[DWARFLinker][DWARFv5] Add support for DW_FORM_addrx*

[avl]

Summary

This patch add support of DWARFv5 attribute forms: DW_FORM_addrx1,
DW_FORM_addrx2, DW_FORM_addrx3, DW_FORM_addrx4.

[clayborg]

Looks good to me.

[dblaikie]

llvm/lib/DWARFLinker/DWARFLinker.cpp

1374

I wouldn't stress about this - no debugger has support for it, it was a prototype for getting data and the version that was standardized is slightly different (takes offset as a uleb).

[avl]

llvm/lib/DWARFLinker/DWARFLinker.cpp

1374

I`ve left a comment to learn it better later. Will remove as it is not useful.

[avl]

Thanks for the review!

llvm/lib/DWARFLinker/DWARFLinker.cpp

if (InputDIE.getTag() == dwarf::DW_TAG_compile_unit &&
     if (uint64_t HighPc = Unit.getHighPc())
       Addr = HighPc;
     else
       return 0;
   } else {
     *Addr += Info.PCOffset;
   }
 
-  dwarf::Form Form = AttrSpec.Form;
-
+  switch (AttrSpec.Form) {
+  case dwarf::DW_FORM_addrx:
+  case dwarf::DW_FORM_addrx1:
+  case dwarf::DW_FORM_addrx2:
+  case dwarf::DW_FORM_addrx3:
+  case dwarf::DW_FORM_addrx4: {
     // DWARFLinker does not use addrx forms since it generates relocated
-  // addresses. Replace DW_FORM_addrx with DW_FORM_addr here.
-  if (Form == dwarf::DW_FORM_addrx)
-    Form = dwarf::DW_FORM_addr;
+    // addresses. Replace DW_FORM_addrx* with DW_FORM_addr here.
+    AttrSpec.Form = dwarf::DW_FORM_addr;
+    break;
+  }
+  default:
+    // Nothing to do.
+    break;
+  }
 
   Die.addValue(DIEAlloc, static_cast<dwarf::Attribute>(AttrSpec.Attr),
-               static_cast<dwarf::Form>(Form), DIEInteger(*Addr));
+               AttrSpec.Form, DIEInteger(*Addr));
   return Unit.getOrigUnit().getAddressByteSize();
 }
 
 unsigned DWARFLinker::DIECloner::cloneScalarAttribute(
     DIE &Die, const DWARFDie &InputDIE, const DWARFFile &File,
     CompileUnit &Unit, AttributeSpec AttrSpec, const DWARFFormValue &Val,
     unsigned AttrSize, AttributesInfo &Info) {
   uint64_t Value;
unsigned DWARFLinker::DIECloner::cloneAttribute(
   case dwarf::DW_FORM_block1:
   case dwarf::DW_FORM_block2:
   case dwarf::DW_FORM_block4:
   case dwarf::DW_FORM_exprloc:
     return cloneBlockAttribute(Die, File, Unit, AttrSpec, Val, AttrSize,
                                IsLittleEndian);
   case dwarf::DW_FORM_addr:
   case dwarf::DW_FORM_addrx:
+  case dwarf::DW_FORM_addrx1:
+  case dwarf::DW_FORM_addrx2:
+  case dwarf::DW_FORM_addrx3:
+  case dwarf::DW_FORM_addrx4:
     return cloneAddressAttribute(Die, InputDIE, AttrSpec, AttrSize, Val, Unit,
                                  Info);
   case dwarf::DW_FORM_data1:
   case dwarf::DW_FORM_data2:
   case dwarf::DW_FORM_data4:
   case dwarf::DW_FORM_data8:
   case dwarf::DW_FORM_udata:
   case dwarf::DW_FORM_sdata:
   case dwarf::DW_FORM_sec_offset:

llvm/test/tools/llvm-dwarfutil/ELF/X86/dwarf5-addrx.test

 # RUN: llvm-dwarfdump -verify %t1 | FileCheck %s
 # RUN: llvm-dwarfdump -a --verbose %t1 | FileCheck %s --check-prefix UPD-DWARF-CHECK
 
 #CHECK: No errors.
 
