[llvm-dwp] Refuse DWARFv5 input DWP files.

[ikudrin]

Summary

The library can parse DWARFv5 unit index sections of DWP files, but llvm-dwp is not ready to process them. Refuse such input files for now.

[grimar]

(I know little about llvm-dwp, but have 2 minor general comments)

llvm/tools/llvm-dwp/llvm-dwp.cpp

622

I'd suggest to include the unsupported version number to new error messages.

e.g: "unsupported cu_index version: 99"

663

Is it tested somewhere? I see you've added a test for "cu_index", but this line is about "tu_index".

[ikudrin]

llvm/tools/llvm-dwp/llvm-dwp.cpp

663

Well, not. This check is added for completeness. To get here, the DWP file should have a "cu_index" section of version 2 and a "tu_index" section of version 5, which seems weird. Or should I add such an odd test anyway?

[dblaikie]

llvm/tools/llvm-dwp/llvm-dwp.cpp

663

Testing error cases is all about weird - yeah, would be good to have such a test case.

[grimar]

llvm/tools/llvm-dwp/llvm-dwp.cpp

663

+1.

[ikudrin]

• Added reporting the version of an unsupported index.
• Added a test for the unsupported TU index version case.
• Removed the actual CU in the test for the CU index version because it is not necessary.

[ikudrin]

Ping.

[ikudrin]

Ping.

[dblaikie]

Looks good - thanks!

llvm/tools/llvm-dwp/llvm-dwp.cpp

664

I think if you make this first thing a StringRef or Twine, rather than a std::string, you should be able to concatenate the rest without needing to wrap the last literal in a std::string, eg:

StringRef("foo")  + utostr(x) + "bar"

[ikudrin]

llvm/tools/llvm-dwp/llvm-dwp.cpp

664

Thanks! I've applied not exactly your variant, but since utostr() returns std::string, the wrappings of the first and the last string are not necessary at all.

llvm/include/llvm/DebugInfo/DWARF/DWARFUnitIndex.h

public:
   DWARFUnitIndex(DWARFSectionKind InfoColumnKind)
       : InfoColumnKind(InfoColumnKind) {}
 
   explicit operator bool() const { return Header.NumBuckets; }
 
   bool parse(DataExtractor IndexData);
   void dump(raw_ostream &OS) const;
 
+  uint32_t getVersion() const { return Header.Version; }
+
   const Entry *getFromOffset(uint32_t Offset) const;
   const Entry *getFromHash(uint64_t Offset) const;
 
   ArrayRef<DWARFSectionKind> getColumnKinds() const {
     return makeArrayRef(ColumnKinds.get(), Header.NumColumns);
   }
 
   ArrayRef<Entry> getRows() const {
     return makeArrayRef(Rows.get(), Header.NumBuckets);
   }
 };
 
 } // end namespace llvm
 
 #endif // LLVM_DEBUGINFO_DWARF_DWARFUNITINDEX_H

llvm/test/tools/llvm-dwp/X86/unsupported_cu_index_version.s

+# RUN: llvm-mc -triple x86_64-unknown-linux %s -filetype=obj -o %t.dwp
+# RUN: not llvm-dwp %t.dwp -o %t 2>&1 | FileCheck %s
+
+# CHECK: error: unsupported cu_index version: 5 (only version 2 is supported)
+
+## To reach the test point, no real data is needed in .debug_info.dwo,
+## but the section should not be empty.
+    .section .debug_info.dwo, "e", @progbits
+    .space 1
+
+    .section .debug_cu_index, "", @progbits
+## Header:
+    .short 5                        # Version
+    .space 2                        # Padding
+    .long 2                         # Section count
+    .long 1                         # Unit count
+    .long 2                         # Slot count
+## Hash Table of Signatures:
+    .quad 0x1100001122222222
+    .quad 0
+## Parallel Table of Indexes:
+    .long 1
+    .long 0
+## Table of Section Offsets:
+## Row 0:
+    .long 1                         # DW_SECT_INFO
+    .long 3                         # DW_SECT_ABBREV
+## Row 1:
+    .long 0
+    .long 0
+## Table of Section Sizes:
+    .long 1
+    .long 1

llvm/test/tools/llvm-dwp/X86/unsupported_tu_index_version.s

+# RUN: llvm-mc -triple x86_64-unknown-linux %s -filetype=obj -o %t.dwp
+# RUN: not llvm-dwp %t.dwp -o %t 2>&1 | FileCheck %s
+
+## Note: To reach the test point, we need a DWP file with a CU, a CU index
+## of version 2 and a TU index of version 5. A valid TU is not required, but
+## the .debug_types.dwo section should not be empty.
+
+# CHECK: error: unsupported tu_index version: 5 (only version 2 is supported)
+
+.section .debug_abbrev.dwo, "e", @progbits
+.LAbbrevBegin:
+    .uleb128 1                      # Abbreviation Code
+    .uleb128 17                     # DW_TAG_compile_unit
+    .byte 1                         # DW_CHILDREN_no
+    .uleb128 0x2131                 # DW_AT_GNU_dwo_id
+    .uleb128 7                      # DW_FORM_data8
+    .byte 0                         # EOM(1)
+    .byte 0                         # EOM(2)
+    .byte 0                         # EOM(3)
+.LAbbrevEnd:
+
+    .section .debug_info.dwo, "e", @progbits
+.LCUBegin:
+    .long .LCUEnd-.LCUVersion       # Length of Unit
+.LCUVersion:
+    .short 4                        # Version
+    .long 0                         # Abbrev offset
+    .byte 8                         # Address size
+    .uleb128 1                      # Abbrev [1] DW_TAG_compile_unit
+    .quad 0x1100001122222222        # DW_AT_GNU_dwo_id
+.LCUEnd:
+
+    .section .debug_types.dwo, "e", @progbits
+    .space 1
+
+    .section .debug_cu_index, "", @progbits
+## Header:
+    .long 2                         # Version
+    .long 2                         # Section count
+    .long 1                         # Unit count
+    .long 2                         # Slot count
+## Hash Table of Signatures:
+    .quad 0x1100001122222222
+    .quad 0
+## Parallel Table of Indexes:
+    .long 1
+    .long 0
+## Table of Section Offsets:
+## Row 0:
+    .long 1                         # DW_SECT_INFO
+    .long 3                         # DW_SECT_ABBREV
+## Row 1:
+    .long 0                         # Offset in .debug_info.dwo
+    .long 0                         # Offset in .debug_abbrev.dwo
+## Table of Section Sizes:
+    .long .LCUEnd-.LCUBegin         # Size in .debug_info.dwo
+    .long .LAbbrevEnd-.LAbbrevBegin # Size in .debug_abbrev.dwo
+
+    .section .debug_tu_index, "", @progbits
+## Header:
+    .short 5                        # Version
+    .space 2                        # Padding
+    .long 2                         # Section count
+    .long 1                         # Unit count
+    .long 2                         # Slot count
+## Hash Table of Signatures:
+    .quad 0x1100003333333333
+    .quad 0
+## Parallel Table of Indexes:
+    .long 1
+    .long 0
+## Table of Section Offsets:
+## Row 0:
+    .long 1                         # DW_SECT_INFO
+    .long 3                         # DW_SECT_ABBREV
+## Row 1:
+    .long 0
+    .long 0
+## Table of Section Sizes:
+    .long 1
+    .long 1

llvm/tools/llvm-dwp/llvm-dwp.cpp

if (CurCUIndexSection.empty()) {
                   ContributionOffsets[getContributionIndex(DW_SECT_EXT_TYPES)]);
       continue;
     }
 
     DWARFUnitIndex CUIndex(DW_SECT_INFO);
     DataExtractor CUIndexData(CurCUIndexSection, Obj.isLittleEndian(), 0);
     if (!CUIndex.parse(CUIndexData))
       return make_error<DWPError>("failed to parse cu_index");
+    if (CUIndex.getVersion() != 2)
+      return make_error<DWPError>(
+          "unsupported cu_index version: " + utostr(CUIndex.getVersion()) +
+          " (only version 2 is supported)");
 
     for (const DWARFUnitIndex::Entry &E : CUIndex.getRows()) {
       auto *I = E.getContributions();
       if (!I)
         continue;
       auto P = IndexEntries.insert(std::make_pair(E.getSignature(), CurEntry));
       Expected<CompileUnitIdentifiers> EID = getCUIdentifiers(
           getSubsection(AbbrevSection, E, DW_SECT_ABBREV),
for (const auto &Input : Inputs) {
 
     if (!CurTypesSection.empty()) {
       if (CurTypesSection.size() != 1)
         return make_error<DWPError>("multiple type unit sections in .dwp file");
       DWARFUnitIndex TUIndex(DW_SECT_EXT_TYPES);
       DataExtractor TUIndexData(CurTUIndexSection, Obj.isLittleEndian(), 0);
       if (!TUIndex.parse(TUIndexData))
         return make_error<DWPError>("failed to parse tu_index");
+      if (TUIndex.getVersion() != 2)
+        return make_error<DWPError>(
+            "unsupported tu_index version: " + utostr(TUIndex.getVersion()) +
+            " (only version 2 is supported)");
+
       addAllTypesFromDWP(
           Out, TypeIndexEntries, TUIndex, TypesSection, CurTypesSection.front(),
           CurEntry,
           ContributionOffsets[getContributionIndex(DW_SECT_EXT_TYPES)]);
     }
   }
 
   // Lie about there being no info contributions so the TU index only includes