[lldb] Fix `DW_AT_decl_file` from `DW_AT_specification` from a different CU

[jankratochvil]

Summary

This would be reproducible in future DWZ category of the testsuite as:

Failed Tests (1):
  lldb-api :: python_api/symbol-context/two-files/TestSymbolContextTwoFiles.py

[labath]

Looks good. Thanks for catching that.

I am particularly happy that we now have an explicit DW_FORM_ref_addr test. It's a very interesting/tricky case, but it does not happen all that often in practice.

lldb/test/Shell/SymbolFile/DWARF/DW_AT_decl_file-DW_AT_specification-crosscu.s

97

These almost-aligned comments are fairly disorienting.

145–153

the offset comments are misaligned and incorrect

[jankratochvil]

Thanks; I see you check very every line even in testcases.

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

case DW_AT_calling_convention:
       calling_convention = form_value.Unsigned();
       break;
 
     case DW_AT_containing_type:
       containing_type = form_value;
       break;
 
     case DW_AT_decl_file:
-      decl.SetFile(die.GetCU()->GetFile(form_value.Unsigned()));
+      // die.GetCU() can differ if DW_AT_specification uses DW_FORM_ref_addr.
+      decl.SetFile(
+          attributes.CompileUnitAtIndex(i)->GetFile(form_value.Unsigned()));
       break;
     case DW_AT_decl_line:
       decl.SetLine(form_value.Unsigned());
       break;
     case DW_AT_decl_column:
       decl.SetColumn(form_value.Unsigned());
       break;
 

lldb/source/Plugins/SymbolFile/DWARF/DWARFAttribute.h

protected:
   DWARFFormValue::ValueType m_value;
 };
 
 class DWARFAttributes {
 public:
   DWARFAttributes();
   ~DWARFAttributes();
 
-  void Append(const DWARFUnit *cu, dw_offset_t attr_die_offset,
-              dw_attr_t attr, dw_form_t form);
-  const DWARFUnit *CompileUnitAtIndex(uint32_t i) const {
-    return m_infos[i].cu;
-  }
+  void Append(DWARFUnit *cu, dw_offset_t attr_die_offset, dw_attr_t attr,
+              dw_form_t form);
+  DWARFUnit *CompileUnitAtIndex(uint32_t i) const { return m_infos[i].cu; }
   dw_offset_t DIEOffsetAtIndex(uint32_t i) const {
     return m_infos[i].die_offset;
   }
   dw_attr_t AttributeAtIndex(uint32_t i) const {
     return m_infos[i].attr.get_attr();
   }
   dw_attr_t FormAtIndex(uint32_t i) const { return m_infos[i].attr.get_form(); }
   bool ExtractFormValueAtIndex(uint32_t i, DWARFFormValue &form_value) const;
   DWARFDIE FormValueAsReferenceAtIndex(uint32_t i) const;
   DWARFDIE FormValueAsReference(dw_attr_t attr) const;
   uint32_t FindAttributeIndex(dw_attr_t attr) const;
   void Clear() { m_infos.clear(); }
   size_t Size() const { return m_infos.size(); }
 
 protected:
   struct AttributeValue {
-    const DWARFUnit *cu;        // Keep the compile unit with each attribute in
-                                // case we have DW_FORM_ref_addr values
+    DWARFUnit *cu; // Keep the compile unit with each attribute in
     dw_offset_t die_offset;
     DWARFAttribute attr;
   };
   typedef llvm::SmallVector<AttributeValue, 8> collection;
   collection m_infos;
 };
 
 #endif // LLDB_SOURCE_PLUGINS_SYMBOLFILE_DWARF_DWARFATTRIBUTE_H

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

uint32_t DWARFAttributes::FindAttributeIndex(dw_attr_t attr) const {
   collection::const_iterator pos;
   for (pos = beg; pos != end; ++pos) {
     if (pos->attr.get_attr() == attr)
       return std::distance(beg, pos);
   }
   return UINT32_MAX;
 }
 