 #DWARF-CHECK: DW_TAG_compile_unit
 #DWARF-CHECK:   DW_AT_name [DW_FORM_strp]  {{.*}}     "CU1"
 #DWARF-CHECK:   DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001130)
-#DWARF-CHECK:   DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000030)
+#DWARF-CHECK:   DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000060)
 #DWARF-CHECK:   DW_TAG_subprogram
 #DWARF-CHECK:     DW_AT_name  [DW_FORM_strp] {{.*}}   "foo1"
 #DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001130)
 #DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
 #DWARF-CHECK:   DW_TAG_subprogram
 #DWARF-CHECK:     DW_AT_name  [DW_FORM_strp] {{.*}}   "foo2"
 #DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001140)
 #DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
 #DWARF-CHECK:   DW_TAG_subprogram
 #DWARF-CHECK:     DW_AT_name  [DW_FORM_strp] {{.*}}   "foo3"
 #DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001150)
 #DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
+#DWARF-CHECK:   DW_TAG_subprogram
+#DWARF-CHECK:     DW_AT_name  [DW_FORM_strp] {{.*}}   "foo4"
+#DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001160)
+#DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
+#DWARF-CHECK:   DW_TAG_subprogram
+#DWARF-CHECK:     DW_AT_name  [DW_FORM_strp] {{.*}}   "foo5"
+#DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001170)
+#DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
+#DWARF-CHECK:   DW_TAG_subprogram
+#DWARF-CHECK:     DW_AT_name  [DW_FORM_strp] {{.*}}   "foo6"
+#DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addr]     (0x0000000000001180)
+#DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
 #DWARF-CHECK=NOT: .debug_addr contents:
 
 #UPD-DWARF-CHECK: DW_TAG_compile_unit
 #UPD-DWARF-CHECK:   DW_AT_name {{.*}}"CU1"
 #UPD-DWARF-CHECK:   DW_AT_low_pc [DW_FORM_addrx]   (indexed (00000000) address = 0x0000000000001130)
-#UPD-DWARF-CHECK:   DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000030)
+#UPD-DWARF-CHECK:   DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000060)
 #UPD-DWARF-CHECK:   DW_TAG_subprogram
 #UPD-DWARF-CHECK:     DW_AT_name  {{.*}}"foo1"
 #UPD-DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addrx]   (indexed (00000000) address = 0x0000000000001130)
 #UPD-DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
 #UPD-DWARF-CHECK:   DW_TAG_subprogram
 #UPD-DWARF-CHECK:     DW_AT_name  {{.*}}"foo2"
 #UPD-DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addrx]   (indexed (00000001) address = 0x0000000000001140)
 #UPD-DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
 #UPD-DWARF-CHECK:   DW_TAG_subprogram
 #UPD-DWARF-CHECK:     DW_AT_name  {{.*}}"foo3"
 #UPD-DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addrx]   (indexed (00000002) address = 0x0000000000001150)
 #UPD-DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
+#UPD-DWARF-CHECK:   DW_TAG_subprogram
+#UPD-DWARF-CHECK:     DW_AT_name  {{.*}}"foo4"
+#UPD-DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addrx1]   (indexed (00000003) address = 0x0000000000001160)
+#UPD-DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
+#UPD-DWARF-CHECK:   DW_TAG_subprogram
+#UPD-DWARF-CHECK:     DW_AT_name  {{.*}}"foo5"
+#UPD-DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addrx2]   (indexed (00000004) address = 0x0000000000001170)
+#UPD-DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
+#UPD-DWARF-CHECK:   DW_TAG_subprogram
+#UPD-DWARF-CHECK:     DW_AT_name  {{.*}}"foo6"
+#UPD-DWARF-CHECK:     DW_AT_low_pc [DW_FORM_addrx4]   (indexed (00000005) address = 0x0000000000001180)
+#UPD-DWARF-CHECK:     DW_AT_high_pc [DW_FORM_data8]   (0x0000000000000010)
 #UPD-DWARF-CHECK: .debug_addr contents:
-#UPD-DWARF-CHECK: 0x00000000: Address table header: length = 0x0000001c, format = DWARF32, version = 0x0005, addr_size = 0x08, seg_size = 0x00
+#UPD-DWARF-CHECK: 0x00000000: Address table header: length = 0x00000034, format = DWARF32, version = 0x0005, addr_size = 0x08, seg_size = 0x00
 #UPD-DWARF-CHECK: Addrs: [
 #UPD-DWARF-CHECK: 0x0000000000001130
 #UPD-DWARF-CHECK: 0x0000000000001140
 #UPD-DWARF-CHECK: 0x0000000000001150
+#UPD-DWARF-CHECK: 0x0000000000001160
+#UPD-DWARF-CHECK: 0x0000000000001170
+#UPD-DWARF-CHECK: 0x0000000000001180
 #UPD-DWARF-CHECK: ]
 
