[LLDB] - Add support for DW_FORM_addrx[1-4]? forms.

[grimar]

Summary

This adds the support for DW_FORM_addrx, DW_FORM_addrx1,
DW_FORM_addrx2, DW_FORM_addrx3, DW_FORM_addrx4 forms.

[grimar]

Planning changes. Going to remove the DW_FORM_rnglistx support from this patch, found it is incomplete and has other issues.

[grimar]

• Removed DW_FORM_rnglistx from this patch.

[grimar]

I have a patch for DW_FORM_rnglistx. Will be happy to land this one first though, as they share a code in DWARFUnit.h/cpp a bit.

lit/Breakpoint/Inputs/debug_addrx.yaml

+--- !ELF
+FileHeader:      
+  Class:           ELFCLASS64
+  Data:            ELFDATA2LSB
+  Type:            ET_EXEC
+  Machine:         EM_X86_64
+  Entry:           0x0000000000201000
+Sections:        
+  - Name:            .text
+    Type:            SHT_PROGBITS
+    Flags:           [ SHF_ALLOC, SHF_EXECINSTR ]
+    Address:         0x0000000000201000
+    AddressAlign:    0x0000000000000010
+    Content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
+  - Name:            .debug_str_offsets
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         44000000050000006B0000005F0000003A0000000E0000001A0000001E000000930000002E0000001600000067000000050000009A0000008E000000000000006900000095000000
+  - Name:            .debug_str
+    Type:            SHT_PROGBITS
+    Flags:           [ SHF_MERGE, SHF_STRINGS ]
+    AddressAlign:    0x0000000000000001
+    Content:         74686973005F5A336261723141005F76707472244100666F6F00696E74005F5F7674626C5F7074725F74797065005F5A4E314133666F6F4576002F686F6D652F756D622F74657374735F323031382F3131326C6C64625F726C657461677300746573742E63630041006100636C616E672076657273696F6E20382E302E3020287472756E6B2033343438333429006D61696E0078006F626A410062617200
+  - Name:            .debug_loc
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         E010200000000000E010200000000000030011009FE010200000000000E1102000000000000500110023019F00000000000000000000000000000000
+  - Name:            .debug_abbrev
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         011101252513050325721710171B25731711015517741700000213011D13360B03250B0B3A0B3B0B0000030D0003254913380B34190000040D00032549133A0B3B0B380B0000052E016E2503253A0B3B0B4C0B4D183C193F191D130000060500491334190000070F0049130000080F00491303250000091500491300000A240003253E0B0B0B00000B2E01111B1206401864137A193B0B471300000C050003254913341900000D3400021703253A0B3B0B491300000E2E01111B120640187A196E2503253A0B3B0B3F1900000F050003253A0B3B0B49130000102E01111B120640187A1903253A0B3B0B49133F19000011340003253A0B3B0B4913000000
+  - Name:            .debug_info
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         C8000000050001080000000001000400010800000000000000020800000000000000000000000C0000000C000000022E000000040910010303035C0000000004066C0000000104080507080105010210002E000000067000000000000761000000086700000005096C0000000A040504072E0000000B000100000001578600000008480000000C0DC60000000D000000000601096C000000000E010100000001570A0B010D0F0E010D2E0000000010020300000001570C01116C000000110F01122E00000000072E00000000
+  - Name:            .debug_rnglists
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         27000000050008000000000007E0102000000000000107F010200000000000010700112000000000000300
+  - Name:            .debug_macinfo
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         '00'
+  - Name:            .debug_addr
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         1C00000005000800E010200000000000F0102000000000000011200000000000
+  - Name:            .debug_line
+    Type:            SHT_PROGBITS
+    AddressAlign:    0x0000000000000001
+    Content:         97000000050008004C000000010101FB0E0D00010101010000000100000101011F010000000003011F020F051E02250000000066F622AD4EB40BCEA506478235D9D438250000000066F622AD4EB40BCEA506478235D9D438000902E0102000000000001905010A150201000101000902F010200000000000030C0105010A140201000101000902001120000000000003100105030A170203000101
+  - Name:            .debug_line_str
+    Type:            SHT_PROGBITS
+    Flags:           [ SHF_MERGE, SHF_STRINGS ]
+    AddressAlign:    0x0000000000000001
+    Content:         2F686F6D652F756D622F74657374735F323031382F3131326C6C64625F726C657461677300746573742E636300
+Symbols:

lit/Breakpoint/debug_addrx.test

+# RUN: yaml2obj %p/Inputs/debug_addrx.yaml > %ttest
+# RUN: lldb-test breakpoints %ttest %s | FileCheck %s
+
+# The following code and invocation were used.
+# clang -gdwarf-5 test.cc -o test_v5 -fuse-ld=lld -fno-rtti -O2 -ffunction-sections
+# Then output was converted to yaml and reduced.
+#
+# struct A {
+#   void foo() {};
+# };
+# 
+# void bar(struct A a) {
+#    a.foo();
+# }
+#
+# int main() {
+#  A objA;
+#  bar(objA);
+#
+#  return 0;
+# }
+#
+# clang and LLD versions were 8.0.0 (trunk 344834)
+#
+# Output file contains DW_FORM_addrx forms.
+
+b bar
+
+# CHECK-LABEL: b bar
+# CHECK: Address: {{.*}}`bar(A) at test.cc:15:1

source/Plugins/SymbolFile/DWARF/DWARFDebugInfoEntry.cpp

for (i = 0; i < numAttributes; ++i) {
             break;
 
           // 0 sized form
           case DW_FORM_flag_present:
             form_size = 0;
             break;
 
           // 1 byte values
+          case DW_FORM_addrx1:
           case DW_FORM_data1:
           case DW_FORM_flag:
           case DW_FORM_ref1:
           case DW_FORM_strx1:
             form_size = 1;
             break;
 
           // 2 byte values
+          case DW_FORM_addrx2:
           case DW_FORM_data2:
           case DW_FORM_ref2:
           case DW_FORM_strx2:
             form_size = 2;
             break;
 
+          // 3 byte values
+          case DW_FORM_addrx3:
           case DW_FORM_strx3:
             form_size = 3;
             break;
 
           // 4 byte values
+          case DW_FORM_addrx4:
           case DW_FORM_data4:
           case DW_FORM_ref4:
           case DW_FORM_strx4:
             form_size = 4;
             break;
 
           // 8 byte values
           case DW_FORM_data8:
           case DW_FORM_ref8:
           case DW_FORM_ref_sig8:
             form_size = 8;
             break;
 
           // signed or unsigned LEB 128 values
+          case DW_FORM_addrx:
           case DW_FORM_sdata:
           case DW_FORM_udata:
           case DW_FORM_ref_udata:
           case DW_FORM_GNU_addr_index:
           case DW_FORM_GNU_str_index:
           case DW_FORM_strx:
             debug_info_data.Skip_LEB128(&offset);
             break;

source/Plugins/SymbolFile/DWARF/DWARFFormValue.cpp

case DW_FORM_sdata:
       m_value.value.sval = data.GetSLEB128(offset_ptr);
       break;
     case DW_FORM_strp:
     case DW_FORM_line_strp:
       assert(m_cu);
       m_value.value.uval =
           data.GetMaxU64(offset_ptr, DWARFUnit::IsDWARF64(m_cu) ? 8 : 4);
       break;
+    case DW_FORM_addrx:
     case DW_FORM_strx:
       m_value.value.uval = data.GetULEB128(offset_ptr);
       break;
+    case DW_FORM_addrx1:
     case DW_FORM_strx1:
       m_value.value.uval = data.GetU8(offset_ptr);
       break;
+    case DW_FORM_addrx2:
     case DW_FORM_strx2:
       m_value.value.uval = data.GetU16(offset_ptr);
       break;
+    case DW_FORM_addrx3:
     case DW_FORM_strx3:
       m_value.value.uval = data.GetMaxU64(offset_ptr, 3);
       break;
+    case DW_FORM_addrx4:
     case DW_FORM_strx4:
       m_value.value.uval = data.GetU32(offset_ptr);
       break;
     //  case DW_FORM_APPLE_db_str:
     case DW_FORM_udata:
       m_value.value.uval = data.GetULEB128(offset_ptr);
       break;
     case DW_FORM_ref_addr:
case DW_FORM_ref_addr:
     return true;
 
