[ELF] - Added support for extern "c++" local symbols in version script.

[grimar]

Summary

Previously we did not support them,
patch implements this functionality.

[ruiu]

ELF/LinkerScript.cpp

1799–1808 ↗

(On Diff #77793)

We have very similar code in readGlobal. Can you factor it out?

[grimar]

• Addressed review comments.

ELF/LinkerScript.cpp

1799–1808 ↗

(On Diff #77793)

Done.

[ruiu]

ELF/LinkerScript.cpp

1810 ↗

(On Diff #77969)

I don't think you need Globals. You could just stop reading until local: in any context.

[grimar]

• Rebased, addressed comment.

[grimar]

ELF/LinkerScript.cpp

1810 ↗

(On Diff #77969)

That what seems obvious and was what I first think of.
But also it would parse something like next I think:

...
local:
<locals1>
local:
<local2>
...
It does not look clean.

Actually if you're ok with that, I am fine either,
as I also do not like that flag.

[grimar]

ELF/LinkerScript.cpp

1810 ↗

(On Diff #77969)

Ah, I see my mistake now, above can't happen, please ignore my comment then.

[ruiu]

LGTM

ELF/SymbolTable.cpp

721 ↗

(On Diff #78150)

Is there any reason to add single quotes? Please remove them because V.Name doesn't contain such quotes.

[grimar]

ELF/SymbolTable.cpp

721 ↗

(On Diff #78150)

Lets add them for V.Name ? That is consistent with other messages we added recently

[ruiu]

ELF/SymbolTable.cpp

721 ↗

(On Diff #78150)

Well, what you pass here is a version name, and that's the right thing to do. If you want to add single quotes, do it in the string in error().

lld/trunk/ELF/LinkerScript.cpp

private:
   Expr readTernary(Expr Cond);
   Expr readParenExpr();
 
   // For parsing version script.
   void readVersionExtern(std::vector<SymbolVersion> *Globals);
   void readVersionDeclaration(StringRef VerStr);
   void readGlobal(StringRef VerStr);
   void readLocal(StringRef VerStr);
+  void readSymbols(std::vector<SymbolVersion> &V);
 
   ScriptConfiguration &Opt = *ScriptConfig;
   bool IsUnderSysroot;
 };
 
 void ScriptParser::readVersionScript() {
   readVersionScriptCommand();
   if (!atEOF())
void ScriptParser::readVersionDeclaration(StringRef VerStr) {
   // "Ver2 { global: foo; local: *; } Ver1;" has "Ver1" as a parent. This
   // version hierarchy is, probably against your instinct, purely for human; the
   // runtime doesn't care about them at all. In LLD, we simply skip the token.
   if (!VerStr.empty() && peek() != ";")
     skip();
   expect(";");
 }
 
+void ScriptParser::readSymbols(std::vector<SymbolVersion> &V) {
+  for (;;) {
+    if (consume("extern"))
+      readVersionExtern(&V);
+
+    StringRef Cur = peek();
+    if (Cur == "}" || Cur == "local:" || Error)
+      return;
+    skip();
+    V.push_back({unquote(Cur), false, hasWildcard(Cur)});
+    expect(";");
+  }
+}
+
 void ScriptParser::readLocal(StringRef VerStr) {
   if (consume("*")) {
     Config->DefaultSymbolVersion = VER_NDX_LOCAL;
     expect(";");
     return;
   }
 
   if (VerStr.empty())
     setError("locals list for anonymous version is not supported");
 
-  std::vector<SymbolVersion> &Locals = Config->VersionDefinitions.back().Locals;
-  while (!Error && peek() != "}") {
-    StringRef Tok = next();
-    Locals.push_back({unquote(Tok), false, hasWildcard(Tok)});
-    expect(";");
-  }
+  readSymbols(Config->VersionDefinitions.back().Locals);
 }
 
