[lld-macho] Support creating N_SO stab for DWARF5 compile units

[BertalanD]

Summary

In DWARF5, the DW_AT_name and DW_AT_comp_dir attributes are encoded
using the strx* forms, which specify an index into __debug_str_offs.
This commit adds that section to DwarfObject, so the debug info parser
can resolve these references.

The test case was manually adapted from stabs-icf.s.

Fixes https://github.com/llvm/llvm-project/issues/51668

[thakis]

Per chat: Given that dwarf5 doesn't currently work in lldb on mac, and probably doesn't work in dsymutil, I'm not sure how useful it is to support this in the linker if the platform (apparently) isn't interested in it.

What does ld64 do with dwarf5 info?

[BertalanD]

What does ld64 do with dwarf5 info?

ld64's dwarf parser is able to understand the strx* forms: https://github.com/apple-opensource/ld64/blob/e28c028b20af187a16a7161d89e91868a450cadc/src/ld/parsers/macho_relocatable_file.cpp#L3721-L3762. Running my test though ld64 produces a sensible result.

[int3]

@phosek what's the DWARF5 use case that motivated https://github.com/llvm/llvm-project/issues/51668?

[phosek]

In D130559#3681289, @int3 wrote:

@phosek what's the DWARF5 use case that motivated https://github.com/llvm/llvm-project/issues/51668?

We use DWARF5 on all platforms except for macOS so mostly consistency. We use debug info primarily with LLVM tools like llvm-readobj so the lack of support in dsymutil or lldb is less of a concern to us.

[int3]

lgtm

lld/test/MachO/stabs-dwarf5.s

28

could you include in a comment the C source used to generate this assembly?

[BertalanD]

lld/test/MachO/stabs-dwarf5.s

28

There isn't really one. The debug info was modified by hand from stabs-icf.s.

[thakis]

Meh :)

lld/MachO/Dwarf.h

public:
 
   llvm::StringRef getAbbrevSection() const override { return abbrevSection; }
   llvm::StringRef getStrSection() const override { return strSection; }
 
   llvm::DWARFSection const &getLineSection() const override {
     return lineSection;
   }
 
+  llvm::DWARFSection const &getStrOffsetsSection() const override {
+    return strOffsSection;
+  }
+
   // Returns an instance of DwarfObject if the given object file has the
   // relevant DWARF debug sections.
   static std::unique_ptr<DwarfObject> create(ObjFile *);
 
 private:
   llvm::DWARFSection infoSection;
   llvm::DWARFSection lineSection;
+  llvm::DWARFSection strOffsSection;
   llvm::StringRef abbrevSection;
   llvm::StringRef strSection;
 };
 
 } // namespace macho
 } // namespace lld
 
 #endif

lld/MachO/Dwarf.cpp

std::unique_ptr<DwarfObject> DwarfObject::create(ObjFile *obj) {
   // to get that path. The debugger will locate the debug info via the object
   // file paths that we emit in our STABS symbols, so we don't need to process &
   // emit them ourselves.
   for (const InputSection *isec : obj->debugSections) {
     if (StringRef *s =
             StringSwitch<StringRef *>(isec->getName())
                 .Case(section_names::debugInfo, &dObj->infoSection.Data)
                 .Case(section_names::debugLine, &dObj->lineSection.Data)
+                .Case(section_names::debugStrOffs, &dObj->strOffsSection.Data)
                 .Case(section_names::debugAbbrev, &dObj->abbrevSection)
                 .Case(section_names::debugStr, &dObj->strSection)
                 .Default(nullptr)) {
       *s = toStringRef(isec->data);
       hasDwarfInfo = true;
     }
   }
 
   if (hasDwarfInfo)
     return dObj;
   return nullptr;
 }

lld/MachO/InputSection.h

 constexpr const char codeSignature[] = "__code_signature";
 constexpr const char common[] = "__common";
 constexpr const char compactUnwind[] = "__compact_unwind";
 constexpr const char data[] = "__data";
 constexpr const char debugAbbrev[] = "__debug_abbrev";
 constexpr const char debugInfo[] = "__debug_info";
 constexpr const char debugLine[] = "__debug_line";
 constexpr const char debugStr[] = "__debug_str";
+constexpr const char debugStrOffs[] = "__debug_str_offs";
 constexpr const char ehFrame[] = "__eh_frame";
 constexpr const char gccExceptTab[] = "__gcc_except_tab";
 constexpr const char export_[] = "__export";
 constexpr const char dataInCode[] = "__data_in_code";
 constexpr const char functionStarts[] = "__func_starts";
 constexpr const char got[] = "__got";
 constexpr const char header[] = "__mach_header";
 constexpr const char indirectSymbolTable[] = "__ind_sym_tab";

