[LLD] [COFF] Fix exporting of functions starting with underscores, etc

[mstorsjo]

Summary

This fixes exporting functions in the following cases:

• functions starting with an underscore in def files
• functions starting with an underscore, via dllexport attributes, for mingw
• fastcall and vectorcall functions when declared undecorated in def files
• vectorcall functions when declared decorated in def files
• stdcall functions when declared decorated in def files for mingw

This still exports the stdcall functions with the wrong name in the normal msvc/link.exe mode, if declared with decoration in the def file though (this is not a regression though). Exporting functions via def files including decoration is not something I believe is routinely done though, but is tested to try to match link.exe's behaviour as far as easily possible.

[ruiu]

COFF/Driver.cpp

216 ↗

(On Diff #119798)

Why is MinGW special?

[ruiu]

LGTM as my question was addressed in a different change.

lld/trunk/COFF/Driver.cpp

if (MBOrErr.second)
       fatal(MBOrErr.second,
             "could not get the buffer for the member defining " + SymName);
     Driver->addArchiveBuffer(takeBuffer(std::move(MBOrErr.first)), SymName,
                              ParentName);
   });
 }
 
 static bool isDecorated(StringRef Sym) {
-  return Sym.startswith("_") || Sym.startswith("@") || Sym.startswith("?");
+  return Sym.startswith("@") || Sym.contains("@@") || Sym.startswith("?") ||
+         (!Config->MinGW && Sym.contains('@'));
 }
 
 // Parses .drectve section contents and returns a list of files
 // specified by /defaultlib.
 void LinkerDriver::parseDirectives(StringRef S) {
   ArgParser Parser;
   // .drectve is always tokenized using Windows shell rules.
   opt::InputArgList Args = Parser.parse(S);
auto E = writeImportLibrary(getImportName(AsLib), getImplibPath(), Exports,
                               Config->Machine, false);
   handleAllErrors(std::move(E),
                   [&](ErrorInfoBase &EIB) { error(EIB.message()); });
 }
 
 static void parseModuleDefs(StringRef Path) {
   std::unique_ptr<MemoryBuffer> MB = check(
     MemoryBuffer::getFile(Path, -1, false, true), "could not open " + Path);
-  COFFModuleDefinition M =
-      check(parseCOFFModuleDefinition(MB->getMemBufferRef(), Config->Machine));
+  COFFModuleDefinition M = check(parseCOFFModuleDefinition(
+      MB->getMemBufferRef(), Config->Machine, Config->MinGW));
 
   if (Config->OutputFile.empty())
     Config->OutputFile = Saver.save(M.OutputFile);
   Config->ImportName = Saver.save(M.ImportName);
   if (M.ImageBase)
     Config->ImageBase = M.ImageBase;
   if (M.StackReserve)
     Config->StackReserve = M.StackReserve;

lld/trunk/COFF/SymbolTable.cpp

 StringRef SymbolTable::findMangle(StringRef Name) {
   if (Symbol *Sym = find(Name))
     if (!isa<Undefined>(Sym->body()))
       return Name;
   if (Config->Machine != I386)
     return findByPrefix(("?" + Name + "@@Y").str());
   if (!Name.startswith("_"))
     return "";
-  // Search for x86 C function.
+  // Search for x86 stdcall function.
   StringRef S = findByPrefix((Name + "@").str());
   if (!S.empty())
     return S;
+  // Search for x86 fastcall function.
+  S = findByPrefix(("@" + Name.substr(1) + "@").str());
+  if (!S.empty())
+    return S;
+  // Search for x86 vectorcall function.
+  S = findByPrefix((Name.substr(1) + "@@").str());
+  if (!S.empty())
+    return S;
   // Search for x86 C++ non-member function.
   return findByPrefix(("?" + Name.substr(1) + "@@Y").str());
 }
 
 void SymbolTable::mangleMaybe(SymbolBody *B) {
   auto *U = dyn_cast<Undefined>(B);
   if (!U || U->WeakAlias)
     return;
   StringRef Alias = findMangle(U->getName());
-  if (!Alias.empty())
+  if (!Alias.empty()) {
+    log(U->getName() + " aliased to " + Alias);
     U->WeakAlias = addUndefined(Alias);
   }
+}
 
 SymbolBody *SymbolTable::addUndefined(StringRef Name) {
   return addUndefined(Name, nullptr, false)->body();
 }
 
 std::vector<StringRef> SymbolTable::compileBitcodeFiles() {
   LTO.reset(new BitcodeCompiler);
   for (BitcodeFile *F : BitcodeFile::Instances)

