DWARF parser: Fix DWARF-2/3 incompatibility: size of DW_FORM_ref_addr is the same as DW_FORM_addr in DWARF2, and is 4/8 bytes on 32/64-bit DWARF starting from DWARF3. Adding a test for this is a huge pain - generating and uploading pre-built binary with DWARF3 debug info is way too ugly, and writing fine-grained unittests for DebugInfo is impossible, as it doesn't expose any headers in include/llvm. That said, I'm going to choose the second approach and submit the patch exposing DebugInfo headers for review soon enough.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179095 91177308-0d34-0410-b5e6-96231b3b80d8

lib/DebugInfo/DWARFCompileUnit.cpp

@@ -165,7 +165,7 @@ size_t DWARFCompileUnit::extractDIEsIfNeeded(bool cu_die_only) {
   // we were told to parse
 
   const uint8_t *fixed_form_sizes =
-    DWARFFormValue::getFixedFormSizesForAddressSize(getAddressByteSize());
+    DWARFFormValue::getFixedFormSizes(getAddressByteSize(), getVersion());
 
   while (offset < next_cu_offset &&
          die.extractFast(this, fixed_form_sizes, &offset)) {

lib/DebugInfo/DWARFFormValue.cpp

@@ -18,9 +18,15 @@
 using namespace llvm;
 using namespace dwarf;
 
-static const uint8_t form_sizes_addr4[] = {
+template <uint8_t AddrSize, uint8_t RefAddrSize>
+struct FixedFormSizes {
+  static const uint8_t sizes[];
+};
+
+template <uint8_t AddrSize, uint8_t RefAddrSize>
+const uint8_t FixedFormSizes<AddrSize, RefAddrSize>::sizes[] = {
   0, // 0x00 unused
-  4, // 0x01 DW_FORM_addr
+  AddrSize, // 0x01 DW_FORM_addr
   0, // 0x02 unused
   0, // 0x03 DW_FORM_block2
   0, // 0x04 DW_FORM_block4
@@ -35,7 +41,7 @@ static const uint8_t form_sizes_addr4[] = {
   0, // 0x0d DW_FORM_sdata
   4, // 0x0e DW_FORM_strp
   0, // 0x0f DW_FORM_udata
-  4, // 0x10 DW_FORM_ref_addr
+  RefAddrSize, // 0x10 DW_FORM_ref_addr
   1, // 0x11 DW_FORM_ref1
   2, // 0x12 DW_FORM_ref2
   4, // 0x13 DW_FORM_ref4
@@ -48,43 +54,23 @@ static const uint8_t form_sizes_addr4[] = {
   8, // 0x20 DW_FORM_ref_sig8
 };
 
-static const uint8_t form_sizes_addr8[] = {
+static uint8_t getRefAddrSize(uint8_t AddrSize, uint16_t Version) {
-  0, // 0x00 unused
+  // FIXME: Support DWARF64.
-  8, // 0x01 DW_FORM_addr
+  return (Version == 2) ? AddrSize : 4;
-  0, // 0x02 unused
+}
-  0, // 0x03 DW_FORM_block2
-  0, // 0x04 DW_FORM_block4
-  2, // 0x05 DW_FORM_data2
-  4, // 0x06 DW_FORM_data4
-  8, // 0x07 DW_FORM_data8
-  0, // 0x08 DW_FORM_string
-  0, // 0x09 DW_FORM_block
-  0, // 0x0a DW_FORM_block1
-  1, // 0x0b DW_FORM_data1
-  1, // 0x0c DW_FORM_flag
-  0, // 0x0d DW_FORM_sdata
-  4, // 0x0e DW_FORM_strp
-  0, // 0x0f DW_FORM_udata
-  8, // 0x10 DW_FORM_ref_addr
-  1, // 0x11 DW_FORM_ref1
-  2, // 0x12 DW_FORM_ref2
-  4, // 0x13 DW_FORM_ref4
-  8, // 0x14 DW_FORM_ref8
-  0, // 0x15 DW_FORM_ref_udata
-  0, // 0x16 DW_FORM_indirect
-  4, // 0x17 DW_FORM_sec_offset
-  0, // 0x18 DW_FORM_exprloc
-  0, // 0x19 DW_FORM_flag_present
-  8, // 0x20 DW_FORM_ref_sig8
-};
 
 const uint8_t *
-DWARFFormValue::getFixedFormSizesForAddressSize(uint8_t addr_size) {
+DWARFFormValue::getFixedFormSizes(uint8_t AddrSize, uint16_t Version) {
-  switch (addr_size) {
+  uint8_t RefAddrSize = getRefAddrSize(AddrSize, Version);
-  case 4: return form_sizes_addr4;
+  if (AddrSize == 4 && RefAddrSize == 4)
-  case 8: return form_sizes_addr8;
+    return FixedFormSizes<4, 4>::sizes;
-  }
+  if (AddrSize == 4 && RefAddrSize == 8)
-  return NULL;
+    return FixedFormSizes<4, 8>::sizes;
+  if (AddrSize == 8 && RefAddrSize == 4)
+    return FixedFormSizes<8, 4>::sizes;
+  if (AddrSize == 8 && RefAddrSize == 8)
+    return FixedFormSizes<8, 8>::sizes;
+  return 0;
 }
 
 bool
@@ -100,14 +86,16 @@ DWARFFormValue::extractValue(DataExtractor data, uint32_t *offset_ptr,
     switch (Form) {
     case DW_FORM_addr:
     case DW_FORM_ref_addr: {
-      RelocAddrMap::const_iterator AI
+      uint16_t AddrSize =
-        = cu->getRelocMap()->find(*offset_ptr);
+          (Form == DW_FORM_addr)
+              ? cu->getAddressByteSize()
+              : getRefAddrSize(cu->getAddressByteSize(), cu->getVersion());
+      RelocAddrMap::const_iterator AI = cu->getRelocMap()->find(*offset_ptr);
       if (AI != cu->getRelocMap()->end()) {
         const std::pair<uint8_t, int64_t> &R = AI->second;
-        Value.uval = data.getUnsigned(offset_ptr, cu->getAddressByteSize()) +
+        Value.uval = data.getUnsigned(offset_ptr, AddrSize) + R.second;
-                     R.second;
       } else
-        Value.uval = data.getUnsigned(offset_ptr, cu->getAddressByteSize());
+        Value.uval = data.getUnsigned(offset_ptr, AddrSize);
       break;
     }
     case DW_FORM_exprloc:
@@ -256,9 +244,11 @@ DWARFFormValue::skipValue(uint16_t form, DataExtractor debug_info_data,
 
     // Compile unit address sized values
     case DW_FORM_addr:
-    case DW_FORM_ref_addr:
       *offset_ptr += cu->getAddressByteSize();
       return true;
+    case DW_FORM_ref_addr:
+      *offset_ptr += getRefAddrSize(cu->getAddressByteSize(), cu->getVersion());
+      return true;
 
     // 0 byte values - implied from the form.
     case DW_FORM_flag_present:

lib/DebugInfo/DWARFFormValue.h

@@ -74,7 +74,7 @@ public:
                         uint32_t *offset_ptr, const DWARFCompileUnit *cu);
   static bool isBlockForm(uint16_t form);
   static bool isDataForm(uint16_t form);
-  static const uint8_t *getFixedFormSizesForAddressSize(uint8_t addr_size);
+  static const uint8_t *getFixedFormSizes(uint8_t AddrSize, uint16_t Version);
 };
 
 }