[lldb] Support DWARF-5 DW_FORM_line_strp (used by GCC)

[jankratochvil]

Summary

LLDB has been failing on binaries produced by new GCC 11.0 with -gdwarf-5.

[labath]

Looks good.

lldb/test/Shell/SymbolFile/DWARF/dwarf5-line-strp.s

51–54

If you make the function return type void, this will go away.

114–125

You can delete this.

[kwk]

@jankratochvil out of curiosity. Can you give a pointer to what the C program looks like that produces the object code? I know GCC might change and no longer produce this code so it's better to have the obj code instead. But still I'm not nearly close to understanding

[jankratochvil]

In D97721#2598798, @kwk wrote:

@jankratochvil out of curiosity. Can you give a pointer to what the C program looks like that produces the object code?

int main() {}
Fedora Rawhide: gcc -o main.s main.c -gdwarf-5 -S -gno-as-loc-support -dA

But it is hand edited, Besides DWARF simplification I removed the .text assembly instructions (and put .long 0 there) to make it arch-independent which was my original intention. But then I remember the DWARF assembly is not fully arch-independent anyway so to prevent some false failures in buildbots again I kept there that # REQUIRES: x86 anyway. One could try to remove the REQUIRES if one really wanted to test the bots. Still I am not sure if it did not cause any buildbots failure as the buildbots are down these days, incl. Apple Green Bot.

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

for (i = 0; i < numAttributes; ++i) {
             break;
 
           case DW_FORM_indirect:
             form_is_indirect = true;
             form = data.GetULEB128(&offset);
             break;
 
           case DW_FORM_strp:
+          case DW_FORM_line_strp:
           case DW_FORM_sec_offset:
             data.GetU32(&offset);
             break;
 
           case DW_FORM_implicit_const:
             form_size = 0;
             break;
 

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

 //===-- DWARFFormValue.cpp ------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include <assert.h>
 
 #include "lldb/Core/Module.h"
 #include "lldb/Core/dwarf.h"
 #include "lldb/Symbol/ObjectFile.h"
 #include "lldb/Utility/Stream.h"
 
 #include "DWARFDebugInfo.h"
 #include "DWARFFormValue.h"
 #include "DWARFUnit.h"
 
bool DWARFFormValue::ExtractValue(const DWARFDataExtractor &data,
 
   return true;
 }
 
 struct FormSize {
   uint8_t valid:1, size:7;
 };
 static FormSize g_form_sizes[] = {
-  {0,0}, // 0x00 unused
-  {0,0}, // 0x01 DW_FORM_addr
-  {0,0}, // 0x02 unused
-  {0,0}, // 0x03 DW_FORM_block2
-  {0,0}, // 0x04 DW_FORM_block4
-  {1,2}, // 0x05 DW_FORM_data2
-  {1,4}, // 0x06 DW_FORM_data4
-  {1,8}, // 0x07 DW_FORM_data8
-  {0,0}, // 0x08 DW_FORM_string
-  {0,0}, // 0x09 DW_FORM_block
-  {0,0}, // 0x0a DW_FORM_block1
-  {1,1}, // 0x0b DW_FORM_data1
-  {1,1}, // 0x0c DW_FORM_flag
-  {0,0}, // 0x0d DW_FORM_sdata
-  {1,4}, // 0x0e DW_FORM_strp
-  {0,0}, // 0x0f DW_FORM_udata
-  {0,0}, // 0x10 DW_FORM_ref_addr (addr size for DWARF2 and earlier, 4 bytes for
-         // DWARF32, 8 bytes for DWARF32 in DWARF 3 and later
-  {1,1}, // 0x11 DW_FORM_ref1
-  {1,2}, // 0x12 DW_FORM_ref2
-  {1,4}, // 0x13 DW_FORM_ref4
-  {1,8}, // 0x14 DW_FORM_ref8
-  {0,0}, // 0x15 DW_FORM_ref_udata
-  {0,0}, // 0x16 DW_FORM_indirect
-  {1,4}, // 0x17 DW_FORM_sec_offset
-  {0,0}, // 0x18 DW_FORM_exprloc
-  {1,0}, // 0x19 DW_FORM_flag_present
-  {0,0}, // 0x1a
-  {0,0}, // 0x1b
-  {0,0}, // 0x1c
-  {0,0}, // 0x1d
-  {0,0}, // 0x1e
-  {0,0}, // 0x1f
-  {1,8}, // 0x20 DW_FORM_ref_sig8
+    {0, 0}, // 0x00 unused
+    {0, 0}, // 0x01 DW_FORM_addr
+    {0, 0}, // 0x02 unused
+    {0, 0}, // 0x03 DW_FORM_block2
+    {0, 0}, // 0x04 DW_FORM_block4
+    {1, 2}, // 0x05 DW_FORM_data2
+    {1, 4}, // 0x06 DW_FORM_data4
+    {1, 8}, // 0x07 DW_FORM_data8
+    {0, 0}, // 0x08 DW_FORM_string
+    {0, 0}, // 0x09 DW_FORM_block
+    {0, 0}, // 0x0a DW_FORM_block1
+    {1, 1}, // 0x0b DW_FORM_data1
+    {1, 1}, // 0x0c DW_FORM_flag
+    {0, 0}, // 0x0d DW_FORM_sdata
+    {1, 4}, // 0x0e DW_FORM_strp
+    {0, 0}, // 0x0f DW_FORM_udata
+    {0, 0}, // 0x10 DW_FORM_ref_addr (addr size for DWARF2 and earlier, 4 bytes
+            // for DWARF32, 8 bytes for DWARF32 in DWARF 3 and later
+    {1, 1},  // 0x11 DW_FORM_ref1
+    {1, 2},  // 0x12 DW_FORM_ref2
+    {1, 4},  // 0x13 DW_FORM_ref4
+    {1, 8},  // 0x14 DW_FORM_ref8
+    {0, 0},  // 0x15 DW_FORM_ref_udata
+    {0, 0},  // 0x16 DW_FORM_indirect
+    {1, 4},  // 0x17 DW_FORM_sec_offset
+    {0, 0},  // 0x18 DW_FORM_exprloc
+    {1, 0},  // 0x19 DW_FORM_flag_present
+    {0, 0},  // 0x1a DW_FORM_strx (ULEB128)
+    {0, 0},  // 0x1b DW_FORM_addrx (ULEB128)
+    {1, 4},  // 0x1c DW_FORM_ref_sup4
+    {0, 0},  // 0x1d DW_FORM_strp_sup (4 bytes for DWARF32, 8 bytes for DWARF64)
+    {1, 16}, // 0x1e DW_FORM_data16
+    {1, 4},  // 0x1f DW_FORM_line_strp
+    {1, 8},  // 0x20 DW_FORM_ref_sig8
 };
 
