the chain as being part of the
   diverge_block (after any *one* of them has separately signalled
   that it needs cleanup, e.g. due to having a destructor to run).

 * Allow for resume terminators to be patched when looking up drop flags.

   (In particular, `MirPatch::new` has an explicit code path,
   presumably previously unreachable, that patches up such resume
   terminators.)

 * Make `Scope` implement `Debug` trait.

 * Expanded a stray comment: we do not emit StorageDead on diverging
   paths, but that end behavior might not be desirable.

Author: pnkfelix

src/librustc/ich/impls_mir.rs

@@ -226,6 +226,9 @@ for mir::StatementKind<'tcx> {
             mir::StatementKind::StorageDead(ref lvalue) => {
                 lvalue.hash_stable(hcx, hasher);
             }
+            mir::StatementKind::EndRegion(ref extents) => {
+                extents.hash_stable(hcx, hasher);
+            }
             mir::StatementKind::Nop => {}
             mir::StatementKind::InlineAsm { ref asm, ref outputs, ref inputs } => {
                 asm.hash_stable(hcx, hasher);

src/librustc/mir/mod.rs

@@ -12,6 +12,7 @@
 
 use graphviz::IntoCow;
 use middle::const_val::ConstVal;
+use middle::region::CodeExtent;
 use rustc_const_math::{ConstUsize, ConstInt, ConstMathErr};
 use rustc_data_structures::indexed_vec::{IndexVec, Idx};
 use rustc_data_structures::control_flow_graph::dominators::{Dominators, dominators};
@@ -804,6 +805,10 @@ pub enum StatementKind<'tcx> {
         inputs: Vec<Operand<'tcx>>
     },
 
+    /// Mark one terminating point of an extent (i.e. static region).
+    /// (The starting point(s) arise implicitly from borrows.)
+    EndRegion(CodeExtent),
+
     /// No-op. Useful for deleting instructions without affecting statement indices.
     Nop,
 }
@@ -813,6 +818,8 @@ impl<'tcx> Debug for Statement<'tcx> {
         use self::StatementKind::*;
         match self.kind {
             Assign(ref lv, ref rv) => write!(fmt, "{:?} = {:?}", lv, rv),
+            // (reuse lifetime rendering policy from ppaux.)
+            EndRegion(ref ce) => write!(fmt, "EndRegion({})", ty::ReScope(*ce)),
             StorageLive(ref lv) => write!(fmt, "StorageLive({:?})", lv),
             StorageDead(ref lv) => write!(fmt, "StorageDead({:?})", lv),
             SetDiscriminant{lvalue: ref lv, variant_index: index} => {
@@ -1472,6 +1479,13 @@ impl<'tcx> TypeFoldable<'tcx> for Statement<'tcx> {
                 outputs: outputs.fold_with(folder),
                 inputs: inputs.fold_with(folder)
             },
+
+            // Note for future: If we want to expose the extents
+            // during the fold, we need to either generalize EndRegion
+            // to carry `[ty::Region]`, or extend the `TypeFolder`
+            // trait with a `fn fold_extent`.
+            EndRegion(ref extent) => EndRegion(extent.clone()),
+
             Nop => Nop,
         };
         Statement {
@@ -1490,6 +1504,13 @@ impl<'tcx> TypeFoldable<'tcx> for Statement<'tcx> {
             StorageDead(ref lvalue) => lvalue.visit_with(visitor),
             InlineAsm { ref outputs, ref inputs, .. } =>
                 outputs.visit_with(visitor) || inputs.visit_with(visitor),
+
+            // Note for future: If we want to expose the extents
+            // during the visit, we need to either generalize EndRegion
+            // to carry `[ty::Region]`, or extend the `TypeVisitor`
+            // trait with a `fn visit_extent`.
+            EndRegion(ref _extent) => false,
+
             Nop => false,
         }
     }

src/librustc/mir/visit.rs

@@ -325,6 +325,7 @@ macro_rules! make_mir_visitor {
                                           ref $($mutability)* rvalue) => {
                         self.visit_assign(block, lvalue, rvalue, location);
                     }
+                    StatementKind::EndRegion(_) => {}
                     StatementKind::SetDiscriminant{ ref $($mutability)* lvalue, .. } => {
                         self.visit_lvalue(lvalue, LvalueContext::Store, location);
                     }

