DebugInfo: Include .dwo file name when hashing multiple CUs in a single file

This is really a workaround for ThinLTO in particular - since it can
import partial CUs that may end up looking very similar/the same as
the same partial import in another ThinLTO compile.

An alternative fix would be to change the DICompileUnit metadata to
include a "primary file" or the like - and when importing for ThinLTO
set the primary file to the name of the DICompileUnit that is being
imported into. This involves changing the schema and would reduce the
excessive uniqueness in the hash that this change creates - allowing
diagnosing of more duplicate CUs than will be caught with this change.

But duplicate CUs can still be caught in non-ThinLTO builds & are mostly
a nuisance rather than a particularly deliberate/effective tool for
finding broken code. (arguably the hash could always include the dwo
file and nothing in fission would break, I think..)

llvm-svn: 304119

Author: dwblaikie

llvm/lib/CodeGen/AsmPrinter/DIEHash.cpp

@@ -380,10 +380,12 @@ void DIEHash::computeHash(const DIE &Die) {
 /// DWARF4 standard. It is an md5 hash of the flattened description of the DIE
 /// with the inclusion of the full CU and all top level CU entities.
 // TODO: Initialize the type chain at 0 instead of 1 for CU signatures.
-uint64_t DIEHash::computeCUSignature(const DIE &Die) {
+uint64_t DIEHash::computeCUSignature(StringRef DWOName, const DIE &Die) {
   Numbering.clear();
   Numbering[&Die] = 1;
 
+  if (!DWOName.empty())
+    Hash.update(DWOName);
   // Hash the DIE.
   computeHash(Die);
 

llvm/lib/CodeGen/AsmPrinter/DIEHash.h

@@ -36,7 +36,7 @@ class DIEHash {
   DIEHash(AsmPrinter *A = nullptr) : AP(A) {}
 
   /// \brief Computes the CU signature.
-  uint64_t computeCUSignature(const DIE &Die);
+  uint64_t computeCUSignature(StringRef DWOName, const DIE &Die);
 
   /// \brief Computes the type signature.
   uint64_t computeTypeSignature(const DIE &Die);

llvm/lib/CodeGen/AsmPrinter/DwarfDebug.cpp

@@ -613,6 +613,13 @@ void DwarfDebug::finalizeModuleInfo() {
 
   finishVariableDefinitions();
 
+  // Include the DWO file name in the hash if there's more than one CU.
+  // This handles ThinLTO's situation where imported CUs may very easily be
+  // duplicate with the same CU partially imported into another ThinLTO unit.
+  StringRef DWOName;
+  if (CUMap.size() > 1)
+    DWOName = Asm->TM.Options.MCOptions.SplitDwarfFile;
+
   // Handle anything that needs to be done on a per-unit basis after
   // all other generation.
   for (const auto &P : CUMap) {
@@ -627,7 +634,8 @@ void DwarfDebug::finalizeModuleInfo() {
     auto *SkCU = TheCU.getSkeleton();
     if (useSplitDwarf()) {
       // Emit a unique identifier for this CU.
-      uint64_t ID = DIEHash(Asm).computeCUSignature(TheCU.getUnitDie());
+      uint64_t ID =
+          DIEHash(Asm).computeCUSignature(DWOName, TheCU.getUnitDie());
       TheCU.addUInt(TheCU.getUnitDie(), dwarf::DW_AT_GNU_dwo_id,
                     dwarf::DW_FORM_data8, ID);
       SkCU->addUInt(SkCU->getUnitDie(), dwarf::DW_AT_GNU_dwo_id,

llvm/test/DebugInfo/Generic/split-dwarf-multiple-cu-hash.ll

@@ -0,0 +1,40 @@
+; RUN: %llc_dwarf -split-dwarf-file=foo.dwo  %s -filetype=obj -o %T/a.o
+; RUN: %llc_dwarf -split-dwarf-file=bar.dwo  %s -filetype=obj -o %T/b.o
+; RUN: llvm-dwarfdump -debug-dump=info %T/a.o %T/b.o | FileCheck %s
+
+; CHECK: dwo_id {{.*}}([[HASH:.*]])
+; CHECK-NOT: dwo_id {{.*}}([[HASH]])
+
+; Function Attrs: noinline nounwind uwtable
+define void @_Z1av() #0 !dbg !9 {
+entry:
+  ret void, !dbg !12
+}
+
+; Function Attrs: noinline nounwind uwtable
+define void @_Z1bv() #0 !dbg !13 {
+entry:
+  ret void, !dbg !14
+}
+
+attributes #0 = { noinline nounwind uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!llvm.dbg.cu = !{!0, !3}
+!llvm.ident = !{!5, !5}
+!llvm.module.flags = !{!6, !7, !8}
+
+!0 = distinct !DICompileUnit(language: DW_LANG_C_plus_plus, file: !1, producer: "clang version 5.0.0 (trunk 304107) (llvm/trunk 304109)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !2)
+!1 = !DIFile(filename: "a.cpp", directory: "/usr/local/google/home/blaikie/dev/scratch")
+!2 = !{}
+!3 = distinct !DICompileUnit(language: DW_LANG_C_plus_plus, file: !4, producer: "clang version 5.0.0 (trunk 304107) (llvm/trunk 304109)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !2)
+!4 = !DIFile(filename: "b.cpp", directory: "/usr/local/google/home/blaikie/dev/scratch")
+!5 = !{!"clang version 5.0.0 (trunk 304107) (llvm/trunk 304109)"}
+!6 = !{i32 2, !"Dwarf Version", i32 4}
+!7 = !{i32 2, !"Debug Info Version", i32 3}
+!8 = !{i32 1, !"wchar_size", i32 4}
+!9 = distinct !DISubprogram(name: "a", linkageName: "_Z1av", scope: !1, file: !1, line: 1, type: !10, isLocal: false, isDefinition: true, scopeLine: 1, flags: DIFlagPrototyped, isOptimized: false, unit: !0, variables: !2)
+!10 = !DISubroutineType(types: !11)
+!11 = !{null}
+!12 = !DILocation(line: 2, column: 1, scope: !9)
+!13 = distinct !DISubprogram(name: "b", linkageName: "_Z1bv", scope: !4, file: !4, line: 1, type: !10, isLocal: false, isDefinition: true, scopeLine: 1, flags: DIFlagPrototyped, isOptimized: false, unit: !3, variables: !2)
+!14 = !DILocation(line: 2, column: 1, scope: !13)