lld/test/MachO/stabs-dwarf5.s

+# REQUIRES: x86
+# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin -dwarf-version=5 %s -o %t.o
+
+# RUN: %lld -lSystem %t.o -o %t
+# RUN: dsymutil -s %t | FileCheck %s -DDIR=%t -DSRC_PATH=%t.o
+
+# CHECK:      (N_SO         ) 00      0000   0000000000000000   '/tmp{{[/\\]}}test.cpp'
+# CHECK-NEXT: (N_OSO        ) 03      0001   {{.*}} '[[SRC_PATH]]'
+# CHECK-NEXT: (N_FUN        ) 01      0000   [[#%.16x,MAIN:]]   '_main'
+# CHECK-NEXT: (N_FUN        ) 00      0000   0000000000000001{{$}}
+# CHECK-DAG:  (     SECT EXT) 01      0000   [[#MAIN]]           '_main'
+# CHECK-DAG:  (       {{.*}}) {{[0-9]+}}                 0010   {{[0-9a-f]+}}      '__mh_execute_header'
+# CHECK-DAG:  (       {{.*}}) {{[0-9]+}}                 0100   0000000000000000   'dyld_stub_binder'
+# CHECK-EMPTY:
+
+.text
+.globl _main
+
+.subsections_via_symbols
+
+_main:
+Lfunc_begin0:
+  retq
+Lfunc_end0:
+
+.section  __DWARF,__debug_str_offs,regular,debug
+Lsection_str_off:
+  .long   12                     ## Length of String Offsets Set
+  .short  5
+  .short  0
+Lstr_offsets_base0:
+.section  __DWARF,__debug_str,regular,debug
+  .asciz  "test.cpp"             ## string offset=0
+  .asciz  "/tmp"                 ## string offset=9
+.section __DWARF,__debug_str_offs,regular,debug
+  .long  0
+  .long  9
+.section  __DWARF,__debug_abbrev,regular,debug
+Lsection_abbrev:
+  .byte  1                       ## Abbreviation Code
+  .byte  17                      ## DW_TAG_compile_unit
+  .byte  1                       ## DW_CHILDREN_yes
+  .byte  17                      ## DW_AT_low_pc
+  .byte  1                       ## DW_FORM_addr
+  .byte  18                      ## DW_AT_high_pc
+  .byte  6                       ## DW_FORM_data4
+  .byte  3                       ## DW_AT_name
+  .byte  37                      ## DW_FORM_strx1
+  .byte  27                      ## DW_AT_comp_dir
+  .byte  37                      ## DW_FORM_strx1
+  .byte  114                     ## DW_AT_str_offsets_base
+  .byte  23                      ## DW_FORM_sec_offset
+  .byte  0                       ## EOM(1)
+.section  __DWARF,__debug_info,regular,debug
+.set Lset0, Ldebug_info_end0-Ldebug_info_start0 ## Length of Unit
+  .long  Lset0
+Ldebug_info_start0:
+  .short  5                       ## DWARF version number
+  .byte   1                       ## DWARF Unit Type
+  .byte   8                       ## Address Size (in bytes)
+.set Lset1, Lsection_abbrev-Lsection_abbrev ## Offset Into Abbrev. Section
+  .long  Lset1
+  .byte  1                       ## Abbrev [1] 0xb:0x48 DW_TAG_compile_unit
+  .quad  Lfunc_begin0            ## DW_AT_low_pc
+.set Lset3, Lfunc_end0-Lfunc_begin0     ## DW_AT_high_pc
+  .long  Lset3
+  .byte  0                       ## DW_AT_name
+  .byte  1                       ## DW_AT_comp_dir
+.set Lset4, Lstr_offsets_base0-Lsection_str_off  ## DW_AT_str_offsets_base
+  .long Lset4
+  .byte  0                       ## End Of Children Mark
+Ldebug_info_end0: