[lld] [llvm] [Object][Wasm] Generate symbol info from name section names (PR #81063)

[Derek Schuff via llvm-commits]

================
@@ -529,26 +546,50 @@ Error WasmObjectFile::parseNameSection(ReadContext &Ctx) {
             return make_error<GenericBinaryError>("invalid function name entry",
                                                   object_error::parse_failed);
 
-          if (isDefinedFunctionIndex(Index))
-            getDefinedFunction(Index).DebugName = Name;
+          if (isDefinedFunctionIndex(Index)) {
+            wasm::WasmFunction &F = getDefinedFunction(Index);
+            F.DebugName = Name;
+            Signature = &Signatures[F.SigIndex];
+            if (F.ExportName) {
+              Info.ExportName = F.ExportName;
+              Info.Flags |= wasm::WASM_SYMBOL_BINDING_GLOBAL;
+            } else {
+              Info.Flags |= wasm::WASM_SYMBOL_BINDING_LOCAL;
+            }
+          } else {
+            Info.Flags |= wasm::WASM_SYMBOL_UNDEFINED;
+          }
         } else if (Type == wasm::WASM_NAMES_GLOBAL) {
-          nameType = wasm::NameType::GLOBAL;
           if (!SeenGlobals.insert(Index).second)
             return make_error<GenericBinaryError>("global named more than once",
                                                   object_error::parse_failed);
           if (!isValidGlobalIndex(Index) || Name.empty())
             return make_error<GenericBinaryError>("invalid global name entry",
                                                   object_error::parse_failed);
+          nameType = wasm::NameType::GLOBAL;
+          Info.Kind = wasm::WASM_SYMBOL_TYPE_GLOBAL;
+          if (isDefinedGlobalIndex(Index)) {
+            GlobalType = &getDefinedGlobal(Index).Type;
+          } else {
+            Info.Flags |= wasm::WASM_SYMBOL_UNDEFINED;
+          }
----------------
dschuff wrote:

Yeah, the text above about limitations kind of gets at that. You'll just see one `DATA` symbol per segment, and one `GLOBAL` symbol per global but they won't be linked.
We could try to be more sophisticated. I guess at least for dylibs we'll know it's a dylib before we parse the name section, right (because there's a custom section before the known sections)? So we could try to get it right in that case. 

https://github.com/llvm/llvm-project/pull/81063