llvm-dwarfdump: Skip tombstoned address ranges

Make the dumper & API a bit more informative by using the new tombstone
addresses to filter out or otherwise render more explicitly dead code
ranges.

Author: dwblaikie

lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp

@@ -946,7 +946,7 @@ DWARFUnit::FindRnglistFromOffset(dw_offset_t offset) {
   llvm::Expected<llvm::DWARFAddressRangesVector> llvm_ranges =
       range_list_or_error->getAbsoluteRanges(
           llvm::object::SectionedAddress{GetBaseAddress()},
-          [&](uint32_t index) {
+          GetAddressByteSize(), [&](uint32_t index) {
             uint32_t index_size = GetAddressByteSize();
             dw_offset_t addr_base = GetAddrBase();
             lldb::offset_t offset = addr_base + index * index_size;

llvm/include/llvm/BinaryFormat/Dwarf.h

@@ -27,6 +27,8 @@
 #include "llvm/Support/FormatVariadicDetails.h"
 #include "llvm/ADT/Triple.h"
 
+#include <limits>
+
 namespace llvm {
 class StringRef;
 
@@ -745,6 +747,11 @@ template <> struct EnumTraits<LocationAtom> : public std::true_type {
   static constexpr char Type[3] = "OP";
   static constexpr StringRef (*StringFn)(unsigned) = &OperationEncodingString;
 };
+
+inline uint64_t computeTombstoneAddress(uint8_t AddressByteSize) {
+  return std::numeric_limits<uint64_t>::max() >> (8 - AddressByteSize) * 8;
+}
+
 } // End of namespace dwarf
 
 /// Dwarf constants format_provider

llvm/include/llvm/DebugInfo/DWARF/DWARFDebugRnglists.h

@@ -48,6 +48,7 @@ class DWARFDebugRnglist : public DWARFListType<RangeListEntry> {
   /// Build a DWARFAddressRangesVector from a rangelist.
   DWARFAddressRangesVector
   getAbsoluteRanges(Optional<object::SectionedAddress> BaseAddr,
+                    uint8_t AddressByteSize,
                     function_ref<Optional<object::SectionedAddress>(uint32_t)>
                         LookupPooledAddress) const;
 

llvm/lib/DebugInfo/DWARF/DWARFDebugLine.cpp

@@ -783,6 +783,18 @@ Error DWARFDebugLine::LineTable::parse(
     *OS << '\n';
     Row::dumpTableHeader(*OS, /*Indent=*/Verbose ? 12 : 0);
   }
+  bool TombstonedAddress = false;
+  auto EmitRow = [&] {
+    if (!TombstonedAddress) {
+      if (Verbose) {
+        *OS << "\n";
+        OS->indent(12);
+      }
+      if (OS)
+        State.Row.dump(*OS);
+      State.appendRowToMatrix();
+    }
+  };
   while (*OffsetPtr < EndOffset) {
     DataExtractor::Cursor Cursor(*OffsetPtr);
 
@@ -834,13 +846,7 @@ Error DWARFDebugLine::LineTable::parse(
         // No need to test the Cursor is valid here, since it must be to get
         // into this code path - if it were invalid, the default case would be
         // followed.
-        if (Verbose) {
-          *OS << "\n";
-          OS->indent(12);
-        }
-        if (OS)
-          State.Row.dump(*OS);
-        State.appendRowToMatrix();
+        EmitRow();
         State.resetRowAndSequence();
         break;
 
@@ -882,6 +888,10 @@ Error DWARFDebugLine::LineTable::parse(
             State.Row.Address.Address = TableData.getRelocatedAddress(
                 Cursor, &State.Row.Address.SectionIndex);
 
