DWARF: Fix debug_info references to the abbreviations section (#2997)

The previous code assumed that each compile unit had its own
abbreviation section, and they are all in order. That's normally how
LLVM emits things, but in #2992 there is a testcase in which linking
of object files with IR files somehow ends up with a different order.

The proper fix is to track the binary offsets of abbreviations in the
abbreviation section. That section is comprised of null-terminated
lists, which each CU has an offset to the beginning of. With those
offsets, we can match things properly.

Add a testcase that crashes without this, to prevent regressions.

Fixes #2992
Fixes #3007

3 files changed, 28 insertions, 7 deletions

diff --git a/third_party/llvm-project/DWARFVisitor.cpp b/third_party/llvm-project/DWARFVisitor.cpp
index ba3d1e0c5..1f4403563 100644
--- a/third_party/llvm-project/DWARFVisitor.cpp
+++ b/third_party/llvm-project/DWARFVisitor.cpp
@@ -44,17 +44,26 @@ static unsigned getRefSize(const DWARFYAML::Unit &Unit) {
 }
 
 template <typename T> void DWARFYAML::VisitorImpl<T>::traverseDebugInfo() {
-  // XXX BINARYEN: Handle multiple linked compile units. Each one has its own
-  // range of values, terminated by a zero. AbbrevStart refers to the start
-  // index for the current unit, and AbbrevEnd to one past the last one
-  // (which is the index of the 0 terminator).
+  // XXX BINARYEN: Handle multiple linked compile units, each of which can
+  // refer to a different abbreviation list.
   // TODO: This code appears to assume that abbreviation codes increment by 1
   // so that lookups are linear. In LLVM output that is true, but it might not
   // be in general.
-  size_t AbbrevStart = 0, AbbrevEnd = -1;
   for (auto &Unit : DebugInfo.CompileUnits) {
-    // Skip the 0 terminator.
-    AbbrevEnd = AbbrevStart = AbbrevEnd + 1;
+    // AbbrOffset is the byte offset into the abbreviation section, which we
+    // need to find among the Abbrev's ListOffsets (which are the byte offsets
+    // of where that abbreviation list begins).
+    // TODO: Optimize this to not be O(#CUs * #abbrevs).
+    size_t AbbrevStart = -1;
+    for (size_t i = 0; i < DebugInfo.AbbrevDecls.size(); i++) {
+      if (DebugInfo.AbbrevDecls[i].ListOffset == Unit.AbbrOffset) {
+        AbbrevStart = i;
+        break;
+      }
+    }
+    assert(AbbrevStart != -1); // must find the list
+    // Find the last entry in this abbreviation list.
+    size_t AbbrevEnd = AbbrevStart;
     while (AbbrevEnd < DebugInfo.AbbrevDecls.size() &&
            DebugInfo.AbbrevDecls[AbbrevEnd].Code) {
       AbbrevEnd++;
diff --git a/third_party/llvm-project/dwarf2yaml.cpp b/third_party/llvm-project/dwarf2yaml.cpp
index 5e22e1989..7f160f7e5 100644
--- a/third_party/llvm-project/dwarf2yaml.cpp
+++ b/third_party/llvm-project/dwarf2yaml.cpp
@@ -27,6 +27,7 @@ void dumpDebugAbbrev(DWARFContext &DCtx, DWARFYAML::Data &Y) {
   auto AbbrevSetPtr = DCtx.getDebugAbbrev();
   if (AbbrevSetPtr) {
     for (auto AbbrvDeclSet : *AbbrevSetPtr) {
+      auto ListOffset = AbbrvDeclSet.second.getOffset();
       for (auto AbbrvDecl : AbbrvDeclSet.second) {
         DWARFYAML::Abbrev Abbrv;
         Abbrv.Code = AbbrvDecl.getCode();
@@ -41,6 +42,7 @@ void dumpDebugAbbrev(DWARFContext &DCtx, DWARFYAML::Data &Y) {
             AttAbrv.Value = Attribute.getImplicitConstValue();
           Abbrv.Attributes.push_back(AttAbrv);
         }
+        Abbrv.ListOffset = ListOffset;
         Y.AbbrevDecls.push_back(Abbrv);
       }
       // XXX BINARYEN: null-terminate the DeclSet. This is needed to separate
diff --git a/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h b/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h
index bd4aa2007..818bddede 100644
--- a/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h
+++ b/third_party/llvm-project/include/llvm/ObjectYAML/DWARFYAML.h
@@ -55,6 +55,16 @@ struct Abbrev {
   llvm::dwarf::Tag Tag;
   llvm::dwarf::Constants Children;
   std::vector<AttributeAbbrev> Attributes;
+  // XXX BINARYEN: Represent the binary offset in the abbreviation section for
+  // this abbreviation's list. The abbreviation section has multiple lists,
+  // each null-terminated, and those lists are what are referred to by compile
+  // units by offset. We need to match the offset in a compile unit to the
+  // abbreviation at that offset (which must be the beginning of an
+  // abbreviation list, that is, either the very first element, or after a null
+  // terminator). All abbreviations in the same list have the same offset
+  // (DWARFAbbreviationDeclarationSet does not track anything else, and we don't
+  // need it).
+  uint64_t ListOffset;
 };
 
 struct ARangeDescriptor {