-void DWARFAttributes::Append(const DWARFUnit *cu, dw_offset_t attr_die_offset,
+void DWARFAttributes::Append(DWARFUnit *cu, dw_offset_t attr_die_offset,
                              dw_attr_t attr, dw_form_t form) {
   AttributeValue attr_value = {
       cu, attr_die_offset, {attr, form, DWARFFormValue::ValueType()}};
   m_infos.push_back(attr_value);
 }
 
 bool DWARFAttributes::ExtractFormValueAtIndex(
     uint32_t i, DWARFFormValue &form_value) const {

lldb/source/Plugins/SymbolFile/DWARF/DWARFDebugInfoEntry.h

public:
 
   void BuildFunctionAddressRangeTable(const DWARFUnit *cu,
                                       DWARFDebugAranges *debug_aranges) const;
 
   bool Extract(const lldb_private::DWARFDataExtractor &data,
                const DWARFUnit *cu, lldb::offset_t *offset_ptr);
 
   using Recurse = DWARFBaseDIE::Recurse;
-  size_t GetAttributes(const DWARFUnit *cu, DWARFAttributes &attrs,
+  size_t GetAttributes(DWARFUnit *cu, DWARFAttributes &attrs,
                        Recurse recurse = Recurse::yes) const {
     return GetAttributes(cu, attrs, recurse, 0 /* curr_depth */);
   }
 
   dw_offset_t
   GetAttributeValue(const DWARFUnit *cu, const dw_attr_t attr,
                     DWARFFormValue &formValue,
                     dw_offset_t *end_attr_offset_ptr = nullptr,
uint32_t m_sibling_idx : 31, // How many to add to "this" to get the sibling.
       // a single NULL terminating child.
       m_has_children : 1;
   uint16_t m_abbr_idx;
   /// A copy of the DW_TAG value so we don't have to go through the compile
   /// unit abbrev table
   dw_tag_t m_tag = llvm::dwarf::DW_TAG_null;
 
 private:
-  size_t GetAttributes(const DWARFUnit *cu, DWARFAttributes &attrs,
-                       Recurse recurse, uint32_t curr_depth) const;
+  size_t GetAttributes(DWARFUnit *cu, DWARFAttributes &attrs, Recurse recurse,
+                       uint32_t curr_depth) const;
 };
 
 #endif // LLDB_SOURCE_PLUGINS_SYMBOLFILE_DWARF_DWARFDEBUGINFOENTRY_H

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

bool DWARFDebugInfoEntry::GetDIENamesAndRanges(
   }
   return !ranges.IsEmpty();
 }
 
 // Get all attribute values for a given DIE, including following any
 // specification or abstract origin attributes and including those in the
 // results. Any duplicate attributes will have the first instance take
 // precedence (this can happen for declaration attributes).
-size_t DWARFDebugInfoEntry::GetAttributes(const DWARFUnit *cu,
+size_t DWARFDebugInfoEntry::GetAttributes(DWARFUnit *cu,
                                           DWARFAttributes &attributes,
                                           Recurse recurse,
                                           uint32_t curr_depth) const {
   const auto *abbrevDecl = GetAbbreviationDeclarationPtr(cu);
   if (abbrevDecl) {
     const DWARFDataExtractor &data = cu->GetData();
     lldb::offset_t offset = GetFirstAttributeOffset();
 

lldb/test/Shell/SymbolFile/DWARF/DW_AT_decl_file-DW_AT_specification-crosscu.s

+# Check that DW_AT_decl_file inherited by DW_AT_specification from a different
+# DW_TAG_compile_unit is using the DW_TAG_compile_unit->DW_AT_stmt_list where the
+# DW_AT_decl_file is located (and not where the DW_AT_specification is located).
+
+# REQUIRES: x86
+
+# RUN: llvm-mc -filetype=obj -triple x86_64-pc-linux %s -o %t
+# RUN: %lldb %t \
+# RUN:   -o "image lookup -v -n main" \
+# RUN:   -o exit | FileCheck %s
+
+# CHECK: FuncType: id = {0x0000002a}, byte-size = 0, decl = filename1:1, compiler_type = "int (void)"
+
+  .text
+  .globl  main                            # -- Begin function main
+  .type main,@function
+main:                                   # @main
+.Lfunc_begin0:
+        .byte 0
+.Lfunc_end0:
+  .size main, .Lfunc_end0-main
+                                        # -- End function
+  .section  .debug_abbrev,"",@progbits
+  .byte 1                               # Abbreviation Code
+  .byte 17                              # DW_TAG_compile_unit
+  .byte 1                               # DW_CHILDREN_yes
+  .byte 37                              # DW_AT_producer
+  .byte 14                              # DW_FORM_strp
+  .byte 19                              # DW_AT_language
+  .byte 5                               # DW_FORM_data2
+  .byte 3                               # DW_AT_name
+  .byte 14                              # DW_FORM_strp
+  .byte 16                              # DW_AT_stmt_list
+  .byte 23                              # DW_FORM_sec_offset
+  .byte 27                              # DW_AT_comp_dir
+  .byte 14                              # DW_FORM_strp
+  .byte 17                              # DW_AT_low_pc
+  .byte 1                               # DW_FORM_addr
+  .byte 18                              # DW_AT_high_pc
+  .byte 6                               # DW_FORM_data4
+  .byte 0                               # EOM(1)
+  .byte 0                               # EOM(2)
+  .byte 2                               # Abbreviation Code
+  .byte 46                              # DW_TAG_subprogram
+  .byte 0                               # DW_CHILDREN_no
+  .byte 3                               # DW_AT_name
+  .byte 14                              # DW_FORM_strp
+  .byte 17                              # DW_AT_low_pc
+  .byte 1                               # DW_FORM_addr
+  .byte 18                              # DW_AT_high_pc
+  .byte 6                               # DW_FORM_data4
+  .byte 64                              # DW_AT_frame_base
+  .byte 24                              # DW_FORM_exprloc
+  .byte 71                              # DW_AT_specification
+  .byte 16                              # DW_FORM_ref_addr
+  .byte 0                               # EOM(1)
+  .byte 0                               # EOM(2)
+  .byte 3                               # Abbreviation Code
+  .byte 36                              # DW_TAG_base_type
+  .byte 0                               # DW_CHILDREN_no
+  .byte 3                               # DW_AT_name
+  .byte 14                              # DW_FORM_strp
+  .byte 62                              # DW_AT_encoding
+  .byte 11                              # DW_FORM_data1
+  .byte 11                              # DW_AT_byte_size
+  .byte 11                              # DW_FORM_data1
+  .byte 0                               # EOM(1)
+  .byte 0                               # EOM(2)
+  .byte 4                               # Abbreviation Code
+  .byte 46                              # DW_TAG_subprogram
+  .byte 0                               # DW_CHILDREN_no
+  .byte 58                              # DW_AT_decl_file
+  .byte 11                              # DW_FORM_data1
+  .byte 59                              # DW_AT_decl_line
+  .byte 11                              # DW_FORM_data1
+  .byte 39                              # DW_AT_prototyped
+  .byte 25                              # DW_FORM_flag_present
+  .byte 73                              # DW_AT_type
+  .byte 19                              # DW_FORM_ref4
+  .byte 63                              # DW_AT_external
+  .byte 25                              # DW_FORM_flag_present
+  .byte 60                              # DW_AT_declaration
+  .byte 25                              # DW_FORM_flag_present
+  .byte 0                               # EOM(1)
+  .byte 0                               # EOM(2)
+  .byte 0                               # EOM(3)
+
+  .section  .debug_info,"",@progbits
+.Ldebug_cu0:
+
+  .long .Ldebug_info_end0-.Ldebug_info_start0 # Length of Unit
+.Ldebug_info_start0:
+  .short  4                               # DWARF version number
+  .long .debug_abbrev                   # Offset Into Abbrev. Section
+  .byte 8                               # Address Size (in bytes)
+  .byte 1                               # Abbrev [1] 0xb:0x40 DW_TAG_compile_unit
+  .long .Linfo_string_producer                  # DW_AT_producer
+  .short  12                              # DW_AT_language
+  .long .Linfo_string_source                  # DW_AT_name
+  .long .Lline_table_start0             # DW_AT_stmt_list
+  .long .Linfo_string_directory                  # DW_AT_comp_dir
+  .quad .Lfunc_begin0                   # DW_AT_low_pc
+  .long .Lfunc_end0-.Lfunc_begin0       # DW_AT_high_pc
+  .byte 2                               # Abbrev [2] 0x2a:0x19 DW_TAG_subprogram
+  .long .Linfo_string_main                  # DW_AT_name
+  .quad .Lfunc_begin0                   # DW_AT_low_pc
+  .long .Lfunc_end0-.Lfunc_begin0       # DW_AT_high_pc
+  .byte 1                               # DW_AT_frame_base
+  .byte 86
+  .long .Ldebug_info1_main - .Ldebug_cu0 # DW_AT_specification
+  .byte 0                               # End Of Children Mark
+.Ldebug_info_end0:
+
+.Ldebug_info1:
+  .long .Ldebug_info_end1-.Ldebug_info_start1 # Length of Unit
+.Ldebug_info_start1:
+  .short  4                               # DWARF version number
+  .long .debug_abbrev                   # Offset Into Abbrev. Section
+  .byte 8                               # Address Size (in bytes)
+  .byte 1                               # Abbrev [1] 0xb:0x40 DW_TAG_compile_unit
+  .long .Linfo_string_producer                  # DW_AT_producer
+  .short  12                              # DW_AT_language
+  .long .Linfo_string_source                  # DW_AT_name
+  .long .Lline_table_start1             # DW_AT_stmt_list
+  .long .Linfo_string_directory                  # DW_AT_comp_dir
+  .quad .Lfunc_begin0                   # DW_AT_low_pc
+  .long .Lfunc_end0-.Lfunc_begin0       # DW_AT_high_pc
+.Ldebug_info1_main:
+  .byte 4                               # Abbrev [2] 0x2a:0x19 DW_TAG_subprogram
+  .byte 1                               # DW_AT_decl_file
+  .byte 1                               # DW_AT_decl_line
+                                        # DW_AT_prototyped
+  .long .Ldebug_info1_int - .Ldebug_info1 # DW_AT_type
+                                        # DW_AT_external
+.Ldebug_info1_int:
+  .byte 3                               # Abbrev [3] 0x43:0x7 DW_TAG_base_type
+  .long .Linfo_string_int                  # DW_AT_name
+  .byte 5                               # DW_AT_encoding
+  .byte 4                               # DW_AT_byte_size
+  .byte 0                               # End Of Children Mark
+.Ldebug_info_end1:
+
+  .section  .debug_str,"MS",@progbits,1
+.Linfo_string_producer:
+  .asciz  "clang version 12.0.0" # string offset=0
+.Linfo_string_source:
+  .asciz  "source.c" # string offset=130
+.Linfo_string_directory:
+  .asciz  "/directory" # string offset=196
+.Linfo_string_main:
+  .asciz  "main"                          # string offset=232
+.Linfo_string_int:
+  .asciz  "int"                           # string offset=237
+  .ident  "clang version 12.0.0"
+  .section  ".note.GNU-stack","",@progbits
+  .addrsig
+
+  .section  .debug_line,"",@progbits
+
+.Lline_table_start0:
+        .long .Lline_table_end0 - 1f  # length from next byte
+1:
+        .short 4                      # version
+        .long .Lline_table_end0 - 2f  # prologue length from next byte
+2:
+        .byte 1                       # minimum instruction length
+        .byte 1                       # maximum ops per instruction
+        .byte 1                       # initial value of is_stmt
+        .byte -5                      # line base
+        .byte 14                      # line range
+        .byte 1                       # opcode base
+                                      # no standard opcodes
+        .asciz "directory0"
+        .byte 0                       # last directory
+        .asciz "filename0"
+        .uleb128 1                    # directory entry
+        .uleb128 0                    # time
+        .uleb128 0                    # file length
+        .byte 0                       # last filename
+.Lline_table_end0:
+
+.Lline_table_start1:
+        .long .Lline_table_end1 - 1f  # length from next byte
+1:
+        .short 4                      # version
+        .long .Lline_table_end1 - 2f  # prologue length from next byte
+2:
+        .byte 1                       # minimum instruction length
+        .byte 1                       # maximum ops per instruction
+        .byte 1                       # initial value of is_stmt
+        .byte -5                      # line base
+        .byte 14                      # line range
+        .byte 1                       # opcode base
+                                      # no standard opcodes
+        .asciz "directory1"
+        .byte 0                       # last directory
+        .asciz "filename1"
+        .uleb128 1                    # directory entry
+        .uleb128 0                    # time
+        .uleb128 0                    # file length
+        .byte 0                       # last filename
+.Lline_table_end1: