[Object][Wasm] Use offset instead of index for Global address and store size (#81781)

Currently the address reported by binutils for a global is its index;
but its offset (in the file or section) is more useful for binary size
attribution.
This PR treats globals similarly to functions, and tracks their offset
and size. It also centralizes the logic differentiating linked from object
and dylib files (where section addresses are 0).

Author: dschuff

llvm/include/llvm/BinaryFormat/Wasm.h

@@ -350,6 +350,8 @@ struct WasmGlobal {
   WasmGlobalType Type;
   WasmInitExpr InitExpr;
   StringRef SymbolName; // from the "linking" section
+  uint32_t Offset; // Offset of the definition in the binary's Global section
+  uint32_t Size;   // Size of the definition in the binary's Global section
 };
 
 struct WasmTag {

llvm/include/llvm/Object/Wasm.h

@@ -239,7 +239,7 @@ class WasmObjectFile : public ObjectFile {
   bool isValidSectionSymbol(uint32_t Index) const;
   wasm::WasmFunction &getDefinedFunction(uint32_t Index);
   const wasm::WasmFunction &getDefinedFunction(uint32_t Index) const;
-  wasm::WasmGlobal &getDefinedGlobal(uint32_t Index);
+  const wasm::WasmGlobal &getDefinedGlobal(uint32_t Index) const;
   wasm::WasmTag &getDefinedTag(uint32_t Index);
 
   const WasmSection &getWasmSection(DataRefImpl Ref) const;

llvm/lib/Object/WasmObjectFile.cpp

@@ -1400,18 +1400,20 @@ Error WasmObjectFile::parseTagSection(ReadContext &Ctx) {
 
 Error WasmObjectFile::parseGlobalSection(ReadContext &Ctx) {
   GlobalSection = Sections.size();
+  const uint8_t *SectionStart = Ctx.Ptr;
   uint32_t Count = readVaruint32(Ctx);
   Globals.reserve(Count);
   while (Count--) {
     wasm::WasmGlobal Global;
     Global.Index = NumImportedGlobals + Globals.size();
+    const uint8_t *GlobalStart = Ctx.Ptr;
+    Global.Offset = static_cast<uint32_t>(GlobalStart - SectionStart);
     auto GlobalOpcode = readVaruint32(Ctx);
-    auto GlobalType = parseValType(Ctx, GlobalOpcode);
-    // assert(GlobalType <= std::numeric_limits<wasm::ValType>::max());
-    Global.Type.Type = (uint8_t)GlobalType;
+    Global.Type.Type = (uint8_t)parseValType(Ctx, GlobalOpcode);
     Global.Type.Mutable = readVaruint1(Ctx);
     if (Error Err = readInitExpr(Global.InitExpr, Ctx))
       return Err;
+    Global.Size = static_cast<uint32_t>(Ctx.Ptr - GlobalStart);
     Globals.push_back(Global);
   }
   if (Ctx.Ptr != Ctx.End)
@@ -1564,7 +1566,7 @@ WasmObjectFile::getDefinedFunction(uint32_t Index) const {
   return Functions[Index - NumImportedFunctions];
 }
 
-wasm::WasmGlobal &WasmObjectFile::getDefinedGlobal(uint32_t Index) {
+const wasm::WasmGlobal &WasmObjectFile::getDefinedGlobal(uint32_t Index) const {
   assert(isDefinedGlobalIndex(Index));
   return Globals[Index - NumImportedGlobals];
 }
@@ -1815,18 +1817,22 @@ Expected<StringRef> WasmObjectFile::getSymbolName(DataRefImpl Symb) const {
 
 Expected<uint64_t> WasmObjectFile::getSymbolAddress(DataRefImpl Symb) const {
   auto &Sym = getWasmSymbol(Symb);
+  if (!Sym.isDefined())
+    return 0;
+  Expected<section_iterator> Sec = getSymbolSection(Symb);
+  if (!Sec)
+    return Sec.takeError();
+  uint32_t SectionAddress = getSectionAddress(Sec.get()->getRawDataRefImpl());
   if (Sym.Info.Kind == wasm::WASM_SYMBOL_TYPE_FUNCTION &&
       isDefinedFunctionIndex(Sym.Info.ElementIndex)) {
-    // For object files, use the section offset. The linker relies on this.
-    // For linked files, use the file offset. This behavior matches the way
-    // browsers print stack traces and is useful for binary size analysis.
-    // (see https://webassembly.github.io/spec/web-api/index.html#conventions)
-    uint32_t Adjustment = isRelocatableObject() || isSharedObject()
-                              ? 0
-                              : Sections[CodeSection].Offset;
     return getDefinedFunction(Sym.Info.ElementIndex).CodeSectionOffset +
-           Adjustment;
+           SectionAddress;
   }
+  if (Sym.Info.Kind == wasm::WASM_SYMBOL_TYPE_GLOBAL &&
+      isDefinedGlobalIndex(Sym.Info.ElementIndex)) {
+    return getDefinedGlobal(Sym.Info.ElementIndex).Offset + SectionAddress;
+  }
+
   return getSymbolValue(Symb);
 }
 
@@ -1936,6 +1942,8 @@ uint32_t WasmObjectFile::getSymbolSize(SymbolRef Symb) const {
   const WasmSymbol &Sym = getWasmSymbol(Symb);
   if (!Sym.isDefined())
     return 0;
+  if (Sym.isTypeGlobal())
+    return getDefinedGlobal(Sym.Info.ElementIndex).Size;
   if (Sym.isTypeData())
     return Sym.Info.DataRef.Size;
   if (Sym.isTypeFunction())

llvm/test/tools/llvm-nm/wasm/exports.yaml

@@ -64,4 +64,4 @@ Sections:
 
 # CHECK:      00000000 D dexport
 # CHECK-NEXT: 00000001 T fexport
-# CHECK-NEXT: 00000000 D gexport
+# CHECK-NEXT: 00000001 D gexport

llvm/test/tools/llvm-nm/wasm/weak-symbols.yaml

@@ -80,6 +80,6 @@ Sections:
 
 # CHECK:      00000000 W weak_defined_data
 # CHECK-NEXT: 00000001 W weak_defined_func
-# CHECK-NEXT: 00000000 W weak_defined_global
+# CHECK-NEXT: 00000001 W weak_defined_global
 # CHECK-NEXT:          w weak_import_data
 # CHECK-NEXT:          w weak_import_func

llvm/test/tools/llvm-objdump/wasm/linked-symbol-table-namesec.yaml

@@ -3,10 +3,11 @@
 #
 # CHECK:      SYMBOL TABLE:
 # CHECK-NEXT: 00000000   F *UND* 00000000 my_func_import_name
-# CHECK-NEXT: 00000083 g F CODE 00000003 my_func_export_name
-# CHECK-NEXT: 00000086 l F CODE 00000003 my_func_local_name
+# CHECK-NEXT: 0000008c g F CODE 00000003 my_func_export_name
+# CHECK-NEXT: 0000008f l F CODE 00000003 my_func_local_name
 # CHECK-NEXT: 00000000    *UND* 00000000 my_global_import_name
-# CHECK-NEXT: 00000001 g  GLOBAL 00000000 my_global_export_name
+# CHECK-NEXT: 0000004c g  GLOBAL 00000005 my_global_export_name
+# CHECK-NEXT: 00000051 g  GLOBAL 00000009 my_global_local_name
 # CHECK-NEXT: 00000000 l O DATA 00000004 my_datasegment_name
 
 --- !WASM
@@ -44,6 +45,12 @@ Sections:
         InitExpr:
           Opcode:          I32_CONST
           Value:           42
+      - Index:           2
+        Mutable:         true
+        Type:            I64
+        InitExpr:
+          Opcode:          I64_CONST
+          Value:           5000000000
   - Type:            EXPORT
     Exports:
       - Name:            my_func_export
@@ -82,6 +89,8 @@ Sections:
        Name:         my_global_import_name
      - Index:        1
        Name:         my_global_export_name
+     - Index:        2
+       Name:         my_global_local_name
     DataSegmentNames:
      - Index:        0
        Name:         my_datasegment_name