 llvm::Optional<uint8_t>
 DWARFFormValue::GetFixedSize(dw_form_t form, const DWARFUnit *u) {
   if (form <= DW_FORM_ref_sig8 && g_form_sizes[form].valid)
     return g_form_sizes[form].size;
   if (form == DW_FORM_addr && u)
     return u->GetAddressByteSize();
case DW_FORM_implicit_const:
     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:
+    case DW_FORM_line_strp:
       *offset_ptr += 4;
       return true;
 
     // 4 byte values
     case DW_FORM_addrx4:
     case DW_FORM_data4:
     case DW_FORM_ref4:
     case DW_FORM_strx4:
case DW_FORM_block4:
     break;
 
   case DW_FORM_sdata:
     s.PutSLEB128(uvalue);
     break;
   case DW_FORM_udata:
     s.PutULEB128(uvalue);
     break;
-  case DW_FORM_strp: {
+  case DW_FORM_strp:
+  case DW_FORM_line_strp: {
     const char *dbg_str = AsCString();
     if (dbg_str) {
       s.QuotedCString(dbg_str);
     } else {
       s.PutHex32(uvalue);
     }
   } break;
 
switch (form) {
     case DW_FORM_data8:
     case DW_FORM_string:
     case DW_FORM_block:
     case DW_FORM_block1:
     case DW_FORM_data1:
     case DW_FORM_flag:
     case DW_FORM_sdata:
     case DW_FORM_strp:
+    case DW_FORM_line_strp:
     case DW_FORM_strx:
     case DW_FORM_strx1:
     case DW_FORM_strx2:
     case DW_FORM_strx3:
     case DW_FORM_strx4:
     case DW_FORM_udata:
     case DW_FORM_ref_addr:
     case DW_FORM_ref1:

lldb/test/Shell/SymbolFile/DWARF/dwarf5-line-strp.s

+# Test handling of DWARF5 DW_FORM_line_strp from .debug_info as used by GCC.
+
+# REQUIRES: x86
+
+# RUN: llvm-mc -filetype=obj -o %t -triple x86_64-pc-linux %s
+# RUN: %lldb %t -o "p main" \
+# RUN:   -o exit | FileCheck %s
+
+# CHECK: (int (*)()) $0 = 0x0000000000000000
+
+  .text
+.Ltext0:
+  .globl  main
+  .type main, @function
+main:
+.LFB0:
+.LM1:
+        .long 0
+.LM2:
+        .long 0
+.LFE0:
+  .size main, .-main
+.Letext0:
+  .section  .debug_info,"",@progbits
+.Ldebug_info0:
+  .long 0x50  # Length of Compilation Unit Info
+  .value  0x5 # DWARF version number
+  .byte 0x1 # DW_UT_compile
+  .byte 0x8 # Pointer Size (in bytes)
+  .long .Ldebug_abbrev0 # Offset Into Abbrev. Section
+  .uleb128 0x1  # (DIE (0xc) DW_TAG_compile_unit)
+  .long .LASF2  # DW_AT_producer: "GNU C17 11.0.0 20210210 (Red Hat 11.0.0-0) -mtune=generic -march=x86-64 -gdwarf-5 -gno-as-loc-support"
+  .byte 0x1d  # DW_AT_language
+  .long .LASF0  # DW_AT_name: "main.c"
+  .long .LASF1  # DW_AT_comp_dir: ""
+  .quad .Ltext0 # DW_AT_low_pc
+  .quad .Letext0-.Ltext0  # DW_AT_high_pc
+  .long .Ldebug_line0 # DW_AT_stmt_list
+  .uleb128 0x2  # (DIE (0x2e) DW_TAG_subprogram)
+      # DW_AT_external
+  .long .LASF3  # DW_AT_name: "main"
+  .byte 0x1 # DW_AT_decl_file (main.c)
+  .byte 0x1 # DW_AT_decl_line
+  .byte 0x5 # DW_AT_decl_column
+  .long 0x4c  # DW_AT_type
+  .quad .LFB0 # DW_AT_low_pc
+  .quad .LFE0-.LFB0 # DW_AT_high_pc
+  .uleb128 0x1  # DW_AT_frame_base
+  .byte 0x9c  # DW_OP_call_frame_cfa
+      # DW_AT_call_all_calls
+  .uleb128 0x3  # (DIE (0x4c) DW_TAG_base_type)
+  .byte 0x4 # DW_AT_byte_size
+  .byte 0x5 # DW_AT_encoding
+  .ascii "int\0"  # DW_AT_name
+  .byte 0 # end of children of DIE 0xc
+  .section  .debug_abbrev,"",@progbits
+.Ldebug_abbrev0:
+  .uleb128 0x1  # (abbrev code)
+  .uleb128 0x11 # (TAG: DW_TAG_compile_unit)
+  .byte 0x1 # DW_children_yes
+  .uleb128 0x25 # (DW_AT_producer)
+  .uleb128 0xe  # (DW_FORM_strp)
+  .uleb128 0x13 # (DW_AT_language)
+  .uleb128 0xb  # (DW_FORM_data1)
+  .uleb128 0x3  # (DW_AT_name)
+  .uleb128 0x1f # (DW_FORM_line_strp)
+  .uleb128 0x1b # (DW_AT_comp_dir)
+  .uleb128 0x1f # (DW_FORM_line_strp)
+  .uleb128 0x11 # (DW_AT_low_pc)
+  .uleb128 0x1  # (DW_FORM_addr)
+  .uleb128 0x12 # (DW_AT_high_pc)
+  .uleb128 0x7  # (DW_FORM_data8)
+  .uleb128 0x10 # (DW_AT_stmt_list)
+  .uleb128 0x17 # (DW_FORM_sec_offset)
+  .byte 0
+  .byte 0
+  .uleb128 0x2  # (abbrev code)
+  .uleb128 0x2e # (TAG: DW_TAG_subprogram)
+  .byte 0 # DW_children_no
+  .uleb128 0x3f # (DW_AT_external)
+  .uleb128 0x19 # (DW_FORM_flag_present)
+  .uleb128 0x3  # (DW_AT_name)
+  .uleb128 0xe  # (DW_FORM_strp)
+  .uleb128 0x3a # (DW_AT_decl_file)
+  .uleb128 0xb  # (DW_FORM_data1)
+  .uleb128 0x3b # (DW_AT_decl_line)
+  .uleb128 0xb  # (DW_FORM_data1)
+  .uleb128 0x39 # (DW_AT_decl_column)
+  .uleb128 0xb  # (DW_FORM_data1)
+  .uleb128 0x49 # (DW_AT_type)
+  .uleb128 0x13 # (DW_FORM_ref4)
+  .uleb128 0x11 # (DW_AT_low_pc)
+  .uleb128 0x1  # (DW_FORM_addr)
+  .uleb128 0x12 # (DW_AT_high_pc)
+  .uleb128 0x7  # (DW_FORM_data8)
+  .uleb128 0x40 # (DW_AT_frame_base)
+  .uleb128 0x18 # (DW_FORM_exprloc)
+  .uleb128 0x7a # (DW_AT_call_all_calls)
