[BranchFolding] Merge debug locations from common tail instead of removing

Summary: D25742 improved the precision of debug locations for PGO by removing debug locations from common tail when tail-merging. However, if identical insturctions that are merged into a common tail have the same debug locations, there's no need to remove them. This patch creates a merged debug location of identical instructions across SameTails and assign it to the instruction in the common tail, so that the debug locations are maintained if they are same across identical instructions.

Reviewers: aprantl, probinson, MatzeB, rob.lougher

Reviewed By: aprantl

Subscribers: andreadb, llvm-commits

Differential Revision: https://reviews.llvm.org/D30226

llvm-svn: 297805

lib/CodeGen/BranchFolding.cpp

@@ -32,6 +32,7 @@
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/CodeGen/Passes.h"
 #include "llvm/CodeGen/TargetPassConfig.h"
+#include "llvm/IR/DebugInfoMetadata.h"
 #include "llvm/IR/Function.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/Debug.h"
@@ -753,6 +754,45 @@ bool BranchFolder::CreateCommonTailOnlyBlock(MachineBasicBlock *&PredBB,
   return true;
 }
 
+/// MergeCommonTailDebugLocs - Create merged DebugLocs of identical instructions
+/// across SameTails and assign it to the instruction in common tail.
+void BranchFolder::MergeCommonTailDebugLocs(unsigned commonTailIndex) {
+  MachineBasicBlock *MBB = SameTails[commonTailIndex].getBlock();
+
+  std::vector<MachineBasicBlock::iterator> NextCommonInsts(SameTails.size());
+  for (unsigned int i = 0 ; i != SameTails.size() ; ++i) {
+    if (i != commonTailIndex)
+      NextCommonInsts[i] = SameTails[i].getTailStartPos();
+    else {
+      assert(SameTails[i].getTailStartPos() == MBB->begin() &&
+          "MBB is not a common tail only block");
+    }
+  }
+
+  for (auto &MI : *MBB) {
+    if (MI.isDebugValue())
+      continue;
+    DebugLoc DL = MI.getDebugLoc();
+    for (unsigned int i = 0 ; i < NextCommonInsts.size() ; i++) {
+      if (i == commonTailIndex)
+        continue;
+
+      auto &Pos = NextCommonInsts[i];
+      assert(Pos != SameTails[i].getBlock()->end() &&
+          "Reached BB end within common tail");
+      while (Pos->isDebugValue()) {
+        ++Pos;
+        assert(Pos != SameTails[i].getBlock()->end() &&
+            "Reached BB end within common tail");
+      }
+      assert(MI.isIdenticalTo(*Pos) && "Expected matching MIIs!");
+      DL = DILocation::getMergedLocation(DL, Pos->getDebugLoc());
+      NextCommonInsts[i] = ++Pos;
+    }
+    MI.setDebugLoc(DL);
+  }
+}
+
 static void
 mergeOperations(MachineBasicBlock::iterator MBBIStartPos,
                 MachineBasicBlock &MBBCommon) {
@@ -905,10 +945,8 @@ bool BranchFolder::TryTailMergeBlocks(MachineBasicBlock *SuccBB,
     // Recompute common tail MBB's edge weights and block frequency.
     setCommonTailEdgeWeights(*MBB);
 
-    // Remove the original debug location from the common tail.
-    for (auto &MI : *MBB)
-      if (!MI.isDebugValue())
-        MI.setDebugLoc(DebugLoc());
+    // Merge debug locations across identical instructions for common tail
+    MergeCommonTailDebugLocs(commonTailIndex);
 
     // MBB is common tail.  Adjust all other BB's to jump to this one.
     // Traversal must be forwards so erases work.

lib/CodeGen/BranchFolding.h

@@ -153,6 +153,7 @@ namespace llvm {
                                    MachineBasicBlock *SuccBB,
                                    unsigned maxCommonTailLength,
                                    unsigned &commonTailIndex);
+    void MergeCommonTailDebugLocs(unsigned commonTailIndex);
 
     bool OptimizeBranches(MachineFunction &MF);
     bool OptimizeBlock(MachineBasicBlock *MBB);

test/CodeGen/X86/tail-merge-debugloc.ll

@@ -0,0 +1,42 @@
+; RUN: llc -stop-after=branch-folder < %s | FileCheck %s
+;
+; bb2 and bb3 in the IR below will be tail-merged into a single basic block.
+; As br instructions in bb2 and bb3 have the same debug location, make sure that
+; the branch instruction in the merged basic block still maintains the debug 
+; location info.
+; 
+; CHECK:      [[DLOC:![0-9]+]] = !DILocation(line: 2, column: 2, scope: !{{[0-9]+}})
+; CHECK:      TEST64rr{{.*}}%rsi, %rsi, implicit-def %eflags
+; CHECK-NEXT: JNE_1{{.*}}, debug-location [[DLOC]]
+
+target triple = "x86_64-unknown-linux-gnu"
+
+define i32 @foo(i1 %b, i8* %p) {
+bb1: 
+  br i1 %b, label %bb2, label %bb3
+
+bb2:
+  %a1 = icmp eq i8* %p, null
+  br i1 %a1, label %bb4, label %bb5, !dbg !6
+  
+bb3:
+  %a2 = icmp eq i8* %p, null
+  br i1 %a2, label %bb4, label %bb5, !dbg !6
+
+bb4:
+  ret i32 1
+
+bb5:
+  ret i32 0
+}
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!2, !3}
+
+!0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1)
+!1 = !DIFile(filename: "foo.c", directory: "b/")
+!2 = !{i32 2, !"Dwarf Version", i32 4}
+!3 = !{i32 2, !"Debug Info Version", i32 3}
+!4 = distinct !DISubprogram(name: "foo", scope: !1, file: !1, line: 3, isLocal: false, isDefinition: true, scopeLine: 3, flags: DIFlagPrototyped, isOptimized: true, unit: !0)
+!5 = distinct !DILexicalBlock(scope: !4, file: !1, line: 1, column: 1)
+!6 = !DILocation(line: 2, column: 2, scope: !5)