-void ScriptParser::readVersionExtern(std::vector<SymbolVersion> *Globals) {
+void ScriptParser::readVersionExtern(std::vector<SymbolVersion> *V) {
   expect("\"C++\"");
   expect("{");
 
   for (;;) {
     if (peek() == "}" || Error)
       break;
     bool HasWildcard = !peek().startswith("\"") && hasWildcard(peek());
-    Globals->push_back({unquote(next()), true, HasWildcard});
+    V->push_back({unquote(next()), true, HasWildcard});
     expect(";");
   }
 
   expect("}");
   expect(";");
 }
 
 void ScriptParser::readGlobal(StringRef VerStr) {
-  std::vector<SymbolVersion> *Globals;
   if (VerStr.empty())
-    Globals = &Config->VersionScriptGlobals;
+    readSymbols(Config->VersionScriptGlobals);
   else
-    Globals = &Config->VersionDefinitions.back().Globals;
-
-  for (;;) {
-    if (consume("extern"))
-      readVersionExtern(Globals);
-
-    StringRef Cur = peek();
-    if (Cur == "}" || Cur == "local:" || Error)
-      return;
-    skip();
-    Globals->push_back({unquote(Cur), false, hasWildcard(Cur)});
-    expect(";");
-  }
+    readSymbols(Config->VersionDefinitions.back().Globals);
 }
 
 static bool isUnderSysroot(StringRef Path) {
   if (Config->Sysroot == "")
     return false;
   for (; !Path.empty(); Path = sys::path::parent_path(Path))
     if (sys::fs::equivalent(Config->Sysroot, Path))
       return true;

lld/trunk/ELF/SymbolTable.cpp

 }
 
 // A helper function to set a version to a symbol.
 // Essentially, assigning two different versions to the same symbol is an error.
 static void setVersionId(SymbolBody *Body, StringRef VersionName,
                          StringRef Name, uint16_t Version) {
   if (!Body || Body->isUndefined()) {
     if (Config->NoUndefinedVersion)
-      error("version script assignment of " + VersionName + " to symbol " +
+      error("version script assignment of '" + VersionName + "' to symbol " +
             Name + " failed: symbol not defined");
     return;
   }
 
   Symbol *Sym = Body->symbol();
   if (Sym->VersionId != Config->DefaultSymbolVersion)
     warn("duplicate symbol " + Name + " in version script");
   Sym->VersionId = Version;
template <class ELFT> void SymbolTable<ELFT>::scanVersionScript() {
 
   if (Config->VersionDefinitions.empty())
     return;
 
   // Now we have version definitions, so we need to set version ids to symbols.
   // Each version definition has a glob pattern, and all symbols that match
   // with the pattern get that version.
 
-  // First, we assign versions to exact matching symbols,
-  // i.e. version definitions not containing any glob meta-characters.
-  for (VersionDefinition &V : Config->VersionDefinitions) {
-    for (SymbolVersion Ver : V.Globals) {
-      if (Ver.HasWildcards)
-        continue;
+  auto assignVersion = [&](SymbolVersion &Ver, size_t Version,
+                           StringRef VerName) {
+      return;
 
-      StringRef N = Ver.Name;
-      if (Ver.IsExternCpp) {
-        for (SymbolBody *B : findDemangled(N))
-          setVersionId(B, V.Name, N, V.Id);
-        continue;
-      }
-      setVersionId(find(N), V.Name, N, V.Id);
-    }
-    for (SymbolVersion Ver : V.Locals) {
-      if (Ver.HasWildcards)
-        continue;
-      setVersionId(find(Ver.Name), V.Name, Ver.Name, VER_NDX_LOCAL);
+      for (SymbolBody *B : findDemangled(Ver.Name))
+        setVersionId(B, VerName, Ver.Name, Version);
+      return;
     }
+    setVersionId(find(Ver.Name), VerName, Ver.Name, Version);
+  };
+
+  // First, we assign versions to exact matching symbols,
+  // i.e. version definitions not containing any glob meta-characters.
+  for (VersionDefinition &V : Config->VersionDefinitions) {
+    for (SymbolVersion Sym : V.Globals)
+      assignVersion(Sym, V.Id, V.Name);
+    for (SymbolVersion Sym : V.Locals)
+      assignVersion(Sym, VER_NDX_LOCAL, "local");
   }
 