+            uint64_t Tombstone =
+                dwarf::computeTombstoneAddress(OpcodeAddressSize);
+            TombstonedAddress = State.Row.Address.Address == Tombstone;
+
             // Restore the address size if the extractor already had it.
             if (ExtractorAddressSize != 0)
               TableData.setAddressSize(ExtractorAddressSize);
@@ -981,13 +991,7 @@ Error DWARFDebugLine::LineTable::parse(
       case DW_LNS_copy:
         // Takes no arguments. Append a row to the matrix using the
         // current values of the state-machine registers.
-        if (Verbose) {
-          *OS << "\n";
-          OS->indent(12);
-        }
-        if (OS)
-          State.Row.dump(*OS);
-        State.appendRowToMatrix();
+        EmitRow();
         break;
 
       case DW_LNS_advance_pc:
@@ -1152,15 +1156,9 @@ Error DWARFDebugLine::LineTable::parse(
       ParsingState::AddrAndLineDelta Delta =
           State.handleSpecialOpcode(Opcode, OpcodeOffset);
 
-      if (Verbose) {
-        *OS << "address += " << Delta.Address << ",  line += " << Delta.Line
-            << "\n";
-        OS->indent(12);
-      }
-      if (OS)
-        State.Row.dump(*OS);
-
-      State.appendRowToMatrix();
+      if (Verbose)
+        *OS << "address += " << Delta.Address << ",  line += " << Delta.Line;
+      EmitRow();
       *OffsetPtr = Cursor.tell();
     }
 

llvm/lib/DebugInfo/DWARF/DWARFDebugRangeList.cpp

@@ -70,6 +70,9 @@ void DWARFDebugRangeList::dump(raw_ostream &OS) const {
 DWARFAddressRangesVector DWARFDebugRangeList::getAbsoluteRanges(
     llvm::Optional<object::SectionedAddress> BaseAddr) const {
   DWARFAddressRangesVector Res;
+  // debug_addr can't use the max integer tombstone because that's used for the
+  // base address specifier entry - so use max-1.
+  uint64_t Tombstone = dwarf::computeTombstoneAddress(AddressSize) - 1;
   for (const RangeListEntry &RLE : Entries) {
     if (RLE.isBaseAddressSelectionEntry(AddressSize)) {
       BaseAddr = {RLE.EndAddress, RLE.SectionIndex};
@@ -78,12 +81,16 @@ DWARFAddressRangesVector DWARFDebugRangeList::getAbsoluteRanges(
 
     DWARFAddressRange E;
     E.LowPC = RLE.StartAddress;
+    if (E.LowPC == Tombstone)
+      continue;
     E.HighPC = RLE.EndAddress;
     E.SectionIndex = RLE.SectionIndex;
     // Base address of a range list entry is determined by the closest preceding
     // base address selection entry in the same range list. It defaults to the
     // base address of the compilation unit if there is no such entry.
     if (BaseAddr) {
+      if (BaseAddr->Address == Tombstone)
+        continue;
       E.LowPC += BaseAddr->Address;
       E.HighPC += BaseAddr->Address;
       if (E.SectionIndex == -1ULL)

llvm/lib/DebugInfo/DWARF/DWARFDebugRnglists.cpp

@@ -89,16 +89,17 @@ Error RangeListEntry::extract(DWARFDataExtractor Data, uint64_t *OffsetPtr) {
 
 DWARFAddressRangesVector DWARFDebugRnglist::getAbsoluteRanges(
     llvm::Optional<object::SectionedAddress> BaseAddr, DWARFUnit &U) const {
-  return getAbsoluteRanges(BaseAddr, [&](uint32_t Index) {
-    return U.getAddrOffsetSectionItem(Index);
-  });
+  return getAbsoluteRanges(
+      BaseAddr, U.getAddressByteSize(),
+      [&](uint32_t Index) { return U.getAddrOffsetSectionItem(Index); });
 }
 