 --- !ELF
 FileHeader:
   Class:    ELFCLASS64
   Data:     ELFDATA2LSB
   Type:     ET_REL
   Machine:  EM_X86_64
 Sections:
   - Name:            .text
     Type:            SHT_PROGBITS
     Flags:           [ SHF_ALLOC, SHF_EXECINSTR ]
     Address:         0x1130
-    Size:            0x30
+    Size:            0x60
 DWARF:
   debug_abbrev:
     - Table:
       - Tag:      DW_TAG_compile_unit
         Children: DW_CHILDREN_yes
         Attributes:
           - Attribute: DW_AT_producer
             Form:      DW_FORM_string
- Table:
           - Attribute: DW_AT_name
             Form:      DW_FORM_string
           - Attribute: DW_AT_low_pc
             Form:      DW_FORM_addrx
           - Attribute: DW_AT_high_pc
             Form:      DW_FORM_data8
           - Attribute: DW_AT_type
             Form:      DW_FORM_ref4
+      - Tag:      DW_TAG_subprogram
+        Children: DW_CHILDREN_yes
+        Attributes:
+          - Attribute: DW_AT_name
+            Form:      DW_FORM_string
+          - Attribute: DW_AT_low_pc
+            Form:      DW_FORM_addrx1
+          - Attribute: DW_AT_high_pc
+            Form:      DW_FORM_data8
+          - Attribute: DW_AT_type
+            Form:      DW_FORM_ref4
+      - Tag:      DW_TAG_subprogram
+        Children: DW_CHILDREN_yes
+        Attributes:
+          - Attribute: DW_AT_name
+            Form:      DW_FORM_string
+          - Attribute: DW_AT_low_pc
+            Form:      DW_FORM_addrx2
+          - Attribute: DW_AT_high_pc
+            Form:      DW_FORM_data8
+          - Attribute: DW_AT_type
+            Form:      DW_FORM_ref4
+      - Tag:      DW_TAG_subprogram
+        Children: DW_CHILDREN_yes
+        Attributes:
+          - Attribute: DW_AT_name
+            Form:      DW_FORM_string
+          - Attribute: DW_AT_low_pc
+            Form:      DW_FORM_addrx4
+          - Attribute: DW_AT_high_pc
+            Form:      DW_FORM_data8
+          - Attribute: DW_AT_type
+            Form:      DW_FORM_ref4
       - Tag:      DW_TAG_base_type
         Children: DW_CHILDREN_no
         Attributes:
           - Attribute: DW_AT_name
             Form:      DW_FORM_string
   debug_info:
     - Version: 5
       UnitType:   DW_UT_compile
       Entries:
         - AbbrCode: 1
           Values:
             - CStr: by_hand
             - Value:  0x04
             - CStr: CU1
             - Value:  0x0
-            - Value:  0x30
+            - Value:  0x60
             - Value:  0x8
         - AbbrCode: 2
           Values:
             - CStr: foo1
             - Value: 0x0
             - Value: 0x10
-            - Value: 0x64
+            - Value: 0xa4
         - AbbrCode: 0
         - AbbrCode: 2
           Values:
             - CStr: foo2
             - Value: 0x01
             - Value: 0x10
-            - Value: 0x64
+            - Value: 0xa4
         - AbbrCode: 0
         - AbbrCode: 2
           Values:
             - CStr: foo3
             - Value: 0x02
             - Value: 0x10
-            - Value: 0x64
+            - Value: 0xa4
         - AbbrCode: 0
         - AbbrCode: 3
           Values:
+            - CStr: foo4
+            - Value: 0x03
+            - Value: 0x10
+            - Value: 0xa4
+        - AbbrCode: 0
+        - AbbrCode: 4
+          Values:
+            - CStr: foo5
+            - Value: 0x04
+            - Value: 0x10
+            - Value: 0xa4
+        - AbbrCode: 0
+        - AbbrCode: 5
+          Values:
+            - CStr: foo6
+            - Value: 0x05
+            - Value: 0x10
+            - Value: 0xa4
+        - AbbrCode: 0
+        - AbbrCode: 6
+          Values:
             - CStr: int
         - AbbrCode: 0
   debug_addr:
     - Version: 5
       AddressSize: 0x08
       Entries:
         - Address: 0x1130
         - Address: 0x1140
         - Address: 0x1150
+        - Address: 0x1160
+        - Address: 0x1170
+        - Address: 0x1180
 ...