+  .uleb128 0x19 # (DW_FORM_flag_present)
+  .byte 0
+  .byte 0
+  .uleb128 0x3  # (abbrev code)
+  .uleb128 0x24 # (TAG: DW_TAG_base_type)
+  .byte 0 # DW_children_no
+  .uleb128 0xb  # (DW_AT_byte_size)
+  .uleb128 0xb  # (DW_FORM_data1)
+  .uleb128 0x3e # (DW_AT_encoding)
+  .uleb128 0xb  # (DW_FORM_data1)
+  .uleb128 0x3  # (DW_AT_name)
+  .uleb128 0x8  # (DW_FORM_string)
+  .byte 0
+  .byte 0
+  .byte 0
+  .section  .debug_aranges,"",@progbits
+  .long 0x2c  # Length of Address Ranges Info
+  .value  0x2 # DWARF aranges version
+  .long .Ldebug_info0 # Offset of Compilation Unit Info
+  .byte 0x8 # Size of Address
+  .byte 0 # Size of Segment Descriptor
+  .value  0 # Pad to 16 byte boundary
+  .value  0
+  .quad .Ltext0 # Address
+  .quad .Letext0-.Ltext0  # Length
+  .quad 0
+  .quad 0
+  .section  .debug_line,"",@progbits
+.Ldebug_line0:
+  .long .LELT0-.LSLT0 # Length of Source Line Info
+.LSLT0:
+  .value  0x5 # DWARF version number
+  .byte 0x8 # Address Size
+  .byte 0 # Segment Size
+  .long .LELTP0-.LASLTP0  # Prolog Length
+.LASLTP0:
+  .byte 0x1 # Minimum Instruction Length
+  .byte 0x1 # Maximum Operations Per Instruction
+  .byte 0x1 # Default is_stmt_start flag
+  .byte 0xf6  # Line Base Value (Special Opcodes)
+  .byte 0xf2  # Line Range Value (Special Opcodes)
+  .byte 0xd # Special Opcode Base
+  .byte 0 # opcode: 0x1 has 0 args
+  .byte 0x1 # opcode: 0x2 has 1 args
+  .byte 0x1 # opcode: 0x3 has 1 args
+  .byte 0x1 # opcode: 0x4 has 1 args
+  .byte 0x1 # opcode: 0x5 has 1 args
+  .byte 0 # opcode: 0x6 has 0 args
+  .byte 0 # opcode: 0x7 has 0 args
+  .byte 0 # opcode: 0x8 has 0 args
+  .byte 0x1 # opcode: 0x9 has 1 args
+  .byte 0 # opcode: 0xa has 0 args
+  .byte 0 # opcode: 0xb has 0 args
+  .byte 0x1 # opcode: 0xc has 1 args
+  .byte 0x1 # Directory entry format count
+  .uleb128 0x1  # DW_LNCT_path
+  .uleb128 0x1f # DW_FORM_line_strp
+  .uleb128 0x2  # Directories count
+  .long .LASF1  # Directory Entry: 0: ""
+  .long .LASF4  # Directory Entry: 0: ""
+  .byte 0x2 # File name entry format count
+  .uleb128 0x1  # DW_LNCT_path
+  .uleb128 0x1f # DW_FORM_line_strp
+  .uleb128 0x2  # DW_LNCT_directory_index
+  .uleb128 0xb  # DW_FORM_data1
+  .uleb128 0x2  # File names count
+  .long .LASF0  # File Entry: 0: "main.c"
+  .byte 0
+  .long .LASF5  # File Entry: 0: "main.c"
+  .byte 0x1
+.LELTP0:
+  .byte 0 # set address *.LM1
+  .uleb128 0x9
+  .byte 0x2
+  .quad .LM1
+  .byte 0x1 # copy line 1
+  .byte 0x5 # column 12
+  .uleb128 0xc  # 12
+  .byte 0 # set address *.LM2
+  .uleb128 0x9
+  .byte 0x2
+  .quad .LM2
+  .byte 0x1 # copy line 1
+  .byte 0x5 # column 13
+  .uleb128 0xd  # 13
+  .byte 0 # set address *.Letext0
+  .uleb128 0x9
+  .byte 0x2
+  .quad .Letext0
+  .byte 0 # end sequence
+  .uleb128 0x1
+  .byte 0x1
+.LELT0:
+  .section  .debug_str,"MS",@progbits,1
+.LASF2:
+  .string "GNU C17 11.0.0 20210210 (Red Hat 11.0.0-0) -mtune=generic -march=x86-64 -gdwarf-5 -gno-as-loc-support"
+.LASF3:
+  .string "main"
+  .section  .debug_line_str,"MS",@progbits,1
+.LASF1:
+  .string ""
+.LASF4:
+  .string ""
+.LASF0:
+  .string "main.c"
+.LASF5:
+  .string "main.c"
+  .ident  "GCC: (GNU) 11.0.0 20210210 (Red Hat 11.0.0-0)"
+  .section  .note.GNU-stack,"",@progbits