lld/trunk/test/COFF/def-export-stdcall.s

 # REQUIRES: x86
 # RUN: llvm-mc -filetype=obj -triple=i686-windows-msvc %s -o %t.obj
-# RUN: echo -e "LIBRARY foo\nEXPORTS\n  stdcall" > %t.def
+# RUN: echo -e "LIBRARY foo\nEXPORTS\n  stdcall\n  fastcall\n  vectorcall\n  _underscored" > %t.def
 # RUN: lld-link -entry:dllmain -dll -def:%t.def %t.obj -out:%t.dll -implib:%t.lib
-# RUN: llvm-readobj %t.lib | FileCheck %s
-# CHECK: Name type: undecorate
-# CHECK: __imp__stdcall@8
-# CHECK: _stdcall@8
+# RUN: llvm-readobj %t.lib | FileCheck -check-prefix UNDECORATED-IMPLIB %s
+# RUN: llvm-readobj -coff-exports %t.dll | FileCheck -check-prefix UNDECORATED-EXPORTS %s
+
+# UNDECORATED-IMPLIB: Name type: noprefix
+# UNDECORATED-IMPLIB-NEXT: __imp___underscored
+# UNDECORATED-IMPLIB-NEXT: __underscored
+# UNDECORATED-IMPLIB: Name type: undecorate
+# UNDECORATED-IMPLIB-NEXT: __imp_@fastcall@8
+# UNDECORATED-IMPLIB-NEXT: fastcall@8
+# UNDECORATED-IMPLIB: Name type: undecorate
+# UNDECORATED-IMPLIB-NEXT: __imp__stdcall@8
+# UNDECORATED-IMPLIB-NEXT: _stdcall@8
+# UNDECORATED-IMPLIB: Name type: undecorate
+# UNDECORATED-IMPLIB-NEXT: __imp_vectorcall@@8
+# UNDECORATED-IMPLIB-NEXT: vectorcall@@8
+
+# UNDECORATED-EXPORTS: Name: _underscored
+# UNDECORATED-EXPORTS: Name: fastcall
+# UNDECORATED-EXPORTS: Name: stdcall
+# UNDECORATED-EXPORTS: Name: vectorcall
+
+
+# RUN: echo -e "LIBRARY foo\nEXPORTS\n  _stdcall@8\n  @fastcall@8\n  vectorcall@@8" > %t.def
+# RUN: lld-link -entry:dllmain -dll -def:%t.def %t.obj -out:%t.dll -implib:%t.lib
+# RUN: llvm-readobj %t.lib | FileCheck -check-prefix DECORATED-IMPLIB %s
+# RUN: llvm-readobj -coff-exports %t.dll | FileCheck -check-prefix DECORATED-EXPORTS %s
+
+# DECORATED-IMPLIB: Name type: name
+# DECORATED-IMPLIB-NEXT: __imp_@fastcall@8
+# DECORATED-IMPLIB-NEXT: @fastcall@8
+# TODO: To match link.exe, this one should also be Name type: name.
+# DECORATED-IMPLIB: Name type: noprefix
+# DECORATED-IMPLIB-NEXT: __imp__stdcall@8
+# DECORATED-IMPLIB-NEXT: _stdcall@8
+# DECORATED-IMPLIB: Name type: name
+# DECORATED-IMPLIB-NEXT: __imp_vectorcall@@8
+# DECORATED-IMPLIB-NEXT: vectorcall@@8
+
+# DECORATED-EXPORTS: Name: @fastcall@8
+# TODO: To match link.exe, this one should actually be _stdcall@8
+# DECORATED-EXPORTS: Name: stdcall@8
+# DECORATED-EXPORTS: Name: vectorcall@@8
+
+
+# RUN: echo -e "LIBRARY foo\nEXPORTS\n  stdcall@8\n  @fastcall@8" > %t.def
+# RUN: lld-link -lldmingw -entry:dllmain -dll -def:%t.def %t.obj -out:%t.dll -implib:%t.lib
+# RUN: llvm-readobj %t.lib | FileCheck -check-prefix DECORATED-MINGW-IMPLIB %s
+# RUN: llvm-readobj -coff-exports %t.dll | FileCheck -check-prefix DECORATED-MINGW-EXPORTS %s
+
+# DECORATED-MINGW-IMPLIB: Name type: name
+# DECORATED-MINGW-IMPLIB-NEXT: __imp_@fastcall@8
+# DECORATED-MINGW-IMPLIB-NEXT: fastcall@8
+# DECORATED-MINGW-IMPLIB: Name type: noprefix
+# DECORATED-MINGW-IMPLIB-NEXT: __imp__stdcall@8
+# DECORATED-MINGW-IMPLIB-NEXT: _stdcall@8
+
+# DECORATED-MINGW-EXPORTS: Name: @fastcall@8
+# DECORATED-MINGW-EXPORTS: Name: stdcall@8
+
 
         .def     _stdcall@8;
         .scl    2;
         .type   32;
         .endef
         .globl  _stdcall@8
+        .globl  @fastcall@8
+        .globl  vectorcall@@8
+        .globl  __underscored
 _stdcall@8:
         movl    8(%esp), %eax
         addl    4(%esp), %eax
         retl    $8
+@fastcall@8:
+        movl    8(%esp), %eax
+        addl    4(%esp), %eax
+        retl    $8
+vectorcall@@8:
+        movl    8(%esp), %eax
+        addl    4(%esp), %eax
+        retl    $8
+__underscored:
+        ret
 
         .def     _dllmain;
         .scl    2;
         .type   32;
         .endef
         .globl  _dllmain
 _dllmain:
         retl
 

lld/trunk/test/COFF/dllexport-mingw.s

 # REQEUIRES: x86
 
 # RUN: llvm-mc -triple=i686-windows-gnu %s -filetype=obj -o %t.obj
 
 # RUN: lld-link -lldmingw -dll -out:%t.dll -entry:main %t.obj -implib:%t.lib
 # RUN: llvm-readobj %t.lib | FileCheck %s
 
+# CHECK: Symbol: __imp___underscoredFunc
+# CHECK: Symbol: __underscoredFunc
 # CHECK: Symbol: __imp__func
 # CHECK: Symbol: _func
 
 .global _main
 .global _func
+.global __underscoredFunc
 .text
 _main:
   ret
 _func:
   ret
+__underscoredFunc:
+  ret
 .section .drectve
-.ascii "-export:func"
+.ascii "-export:func -export:_underscoredFunc"