[clang] Add support for optional func-name argument in #pragma init_seg

[Shriiraz]

This patch implements parsing of the optional [, func-name] argument of MSVC's #pragma init_seg directive. The function identifier is now parsed and stored in Sema, matching MSVC's accepted syntax.

No behavior changes to code generation yet.

Fixes the FIXME comment in ParsePragma.cpp.
#148938

[github-actions]

Thank you for submitting a Pull Request (PR) to the LLVM Project!

This PR will be automatically labeled and the relevant teams will be notified.

If you wish to, you can add reviewers by using the "Reviewers" section on this page.

If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using @ followed by their GitHub username.

If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers.

If you have further questions, they may be answered by the LLVM GitHub User Guide.

You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums.

[llvmbot]

@llvm/pr-subscribers-clang

Author: None (Shriiraz)

Changes

This patch implements parsing of the optional [, func-name] argument of MSVC's #pragma init_seg directive. The function identifier is now parsed and stored in Sema, matching MSVC's accepted syntax.

No behavior changes to code generation yet.

Fixes the FIXME comment in ParsePragma.cpp.

---

Full diff: https://github.com/llvm/llvm-project/pull/167133.diff

5 Files Affected:

• (modified) clang/include/clang/Sema/Sema.h (+9-1)
• (modified) clang/lib/Parse/ParsePragma.cpp (+32-2)
• (modified) clang/lib/Sema/Sema.cpp (+1-1)
• (modified) clang/lib/Sema/SemaAttr.cpp (+10-5)
• (modified) clang/test/SemaCXX/pragma-init_seg.cpp (+3)

diff --git a/clang/include/clang/Sema/Sema.h b/clang/include/clang/Sema/Sema.h
index 0470645a9e7ad..ef9e993ed4ab3 100644
--- a/clang/include/clang/Sema/Sema.h
+++ b/clang/include/clang/Sema/Sema.h
@@ -2075,6 +2075,10 @@ class Sema final : public SemaBase {
   /// Last section used with #pragma init_seg.
   StringLiteral *CurInitSeg;
   SourceLocation CurInitSegLoc;
+  /// Optional function name provided to #pragma init_seg([seg][, func-name]).
+  /// When present, stores the IdentifierInfo and its location.
+  IdentifierInfo *CurInitSegFn;
+  SourceLocation CurInitSegFnLoc;
 
   /// Sections used with #pragma alloc_text.
   llvm::StringMap<std::tuple<StringRef, SourceLocation>> FunctionToSectionMap;
@@ -2245,8 +2249,12 @@ class Sema final : public SemaBase {
                             StringLiteral *SegmentName);
 
   /// Called on well-formed \#pragma init_seg().
+  /// If a function name is provided (", func-name"), it will be passed
+  /// via Func and FuncLoc.
   void ActOnPragmaMSInitSeg(SourceLocation PragmaLocation,
-                            StringLiteral *SegmentName);
+                            StringLiteral *SegmentName,
+                            IdentifierInfo *Func = nullptr,
+                            SourceLocation FuncLoc = SourceLocation());
 
   /// Called on well-formed \#pragma alloc_text().
   void ActOnPragmaMSAllocText(
diff --git a/clang/lib/Parse/ParsePragma.cpp b/clang/lib/Parse/ParsePragma.cpp
index 7c2b9280f0b76..8f7f256c5ce0a 100644
--- a/clang/lib/Parse/ParsePragma.cpp
+++ b/clang/lib/Parse/ParsePragma.cpp
@@ -1229,6 +1229,11 @@ bool Parser::HandlePragmaMSInitSeg(StringRef PragmaName,
 
   // Parse either the known section names or the string section name.
   StringLiteral *SegmentName = nullptr;
+  // Optional function identifier provided as the second argument to
+  // #pragma init_seg([segment][, func-name]). Declared here so it's in scope
+  // for the call to ActOnPragmaMSInitSeg below.
+  IdentifierInfo *FuncII = nullptr;
+  SourceLocation FuncLoc;
   if (Tok.isAnyIdentifier()) {
     auto *II = Tok.getIdentifierInfo();
     StringRef Section = llvm::StringSwitch<StringRef>(II->getName())
@@ -1259,7 +1264,31 @@ bool Parser::HandlePragmaMSInitSeg(StringRef PragmaName,
           << PragmaName;
       return false;
     }
-    // FIXME: Add support for the '[, func-name]' part of the pragma.
+    // Nothing else here; the optional ', func-name' (if present) will be
+    // parsed after the segment parsing so it works whether the segment was
+    // specified as a known identifier or as a string literal.
+  }
+
+  // After parsing the segment name (either via identifier mapping or a
+  // string literal), optionally parse a comma and an identifier naming the
+  // helper function to be used in place of atexit.
+  // Only consume the comma if an identifier follows; otherwise leave it
+  // for ExpectAndConsume(tok::r_paren) to report the error naturally (e.g.,
+  // for malformed pragmas like #pragma init_seg("a", "b")).
+  if (Tok.is(tok::comma)) {
+    // Save current token state for potential backtrack.
+    Token SavedTok = Tok;
+    PP.Lex(Tok); // tentatively consume comma
+    if (Tok.is(tok::identifier)) {
+      // The comma is followed by an identifier; keep both consumed.
+      FuncII = Tok.getIdentifierInfo();
+      FuncLoc = Tok.getLocation();
+      PP.Lex(Tok); // consume identifier and move to next token
+    } else {
+      // Not an identifier after comma; restore and don't consume the comma.
+      // The r_paren check below will produce the expected diagnostics.
+      Tok = SavedTok;
+    }
   }
 