   // 0 bytes values (implied from DW_FORM)
   case DW_FORM_flag_present:
   case DW_FORM_implicit_const:
     return true;
 
     // 1 byte values
+    case DW_FORM_addrx1:
     case DW_FORM_data1:
     case DW_FORM_flag:
     case DW_FORM_ref1:
     case DW_FORM_strx1:
       *offset_ptr += 1;
       return true;
 
     // 2 byte values
+    case DW_FORM_addrx2:
     case DW_FORM_data2:
     case DW_FORM_ref2:
     case DW_FORM_strx2:
       *offset_ptr += 2;
       return true;
 
     // 3 byte values
+    case DW_FORM_addrx3:
     case DW_FORM_strx3:
       *offset_ptr += 3;
       return true;
 
     // 32 bit for DWARF 32, 64 for DWARF 64
     case DW_FORM_sec_offset:
     case DW_FORM_strp:
       assert(cu);
       *offset_ptr += (cu->IsDWARF64() ? 8 : 4);
       return true;
 
     // 4 byte values
+    case DW_FORM_addrx4:
     case DW_FORM_data4:
     case DW_FORM_ref4:
     case DW_FORM_strx4:
       *offset_ptr += 4;
       return true;
 
     // 8 byte values
     case DW_FORM_data8:
     case DW_FORM_ref8:
     case DW_FORM_ref_sig8:
       *offset_ptr += 8;
       return true;
 
     // signed or unsigned LEB 128 values
+    case DW_FORM_addrx:
     case DW_FORM_sdata:
     case DW_FORM_udata:
     case DW_FORM_ref_udata:
     case DW_FORM_GNU_addr_index:
     case DW_FORM_GNU_str_index:
     case DW_FORM_strx:
       debug_info_data.Skip_LEB128(offset_ptr);
       return true;

 
 dw_addr_t DWARFFormValue::Address() const {
   SymbolFileDWARF *symbol_file = m_cu->GetSymbolFileDWARF();
 
   if (m_form == DW_FORM_addr)
     return Unsigned();
 
   assert(m_cu);
-  assert(m_form == DW_FORM_GNU_addr_index);
+  assert(m_form == DW_FORM_GNU_addr_index || m_form == DW_FORM_addrx ||
+         m_form == DW_FORM_addrx1 || m_form == DW_FORM_addrx2 ||
+         m_form == DW_FORM_addrx3 || m_form == DW_FORM_addrx4);
 
   if (!symbol_file)
     return 0;
 
   uint32_t index_size = m_cu->GetAddressByteSize();
   dw_offset_t addr_base = m_cu->GetAddrBase();
   lldb::offset_t offset = addr_base + m_value.value.uval * index_size;
   return symbol_file->get_debug_addr_data().GetMaxU64(&offset, index_size);
default:
     llvm_unreachable("Unhandled DW_FORM");
   }
   return -1;
 }
 
 bool DWARFFormValue::FormIsSupported(dw_form_t form) {
   switch (form) {
     case DW_FORM_addr:
+    case DW_FORM_addrx:
     case DW_FORM_block2:
     case DW_FORM_block4:
     case DW_FORM_data2:
     case DW_FORM_data4:
     case DW_FORM_data8:
     case DW_FORM_string:
     case DW_FORM_block:
     case DW_FORM_block1:

source/Plugins/SymbolFile/DWARF/DWARFUnit.h

public:
   uint32_t GetLength() const { return m_length; }
   uint16_t GetVersion() const { return m_version; }
   const DWARFAbbreviationDeclarationSet *GetAbbreviations() const;
   dw_offset_t GetAbbrevOffset() const;
   uint8_t GetAddressByteSize() const { return m_addr_size; }
   dw_addr_t GetBaseAddress() const { return m_base_addr; }
   dw_addr_t GetAddrBase() const { return m_addr_base; }
   dw_addr_t GetRangesBase() const { return m_ranges_base; }
-  void SetAddrBase(dw_addr_t addr_base, dw_addr_t ranges_base,
-                   dw_offset_t base_obj_offset);
+  void SetAddrBase(dw_addr_t addr_base);
+  void SetRangesBase(dw_addr_t ranges_base);
+  void SetBaseObjOffset(dw_offset_t base_obj_offset);
   void BuildAddressRangeTable(SymbolFileDWARF *dwarf,
                               DWARFDebugAranges *debug_aranges);
 