src/librustc_borrowck/borrowck/mir/dataflow/impls.rs

@@ -474,6 +474,7 @@ impl<'a, 'tcx> BitDenotation for MovingOutStatements<'a, 'tcx> {
             mir::StatementKind::StorageLive(_) |
             mir::StatementKind::StorageDead(_) |
             mir::StatementKind::InlineAsm { .. } |
+            mir::StatementKind::EndRegion(_) |
             mir::StatementKind::Nop => {}
         }
     }

src/librustc_borrowck/borrowck/mir/dataflow/sanity_check.rs

@@ -105,6 +105,7 @@ fn each_block<'a, 'tcx, O>(tcx: TyCtxt<'a, 'tcx, 'tcx>,
             mir::StatementKind::StorageLive(_) |
             mir::StatementKind::StorageDead(_) |
             mir::StatementKind::InlineAsm { .. } |
+            mir::StatementKind::EndRegion(_) |
             mir::StatementKind::Nop => continue,
             mir::StatementKind::SetDiscriminant{ .. } =>
                 span_bug!(stmt.source_info.span,

src/librustc_borrowck/borrowck/mir/elaborate_drops.rs

@@ -585,6 +585,11 @@ impl<'b, 'tcx> ElaborateDropsCtxt<'b, 'tcx> {
                             // drop elaboration should handle that by itself
                             continue
                         }
+                        TerminatorKind::Resume => {
+                            // We can replace resumes with gotos
+                            // jumping to a canonical resume.
+                            continue
+                        }
                         TerminatorKind::DropAndReplace { .. } => {
                             // this contains the move of the source and
                             // the initialization of the destination. We

src/librustc_borrowck/borrowck/mir/gather_moves.rs

@@ -413,6 +413,7 @@ impl<'a, 'tcx> MoveDataBuilder<'a, 'tcx> {
                           "SetDiscriminant should not exist during borrowck");
             }
             StatementKind::InlineAsm { .. } |
+            StatementKind::EndRegion(_) |
             StatementKind::Nop => {}
         }
     }

src/librustc_borrowck/borrowck/mir/mod.rs

@@ -394,6 +394,7 @@ fn drop_flag_effects_for_location<'a, 'tcx, F>(
             mir::StatementKind::StorageLive(_) |
             mir::StatementKind::StorageDead(_) |
             mir::StatementKind::InlineAsm { .. } |
+            mir::StatementKind::EndRegion(_) |
             mir::StatementKind::Nop => {}
         },
         None => {

src/librustc_mir/build/block.rs

@@ -71,7 +71,7 @@ impl<'a, 'gcx, 'tcx> Builder<'a, 'gcx, 'tcx> {
         let outer_visibility_scope = this.visibility_scope;
         let source_info = this.source_info(span);
         for stmt in stmts {
-            let Stmt { span: _, kind, opt_destruction_extent } = this.hir.mirror(stmt);
+            let Stmt { span, kind, opt_destruction_extent } = this.hir.mirror(stmt);
             match kind {
                 StmtKind::Expr { scope, expr } => {
                     unpack!(block = this.in_opt_scope(
@@ -122,7 +122,6 @@ impl<'a, 'gcx, 'tcx> Builder<'a, 'gcx, 'tcx> {
         if let Some(expr) = expr {
             unpack!(block = this.into(destination, block, expr));
         } else {
-            let source_info = this.source_info(span);
             this.cfg.push_assign_unit(block, source_info, destination);
         }
         // Finally, we pop all the let scopes before exiting out from the scope of block

src/librustc_mir/build/cfg.rs

@@ -14,6 +14,7 @@
 //! Routines for manipulating the control-flow graph.
 
 use build::CFG;
+use rustc::middle::region::CodeExtent;
 use rustc::mir::*;
 
 impl<'tcx> CFG<'tcx> {
@@ -43,6 +44,16 @@ impl<'tcx> CFG<'tcx> {
         self.block_data_mut(block).statements.push(statement);
     }
 
+    pub fn push_end_region(&mut self,
+                           block: BasicBlock,
+                           source_info: SourceInfo,
+                           extent: CodeExtent) {
+        self.push(block, Statement {
+            source_info: source_info,
+            kind: StatementKind::EndRegion(extent),
+        });
+    }
+
     pub fn push_assign(&mut self,
                        block: BasicBlock,
                        source_info: SourceInfo,

src/librustc_mir/build/expr/as_temp.rs

@@ -49,7 +49,6 @@ impl<'a, 'gcx, 'tcx> Builder<'a, 'gcx, 'tcx> {
 
         let expr_ty = expr.ty.clone();
         let temp = this.temp(expr_ty.clone(), expr_span);
-        let source_info = this.source_info(expr_span);
 
         if !expr_ty.is_never() && temp_lifetime.is_some() {
             this.cfg.push(block, Statement {

src/librustc_mir/build/scope.rs

@@ -94,10 +94,11 @@ use rustc::ty::subst::{Kind, Subst};
 use rustc::ty::{Ty, TyCtxt};
 use rustc::mir::*;
 use rustc::mir::transform::MirSource;
-use syntax_pos::Span;
+use syntax_pos::{Span};
 use rustc_data_structures::indexed_vec::Idx;
 use rustc_data_structures::fx::FxHashMap;
 
+#[derive(Debug)]
 pub struct Scope<'tcx> {
     /// The visibility scope this scope was created in.
     visibility_scope: VisibilityScope,
@@ -114,7 +115,7 @@ pub struct Scope<'tcx> {
     ///  * pollutting the cleanup MIR with StorageDead creates
     ///    landing pads even though there's no actual destructors
     ///  * freeing up stack space has no effect during unwinding
-    needs_cleanup: bool,
+    pub(super) needs_cleanup: bool,
 
     /// set of lvalues to drop when exiting this scope. This starts
     /// out empty but grows as variables are declared during the
@@ -141,6 +142,7 @@ pub struct Scope<'tcx> {
     cached_exits: FxHashMap<(BasicBlock, CodeExtent), BasicBlock>,
 }
 
+#[derive(Debug)]
 struct DropData<'tcx> {
     /// span where drop obligation was incurred (typically where lvalue was declared)
     span: Span,
@@ -152,6 +154,7 @@ struct DropData<'tcx> {
     kind: DropKind
 }
 
+#[derive(Debug)]
 enum DropKind {
     Value {
         /// The cached block for the cleanups-on-diverge path. This block
@@ -163,6 +166,7 @@ enum DropKind {
     Storage
 }
 
+#[derive(Debug)]
 struct FreeData<'tcx> {
     /// span where free obligation was incurred
     span: Span,
@@ -338,6 +342,8 @@ impl<'a, 'gcx, 'tcx> Builder<'a, 'gcx, 'tcx> {
                                           &self.scopes,
                                           block,
                                           self.arg_count));
+
+        self.cfg.push_end_region(block, extent.1, scope.extent);
         block.unit()
     }
 
@@ -379,6 +385,10 @@ impl<'a, 'gcx, 'tcx> Builder<'a, 'gcx, 'tcx> {
                                               rest,
                                               block,
                                               self.arg_count));
+
+            // End all regions for scopes out of which we are breaking.
+            self.cfg.push_end_region(block, extent.1, scope.extent);
+
             if let Some(ref free_data) = scope.free {
                 let next = self.cfg.start_new_block();
                 let free = build_free(self.hir.tcx(), &tmp, free_data, next);
@@ -640,7 +650,7 @@ impl<'a, 'gcx, 'tcx> Builder<'a, 'gcx, 'tcx> {
             resumeblk
         };
 
-        for scope in scopes.iter_mut().filter(|s| s.needs_cleanup) {
+        for scope in scopes.iter_mut() {
             target = build_diverge_scope(hir.tcx(), cfg, &unit_temp, span, scope, target);
         }
         Some(target)
@@ -775,9 +785,9 @@ fn build_diverge_scope<'a, 'gcx, 'tcx>(tcx: TyCtxt<'a, 'gcx, 'tcx>,
     // Build up the drops in **reverse** order. The end result will
     // look like:
     //
-    //    [drops[n]] -...-> [drops[0]] -> [Free] -> [target]
-    //    |                                    |
-    //    +------------------------------------+
+    //    [EndRegion Block] -> [drops[n]] -...-> [drops[0]] -> [Free] -> [target]
+    //    |                                                         |
+    //    +---------------------------------------------------------+
     //     code for scope
     //
     // The code in this function reads from right to left. At each
@@ -807,9 +817,16 @@ fn build_diverge_scope<'a, 'gcx, 'tcx>(tcx: TyCtxt<'a, 'gcx, 'tcx>,
     // Next, build up the drops. Here we iterate the vector in
     // *forward* order, so that we generate drops[0] first (right to
     // left in diagram above).
-    for drop_data in &mut scope.drops {
+    for (j, drop_data) in scope.drops.iter_mut().enumerate() {
+        debug!("build_diverge_scope drop_data[{}]: {:?}", j, drop_data);
         // Only full value drops are emitted in the diverging path,
         // not StorageDead.
+        //
+        // Note: This may not actually be what we desire (are we
+        // "freeing" stack storage as we unwind, or merely observing a
+        // frozen stack)? In particular, the intent may have been to
+        // match the behavior of clang, but on inspection eddyb says
+        // this is not what clang does.
         let cached_block = match drop_data.kind {
             DropKind::Value { ref mut cached_block } => cached_block,
             DropKind::Storage => continue
@@ -829,6 +846,15 @@ fn build_diverge_scope<'a, 'gcx, 'tcx>(tcx: TyCtxt<'a, 'gcx, 'tcx>,
         };
     }
 