llvm/tools/dsymutil/DwarfLinkerForBinary.cpp

bool DwarfLinkerForBinary::AddressManager::isLiveSubprogram(
 
   std::optional<uint32_t> LowPcIdx =
       Abbrev->findAttributeIndex(dwarf::DW_AT_low_pc);
   if (!LowPcIdx)
     return false;
 
   dwarf::Form Form = Abbrev->getFormByIndex(*LowPcIdx);
 
-  if (Form == dwarf::DW_FORM_addr) {
+  switch (Form) {
+  case dwarf::DW_FORM_addr: {
     uint64_t Offset = DIE.getOffset() + getULEB128Size(Abbrev->getCode());
     uint64_t LowPcOffset, LowPcEndOffset;
     std::tie(LowPcOffset, LowPcEndOffset) =
         getAttributeOffsets(Abbrev, *LowPcIdx, Offset, *DIE.getDwarfUnit());
     return hasValidRelocationAt(ValidDebugInfoRelocs, LowPcOffset,
                                 LowPcEndOffset, MyInfo);
   }
-
-  if (Form == dwarf::DW_FORM_addrx) {
+  case dwarf::DW_FORM_addrx:
+  case dwarf::DW_FORM_addrx1:
+  case dwarf::DW_FORM_addrx2:
+  case dwarf::DW_FORM_addrx3:
+  case dwarf::DW_FORM_addrx4: {
     std::optional<DWARFFormValue> AddrValue = DIE.find(dwarf::DW_AT_low_pc);
     if (std::optional<uint64_t> AddrOffsetSectionBase =
             DIE.getDwarfUnit()->getAddrOffsetSectionBase()) {
       uint64_t StartOffset = *AddrOffsetSectionBase + AddrValue->getRawUValue();
       uint64_t EndOffset =
           StartOffset + DIE.getDwarfUnit()->getAddressByteSize();
       return hasValidRelocationAt(ValidDebugAddrRelocs, StartOffset, EndOffset,
                                   MyInfo);
-    } else
-      Linker.reportWarning("no base offset for address table", SrcFileName);
     }
 
+    Linker.reportWarning("no base offset for address table", SrcFileName);
     return false;
   }
+  default:
+    return false;
+  }
+}
 
 uint64_t
 DwarfLinkerForBinary::AddressManager::relocate(const ValidReloc &Reloc) const {
   return Reloc.Mapping->getValue().BinaryAddress + Reloc.Addend;
 }
 
 /// Apply the valid relocations found by findValidRelocs() to
 /// the buffer \p Data, taking into account that Data is at \p BaseOffset