[DWARF] Fix DWARFUnit::getDebugInfoSize() for 64-bit DWARF.

[ikudrin]

Summary

The calculations in that method were correct only for DWARF32. In cases of DWARF64 they might result in overflow which might lead to a crash, for example, in DWARFUnit::extractDIEsToVector().

[probinson]

Hi @aprantl please remember to post a comment along with approval, so that Phab will send an email to the commits list. Thanks!

llvm/trunk/include/llvm/BinaryFormat/Dwarf.h

case DwarfFormat::DWARF64:
       return 8;
     }
     llvm_unreachable("Invalid Format value");
   }
 
   explicit operator bool() const { return Version && AddrSize; }
 };
 
+/// Get the byte size of the unit length field depending on the DWARF format.
+inline uint8_t getUnitLengthFieldByteSize(DwarfFormat Format) {
+  switch (Format) {
+  case DwarfFormat::DWARF32:
+    return 4;
+  case DwarfFormat::DWARF64:
+    return 12;
+  }
+  llvm_unreachable("Invalid Format value");
+}
+
 /// Get the fixed byte size for a given form.
 ///
 /// If the form has a fixed byte size, then an Optional with a value will be
 /// returned. If the form is always encoded using a variable length storage
 /// format (ULEB or SLEB numbers or blocks) then None will be returned.
 ///
 /// \param Form DWARF form to get the fixed byte size for.
 /// \param Params DWARF parameters to help interpret forms.

llvm/trunk/include/llvm/DebugInfo/DWARF/DWARFUnit.h

public:
   const DWARFUnitIndex::Entry *getIndexEntry() const { return IndexEntry; }
   uint64_t getTypeHash() const { return TypeHash; }
   uint32_t getTypeOffset() const { return TypeOffset; }
   uint8_t getUnitType() const { return UnitType; }
   bool isTypeUnit() const {
     return UnitType == dwarf::DW_UT_type || UnitType == dwarf::DW_UT_split_type;
   }
   uint8_t getSize() const { return Size; }
+  uint8_t getUnitLengthFieldByteSize() const {
+    return dwarf::getUnitLengthFieldByteSize(FormParams.Format);
+  }
   uint64_t getNextUnitOffset() const {
-    return Offset + Length +
-           (FormParams.Format == llvm::dwarf::DwarfFormat::DWARF64 ? 4 : 0) +
-           FormParams.getDwarfOffsetByteSize();
+    return Offset + Length + getUnitLengthFieldByteSize();
   }
 };
 
 const DWARFUnitIndex &getDWARFUnitIndex(DWARFContext &Context,
                                         DWARFSectionKind Kind);
 
 /// Describe a collection of units. Intended to hold all units either from
 /// .debug_info and .debug_types, or from .debug_info.dwo and .debug_types.dwo.
public:
 
   virtual void dump(raw_ostream &OS, DIDumpOptions DumpOpts) = 0;
 
   Error tryExtractDIEsIfNeeded(bool CUDieOnly);
 
 private:
   /// Size in bytes of the .debug_info data associated with this compile unit.
   size_t getDebugInfoSize() const {
-    return Header.getLength() + 4 - getHeaderSize();
+    return Header.getLength() + Header.getUnitLengthFieldByteSize() -
+           getHeaderSize();
   }
 
   /// extractDIEsIfNeeded - Parses a compile unit and indexes its DIEs if it
   /// hasn't already been done
   void extractDIEsIfNeeded(bool CUDieOnly);
 
   /// extractDIEsToVector - Appends all parsed DIEs to a vector.
   void extractDIEsToVector(bool AppendCUDie, bool AppendNonCUDIEs,

llvm/trunk/test/tools/llvm-dwarfdump/X86/debug_info_min_dwarf64.s

+# RUN: llvm-mc %s -filetype obj -triple x86_64-unknown-elf -o - | \
+# RUN:   llvm-dwarfdump -debug-info - | \
+# RUN:   FileCheck %s
+
+        .section .debug_abbrev,"",@progbits
+        .byte 0x01  # Abbrev code
+        .byte 0x11  # DW_TAG_compile_unit
+        .byte 0x00  # DW_CHILDREN_no
+        .byte 0x13  # DW_AT_language
+        .byte 0x05  # DW_FORM_data2
+        .byte 0x00  # EOM(1)
+        .byte 0x00  # EOM(2)
+        .byte 0x00  # EOM(3)
+        
+        .section .debug_info,"",@progbits
+# CHECK: .debug_info contents:
+# CHECK-NEXT: 0x00000000: Compile Unit:
+DI_4_64_start:
+        .long 0xffffffff       # DWARF64 mark
+        .quad DI_4_64_end - DI_4_64_version # Length of Unit
+# CHECK-SAME: length = 0x0000000f
+DI_4_64_version:
+        .short 4               # DWARF version number
+# CHECK-SAME: version = 0x0004
+        .quad .debug_abbrev    # Offset Into Abbrev. Section
+# CHECK-SAME: abbr_offset = 0x0000
+        .byte 8                # Address Size (in bytes)
+# CHECK-SAME: addr_size = 0x08
+# CHECK-SAME: (next unit at 0x0000001b)
+
+        .byte 1                # Abbreviation code
+# CHECK: 0x00000017: DW_TAG_compile_unit
+        .short 4               # DW_LANG_C_plus_plus
+# CHECK-NEXT: DW_AT_language (DW_LANG_C_plus_plus)
+        .byte 0                # NULL
+DI_4_64_end:
+