[DebugInfo] Allow GNU macro extension to be read

Allow the GNU .debug_macro extension to be parsed and printed by
llvm-dwarfdump. In an upcoming patch support will be added for emitting
that format also.

Reviewed By: dblaikie

Differential Revision: https://reviews.llvm.org/D82974

Author: dstenb

llvm/include/llvm/BinaryFormat/Dwarf.h

@@ -480,6 +480,7 @@ StringRef LNStandardString(unsigned Standard);
 StringRef LNExtendedString(unsigned Encoding);
 StringRef MacinfoString(unsigned Encoding);
 StringRef MacroString(unsigned Encoding);
+StringRef GnuMacroString(unsigned Encoding);
 StringRef RangeListEncodingString(unsigned Encoding);
 StringRef LocListEncodingString(unsigned Encoding);
 StringRef CallFrameString(unsigned Encoding, Triple::ArchType Arch);

llvm/include/llvm/DebugInfo/DWARF/DWARFDebugMacro.h

@@ -96,6 +96,9 @@ class DWARFDebugMacro {
     MacroHeader Header;
     SmallVector<Entry, 4> Macros;
     uint64_t Offset;
+
+    /// Whether or not this is a .debug_macro section.
+    bool IsDebugMacro;
   };
 
   /// A list of all the macro entries in the debug_macinfo section.

llvm/lib/BinaryFormat/Dwarf.cpp

@@ -488,6 +488,17 @@ StringRef llvm::dwarf::MacroString(unsigned Encoding) {
   }
 }
 
+StringRef llvm::dwarf::GnuMacroString(unsigned Encoding) {
+  switch (Encoding) {
+  default:
+    return StringRef();
+#define HANDLE_DW_MACRO_GNU(ID, NAME)                                          \
+  case DW_MACRO_GNU_##NAME:                                                    \
+    return "DW_MACRO_GNU_" #NAME;
+#include "llvm/BinaryFormat/Dwarf.def"
+  }
+}
+
 unsigned llvm::dwarf::getMacro(StringRef MacroString) {
   return StringSwitch<unsigned>(MacroString)
 #define HANDLE_DW_MACRO(ID, NAME) .Case("DW_MACRO_" #NAME, ID)

llvm/lib/DebugInfo/DWARF/DWARFDebugMacro.cpp

@@ -40,7 +40,7 @@ void DWARFDebugMacro::dump(raw_ostream &OS) const {
   unsigned IndLevel = 0;
   for (const auto &Macros : MacroLists) {
     OS << format("0x%08" PRIx64 ":\n", Macros.Offset);
-    if (Macros.Header.Version >= 5)
+    if (Macros.IsDebugMacro)
       Macros.Header.dumpMacroHeader(OS);
     for (const Entry &E : Macros.Macros) {
       // There should not be DW_MACINFO_end_file when IndLevel is Zero. However,
@@ -52,8 +52,10 @@ void DWARFDebugMacro::dump(raw_ostream &OS) const {
         OS << "  ";
       IndLevel += (E.Type == DW_MACINFO_start_file);
       // Based on which version we are handling choose appropriate macro forms.
-      if (Macros.Header.Version >= 5)
-        WithColor(OS, HighlightColor::Macro).get() << MacroString(E.Type);
+      if (Macros.IsDebugMacro)
+        WithColor(OS, HighlightColor::Macro).get()
+            << (Macros.Header.Version < 5 ? GnuMacroString(E.Type)
+                                          : MacroString(E.Type));
       else
         WithColor(OS, HighlightColor::Macro).get() << MacinfoString(E.Type);
       switch (E.Type) {
@@ -67,6 +69,9 @@ void DWARFDebugMacro::dump(raw_ostream &OS) const {
         // DW_MACRO_start_file == DW_MACINFO_start_file
         // DW_MACRO_end_file   == DW_MACINFO_end_file
         // For readability/uniformity we are using DW_MACRO_*.
+        //
+        // The GNU .debug_macro extension's entries have the same encoding
+        // as DWARF 5's DW_MACRO_* entries, so we only use the latter here.
       case DW_MACRO_define:
       case DW_MACRO_undef:
       case DW_MACRO_define_strp:
@@ -118,6 +123,7 @@ Error DWARFDebugMacro::parseImpl(
       MacroLists.emplace_back();
       M = &MacroLists.back();
       M->Offset = Offset;
+      M->IsDebugMacro = IsMacro;
       if (IsMacro) {
         auto Err = M->Header.parseMacroHeader(Data, &Offset);
         if (Err)

llvm/test/DebugInfo/X86/debug-macro-gnu.s

@@ -0,0 +1,59 @@
+## This test checks that llvm-dwarfdump can dump a .debug_macro section
+## with the GNU extension format.
+
+# RUN: llvm-mc -triple x86_64-unknown-linux -filetype=obj %s -o -| \
+# RUN:   llvm-dwarfdump -debug-macro - | FileCheck -strict-whitespace -match-full-lines %s
+
+#      CHECK:.debug_macro contents:
+# CHECK-NEXT:0x00000000:
+# CHECK-NEXT:macro header: version = 0x0004, flags = 0x02, format = DWARF32, debug_line_offset = 0x00000000
+# CHECK-NEXT:DW_MACRO_GNU_start_file - lineno: 0 filenum: 0
+# CHECK-NEXT:  DW_MACRO_GNU_start_file - lineno: 1 filenum: 6
+# CHECK-NEXT:    DW_MACRO_GNU_define_indirect - lineno: 1 macro: FOO 5
+# CHECK-NEXT:  DW_MACRO_GNU_end_file
+# CHECK-NEXT:  DW_MACRO_GNU_undef_indirect - lineno: 8 macro: WORLD1
+# CHECK-NEXT:  DW_MACRO_GNU_transparent_include - import offset: 0x[[OFFSET:[0-9]+]]
+# CHECK-NEXT:DW_MACRO_GNU_end_file
+
+#      CHECK:0x[[OFFSET]]:
+# CHECK-NEXT:macro header: version = 0x0004, flags = 0x00, format = DWARF32
+# CHECK-NEXT:DW_MACRO_GNU_define_indirect - lineno: 0 macro: WORLD 2
+
+	.section	.debug_macro,"",@progbits
+.Lcu_macro_begin0:
+	.short	4                      # Macro information version
+	.byte	2                       # Flags: 32 bit, debug_line_offset present
+	.long	0                       # debug_line_offset
+	.byte	3                       # DW_MACRO_GNU_start_file
+	.byte	0                       # Line Number
+	.byte	0                       # File Number
+	.byte	3                       # DW_MACRO_GNU_start_file
+	.byte	1                       # Line Number
+	.byte	6                       # File Number
+	.byte	5                       # DW_MACRO_GNU_define_indirect
+	.byte	1                       # Line Number
+	.long	.Linfo_string0          # Macro String
+	.byte	4                       # DW_MACRO_GNU_end_file
+	.byte	6                       # DW_MACRO_GNU_undef_indirect
+	.byte	8                       # Line Number
+	.long	.Linfo_string1          # Macro String
+	.byte	7                       # DW_MACRO_GNU_transparent_include
+	.long	.Lmacro1                # Macro Unit Offset
+	.byte	4                       # DW_MACRO_GNU_end_file
+	.byte	0                       # End Of Macro List Mark
+
+.Lmacro1:
+	.short	4                      # Macro information version
+	.byte	0                       # Flags: 32 bit
+	.byte	5                       # DW_MACRO_GNU_define_indirect
+	.byte	0                       # Line Number
+	.long	.Linfo_string2          # Macro String
+	.byte	0                       # End Of Macro List Mark
+
+	.section	.debug_str,"MS",@progbits,1
+.Linfo_string0:
+	.asciz	"FOO 5"
+.Linfo_string1:
+	.asciz	"WORLD1"
+.Linfo_string2:
+	.asciz	"WORLD 2"