 DWARFAddressRangesVector DWARFDebugRnglist::getAbsoluteRanges(
-    Optional<object::SectionedAddress> BaseAddr,
+    Optional<object::SectionedAddress> BaseAddr, uint8_t AddressByteSize,
     function_ref<Optional<object::SectionedAddress>(uint32_t)>
         LookupPooledAddress) const {
   DWARFAddressRangesVector Res;
+  uint64_t Tombstone = dwarf::computeTombstoneAddress(AddressByteSize);
   for (const RangeListEntry &RLE : Entries) {
     if (RLE.EntryKind == dwarf::DW_RLE_end_of_list)
       break;
@@ -121,8 +122,12 @@ DWARFAddressRangesVector DWARFDebugRnglist::getAbsoluteRanges(
     switch (RLE.EntryKind) {
     case dwarf::DW_RLE_offset_pair:
       E.LowPC = RLE.Value0;
+      if (E.LowPC == Tombstone)
+        continue;
       E.HighPC = RLE.Value1;
       if (BaseAddr) {
+        if (BaseAddr->Address == Tombstone)
+          continue;
         E.LowPC += BaseAddr->Address;
         E.HighPC += BaseAddr->Address;
       }
@@ -149,6 +154,8 @@ DWARFAddressRangesVector DWARFDebugRnglist::getAbsoluteRanges(
       // so we should not run into any here.
       llvm_unreachable("Unsupported range list encoding");
     }
+    if (E.LowPC == Tombstone)
+      continue;
     Res.push_back(E);
   }
   return Res;
@@ -181,6 +188,8 @@ void RangeListEntry::dump(
       OS << ": ";
   }
 
+  uint64_t Tombstone = dwarf::computeTombstoneAddress(AddrSize);
+
   switch (EntryKind) {
   case dwarf::DW_RLE_end_of_list:
     OS << (DumpOpts.Verbose ? "" : "<End of list>");
@@ -208,8 +217,11 @@ void RangeListEntry::dump(
     break;
   case dwarf::DW_RLE_offset_pair:
     PrintRawEntry(OS, *this, AddrSize, DumpOpts);
-    DWARFAddressRange(Value0 + CurrentBase, Value1 + CurrentBase)
-        .dump(OS, AddrSize, DumpOpts);
+    if (CurrentBase != Tombstone)
+      DWARFAddressRange(Value0 + CurrentBase, Value1 + CurrentBase)
+          .dump(OS, AddrSize, DumpOpts);
+    else
+      OS << "dead code";
     break;
   case dwarf::DW_RLE_start_end:
     DWARFAddressRange(Value0, Value1).dump(OS, AddrSize, DumpOpts);

llvm/lib/DebugInfo/DWARF/DWARFDie.cpp

@@ -268,8 +268,18 @@ static void dumpAttribute(raw_ostream &OS, const DWARFDie &Die,
     WithColor(OS, Color) << Name;
   else if (Attr == DW_AT_decl_line || Attr == DW_AT_call_line)
     OS << *FormValue.getAsUnsignedConstant();
-  else if (Attr == DW_AT_high_pc && !DumpOpts.ShowForm && !DumpOpts.Verbose &&
-           FormValue.getAsUnsignedConstant()) {
+  else if (Attr == DW_AT_low_pc &&
+           (FormValue.getAsAddress() ==
+            dwarf::computeTombstoneAddress(U->getAddressByteSize()))) {
+    if (DumpOpts.Verbose) {
+      FormValue.dump(OS, DumpOpts);
+      OS << " (";
+    }
+    OS << "dead code";
+    if (DumpOpts.Verbose)
+      OS << ')';
+  } else if (Attr == DW_AT_high_pc && !DumpOpts.ShowForm && !DumpOpts.Verbose &&
+             FormValue.getAsUnsignedConstant()) {
     if (DumpOpts.ShowAddresses) {
       // Print the actual address rather than the offset.
       uint64_t LowPC, HighPC, Index;
@@ -415,6 +425,9 @@ Optional<uint64_t> DWARFDie::getLocBaseAttribute() const {
 }
 
 Optional<uint64_t> DWARFDie::getHighPC(uint64_t LowPC) const {
+  uint64_t Tombstone = dwarf::computeTombstoneAddress(U->getAddressByteSize());
+  if (LowPC == Tombstone)
+    return None;
   if (auto FormValue = find(DW_AT_high_pc)) {
     if (auto Address = FormValue->getAsAddress()) {
       // High PC is an address.