   lldb::ByteOrder GetByteOrder() const;
 
   lldb_private::TypeSystem *GetTypeSystem();
 
   const DWARFDebugAranges &GetFunctionAranges();

source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp

if (log && log->GetVerbose()) {
     else
       m_die_array[0].Dump(m_dwarf, this, strm, UINT32_MAX);
     log->PutString(strm.GetString());
   }
 }
 
 // m_die_array_mutex must be already held as read/write.
 void DWARFUnit::AddUnitDIE(const DWARFDebugInfoEntry &cu_die) {
+  SetAddrBase(
+      cu_die.GetAttributeValueAsUnsigned(m_dwarf, this, DW_AT_addr_base, 0));
+
   uint64_t base_addr = cu_die.GetAttributeValueAsAddress(
       m_dwarf, this, DW_AT_low_pc, LLDB_INVALID_ADDRESS);
   if (base_addr == LLDB_INVALID_ADDRESS)
     base_addr = cu_die.GetAttributeValueAsAddress(
         m_dwarf, this, DW_AT_entry_pc, 0);
   SetBaseAddress(base_addr);
 
   std::unique_ptr<SymbolFileDWARFDwo> dwo_symbol_file =
void DWARFUnit::AddUnitDIE(const DWARFDebugInfoEntry &cu_die) {
   if (main_dwo_id != sub_dwo_id)
     return; // The 2 dwo ID isn't match. Don't use the dwo file as it belongs to
   // a differectn compilation.
 
   m_dwo_symbol_file = std::move(dwo_symbol_file);
 
   dw_addr_t addr_base =
       cu_die.GetAttributeValueAsUnsigned(m_dwarf, this, DW_AT_GNU_addr_base, 0);
+  dwo_cu->SetAddrBase(addr_base);
+
   dw_addr_t ranges_base = cu_die.GetAttributeValueAsUnsigned(
       m_dwarf, this, DW_AT_GNU_ranges_base, 0);
-  dwo_cu->SetAddrBase(addr_base, ranges_base, m_offset);
+  dwo_cu->SetRangesBase(ranges_base);
+
+  dwo_cu->SetBaseObjOffset(m_offset);
 }
 
 DWARFDIE DWARFUnit::LookupAddress(const dw_addr_t address) {
   if (DIE()) {
     const DWARFDebugAranges &func_aranges = GetFunctionAranges();
 
     // Re-check the aranges auto pointer contents in case it was created above
     if (!func_aranges.IsEmpty())

 const DWARFAbbreviationDeclarationSet *DWARFUnit::GetAbbreviations() const {
   return m_abbrevs;
 }
 
 dw_offset_t DWARFUnit::GetAbbrevOffset() const {
   return m_abbrevs ? m_abbrevs->GetOffset() : DW_INVALID_OFFSET;
 }
 
-void DWARFUnit::SetAddrBase(dw_addr_t addr_base,
-                            dw_addr_t ranges_base,
-                            dw_offset_t base_obj_offset) {
-  m_addr_base = addr_base;
+void DWARFUnit::SetAddrBase(dw_addr_t addr_base) { m_addr_base = addr_base; }
+
+void DWARFUnit::SetRangesBase(dw_addr_t ranges_base) {
   m_ranges_base = ranges_base;
+}
+
+void DWARFUnit::SetBaseObjOffset(dw_offset_t base_obj_offset) {
   m_base_obj_offset = base_obj_offset;
 }
 
 // It may be called only with m_die_array_mutex held R/W.
 void DWARFUnit::ClearDIEsRWLocked() {
   m_die_array.clear();
   m_die_array.shrink_to_fit();