   // Next, we assign versions to fuzzy matching symbols,
   // i.e. version definitions containing glob meta-characters.
   // Note that because the last match takes precedence over previous matches,
   // we iterate over the definitions in the reverse order.
   for (size_t I = Config->VersionDefinitions.size() - 1; I != (size_t)-1; --I) {
     VersionDefinition &V = Config->VersionDefinitions[I];

lld/trunk/test/ELF/version-script-locals-extern.s

+# REQUIRES: x86
+
+# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o
+# RUN: echo "FOO { local: extern \"C++\" { \"abb(int)\"; }; };" > %t.script
+# RUN: ld.lld --version-script %t.script -shared %t.o -o %t.so
+# RUN: llvm-readobj -V %t.so | FileCheck %s --check-prefix=ABB
+# ABB:      Symbols [
+# ABB-NEXT:   Symbol {
+# ABB-NEXT:     Version: 0
+# ABB-NEXT:     Name: @
+# ABB-NEXT:   }
+# ABB-NEXT:   Symbol {
+# ABB-NEXT:     Version: 1
+# ABB-NEXT:     Name: _Z3abci@
+# ABB-NEXT:   }
+# ABB-NEXT: ]
+
+# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o
+# RUN: echo "FOO { local: extern \"C++\" { abb*; }; };" > %t.script
+# RUN: ld.lld --version-script %t.script -shared %t.o -o %t.so
+# RUN: llvm-readobj -V %t.so | FileCheck %s --check-prefix=ABB
+
+# RUN: echo "FOO { local: extern \"C++\" { abc*; }; };" > %t.script
+# RUN: ld.lld --version-script %t.script -shared %t.o -o %t.so
+# RUN: llvm-readobj -V %t.so | FileCheck %s --check-prefix=ABC
+# ABC:      Symbols [
+# ABC-NEXT:   Symbol {
+# ABC-NEXT:     Version: 0
+# ABC-NEXT:     Name: @
+# ABC-NEXT:   }
+# ABC-NEXT:   Symbol {
+# ABC-NEXT:     Version: 1
+# ABC-NEXT:     Name: _Z3abbi@
+# ABC-NEXT:   }
+# ABC-NEXT: ]
+
+.globl _Z3abbi
+.type _Z3abbi,@function
+_Z3abbi:
+retq
+
+.globl _Z3abci
+.type _Z3abci,@function
+_Z3abci:
+retq

lld/trunk/test/ELF/version-script-locals.s

 # RUN: ld.lld --version-script %t.script -shared %t.o -o %t.so
 # RUN: llvm-readobj -dyn-symbols %t.so | FileCheck --check-prefix=MIX %s
 # MIX:  DynamicSymbols [
 # MIX:      _start@@VERSION_1.0
 # MIX-NOT:  foo1
 # MIX-NOT:  foo2
 # MIX-NOT:  foo3
 
-# RUN: echo "VERSION_1.0 { global: *; local: extern \"C++\" { foo*; } };" > %t.script
-# RUN: not ld.lld --version-script %t.script -shared %t.o -o %t.so 2>&1 \
-# RUN:   | FileCheck --check-prefix=EXTERNERR %s
-# EXTERNERR: ; expected, but got "C++"
-
 .globl foo1
 foo1:
   ret
 
 .globl foo2
 foo2:
   ret
 
 .globl foo3
 foo3:
   ret
 
 .globl _start
 _start:
   ret

lld/trunk/test/ELF/version-script-noundef.s

 # REQUIRES: x86
 
 # RUN: echo "VERSION_1.0 { global: bar; };" > %t.script
 # RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t.o
 # RUN: not ld.lld --version-script %t.script -shared --no-undefined-version \
 # RUN:   %t.o -o %t.so 2>&1 | FileCheck -check-prefix=ERR1 %s
-# ERR1: version script assignment of VERSION_1.0 to symbol bar failed: symbol not defined
+# ERR1: version script assignment of 'VERSION_1.0' to symbol bar failed: symbol not defined
 
 # RUN: echo "VERSION_1.0 { global: und; };" > %t2.script
 # RUN: not ld.lld --version-script %t2.script -shared --no-undefined-version \
 # RUN:   %t.o -o %t.so 2>&1 | FileCheck -check-prefix=ERR2 %s
-# ERR2: version script assignment of VERSION_1.0 to symbol und failed: symbol not defined
+# ERR2: version script assignment of 'VERSION_1.0' to symbol und failed: symbol not defined
+
+# RUN: echo "VERSION_1.0 { local: und; };" > %t3.script
+# RUN: not ld.lld --version-script %t3.script -shared --no-undefined-version \
+# RUN:   %t.o -o %t.so 2>&1 | FileCheck -check-prefix=ERR3 %s
+# ERR3: version script assignment of 'local' to symbol und failed: symbol not defined
 
 .text
 .globl foo
 .type foo,@function
 foo:
 callq und@PLT