   if (!SegmentName) {
@@ -1273,7 +1302,8 @@ bool Parser::HandlePragmaMSInitSeg(StringRef PragmaName,
                        PragmaName))
     return false;
 
-  Actions.ActOnPragmaMSInitSeg(PragmaLocation, SegmentName);
+  Actions.ActOnPragmaMSInitSeg(PragmaLocation, SegmentName, FuncII,
+                               FuncLoc);
   return true;
 }
 
diff --git a/clang/lib/Sema/Sema.cpp b/clang/lib/Sema/Sema.cpp
index 46addea232b03..1620c2890d911 100644
--- a/clang/lib/Sema/Sema.cpp
+++ b/clang/lib/Sema/Sema.cpp
@@ -313,7 +313,7 @@ Sema::Sema(Preprocessor &pp, ASTContext &ctxt, ASTConsumer &consumer,
       DataSegStack(nullptr), BSSSegStack(nullptr), ConstSegStack(nullptr),
       CodeSegStack(nullptr), StrictGuardStackCheckStack(false),
       FpPragmaStack(FPOptionsOverride()), CurInitSeg(nullptr),
-      VisContext(nullptr), PragmaAttributeCurrentTargetDecl(nullptr),
+  CurInitSegFn(nullptr), CurInitSegFnLoc(), VisContext(nullptr), PragmaAttributeCurrentTargetDecl(nullptr),
       StdCoroutineTraitsCache(nullptr), IdResolver(pp),
       OriginalLexicalContext(nullptr), StdInitializerList(nullptr),
       StdTypeIdentity(nullptr),
diff --git a/clang/lib/Sema/SemaAttr.cpp b/clang/lib/Sema/SemaAttr.cpp
index 8411a3da8322d..eec8326dabad7 100644
--- a/clang/lib/Sema/SemaAttr.cpp
+++ b/clang/lib/Sema/SemaAttr.cpp
@@ -888,12 +888,17 @@ void Sema::ActOnPragmaMSSection(SourceLocation PragmaLocation,
 }
 
 void Sema::ActOnPragmaMSInitSeg(SourceLocation PragmaLocation,
-                                StringLiteral *SegmentName) {
-  // There's no stack to maintain, so we just have a current section.  When we
-  // see the default section, reset our current section back to null so we stop
-  // tacking on unnecessary attributes.
-  CurInitSeg = SegmentName->getString() == ".CRT$XCU" ? nullptr : SegmentName;
+                                StringLiteral *SegmentName,
+                                IdentifierInfo *Func,
+                                SourceLocation FuncLoc) {
+
+  // When we see the default section, reset back to null.
+  CurInitSeg = (SegmentName->getString() == ".CRT$XCU") ? nullptr : SegmentName;
   CurInitSegLoc = PragmaLocation;
+
+  // Store optional function identifier (may be null).
+  CurInitSegFn = Func;
+  CurInitSegFnLoc = Func ? FuncLoc : SourceLocation();
 }
 
 void Sema::ActOnPragmaMSAllocText(
diff --git a/clang/test/SemaCXX/pragma-init_seg.cpp b/clang/test/SemaCXX/pragma-init_seg.cpp
index 1b22939f18e33..7e5d58e608ce1 100644
--- a/clang/test/SemaCXX/pragma-init_seg.cpp
+++ b/clang/test/SemaCXX/pragma-init_seg.cpp
@@ -17,6 +17,9 @@
 #pragma init_seg("a" L"b") // expected-warning {{expected non-wide string literal in '#pragma init_seg'}}
 
 #pragma init_seg(compiler)
+// Test optional func-name parsing: declare an identifier and use it.
+int myexit();
+#pragma init_seg(compiler, myexit)
 #else
 #pragma init_seg(compiler) // expected-warning {{'#pragma init_seg' is only supported when targeting a Microsoft environment}}
 #endif