+    // Finally, push the EndRegion block, used by mir-borrowck. (Block
+    // becomes trivial goto after pass that removes all EndRegions.)
+    {
+        let block = cfg.start_new_cleanup_block();
+        cfg.push_end_region(block, source_info(span), scope.extent);
+        cfg.terminate(block, source_info(span), TerminatorKind::Goto { target: target });
+        target = block
+    }
+
     target
 }
 

src/librustc_mir/transform/erase_regions.rs

@@ -65,6 +65,15 @@ impl<'a, 'tcx> MutVisitor<'tcx> for EraseRegionsVisitor<'a, 'tcx> {
                             substs: &mut ClosureSubsts<'tcx>) {
         *substs = self.tcx.erase_regions(substs);
     }
+
+    fn visit_statement(&mut self,
+                       _block: BasicBlock,
+                       statement: &mut Statement<'tcx>,
+                       _location: Location) {
+        if let StatementKind::EndRegion(_) = statement.kind {
+            statement.kind = StatementKind::Nop;
+        }
+    }
 }
 
 pub struct EraseRegions;

src/librustc_mir/transform/qualify_consts.rs

@@ -907,6 +907,7 @@ impl<'a, 'tcx> Visitor<'tcx> for Qualifier<'a, 'tcx, 'tcx> {
                 StatementKind::StorageLive(_) |
                 StatementKind::StorageDead(_) |
                 StatementKind::InlineAsm {..} |
+                StatementKind::EndRegion(_) |
                 StatementKind::Nop => {}
             }
         });

src/librustc_mir/transform/type_check.rs

@@ -413,6 +413,7 @@ impl<'a, 'gcx, 'tcx> TypeChecker<'a, 'gcx, 'tcx> {
                 }
             }
             StatementKind::InlineAsm { .. } |
+            StatementKind::EndRegion(_) |
             StatementKind::Nop => {}
         }
     }

src/librustc_mir/util/patch.rs

@@ -46,6 +46,7 @@ impl<'tcx> MirPatch<'tcx> {
         for (bb, block) in mir.basic_blocks().iter_enumerated() {
             if let TerminatorKind::Resume = block.terminator().kind {
                 if block.statements.len() > 0 {
+                    assert!(resume_stmt_block.is_none());
                     resume_stmt_block = Some(bb);
                 } else {
                     resume_block = Some(bb);

src/librustc_passes/mir_stats.rs

@@ -125,6 +125,7 @@ impl<'a, 'tcx> mir_visit::Visitor<'tcx> for StatCollector<'a, 'tcx> {
         self.record("Statement", statement);
         self.record(match statement.kind {
             StatementKind::Assign(..) => "StatementKind::Assign",
+            StatementKind::EndRegion(..) => "StatementKind::EndRegion",
             StatementKind::SetDiscriminant { .. } => "StatementKind::SetDiscriminant",
             StatementKind::StorageLive(..) => "StatementKind::StorageLive",
             StatementKind::StorageDead(..) => "StatementKind::StorageDead",

src/librustc_trans/mir/constant.rs

@@ -284,6 +284,7 @@ impl<'a, 'tcx> MirConstContext<'a, 'tcx> {
                     }
                     mir::StatementKind::StorageLive(_) |
                     mir::StatementKind::StorageDead(_) |
+                    mir::StatementKind::EndRegion(_) |
                     mir::StatementKind::Nop => {}
                     mir::StatementKind::InlineAsm { .. } |
                     mir::StatementKind::SetDiscriminant{ .. } => {

src/librustc_trans/mir/statement.rs

@@ -86,6 +86,7 @@ impl<'a, 'tcx> MirContext<'a, 'tcx> {
                 asm::trans_inline_asm(&bcx, asm, outputs, input_vals);
                 bcx
             }
+            mir::StatementKind::EndRegion(_) |
             mir::StatementKind::Nop => bcx,
         }
     }