✨ Speedy #embed implementation

⚡ [Lex] Better reservations for improved performance/memory usage.

🛠 [Lex, Frontend] Remove comma hardcoding since we are servicing a full file

apply suggestions from git-clang-format

clang/include/clang/AST/Expr.h

@@ -4805,6 +4805,57 @@ class SourceLocExpr final : public Expr {
   friend class ASTStmtReader;
 };
 
+/// Represents a function call to __builtin_pp_embed().
+class PPEmbedExpr final : public Expr {
+  SourceLocation BuiltinLoc, RParenLoc;
+  DeclContext *ParentContext;
+  StringLiteral *Filename;
+  StringLiteral *BinaryData;
+
+public:
+  enum Action {
+    NotFound,
+    FoundOne,
+    Expanded,
+  };
+
+  PPEmbedExpr(const ASTContext &Ctx, QualType ResultTy, StringLiteral *Filename,
+              StringLiteral *BinaryData, SourceLocation BLoc,
+              SourceLocation RParenLoc, DeclContext *Context);
+
+  /// Build an empty call expression.
+  explicit PPEmbedExpr(EmptyShell Empty) : Expr(SourceLocExprClass, Empty) {}
+
+  /// If the PPEmbedExpr has been resolved return the subexpression
+  /// representing the resolved value. Otherwise return null.
+  const DeclContext *getParentContext() const { return ParentContext; }
+  DeclContext *getParentContext() { return ParentContext; }
+
+  SourceLocation getLocation() const { return BuiltinLoc; }
+  SourceLocation getBeginLoc() const { return BuiltinLoc; }
+  SourceLocation getEndLoc() const { return RParenLoc; }
+
+  StringLiteral *getFilenameStringLiteral() const { return Filename; }
+  StringLiteral *getDataStringLiteral() const { return BinaryData; }
+
+  size_t getDataElementCount(ASTContext &Context) const;
+
+  child_range children() {
+    return child_range(child_iterator(), child_iterator());
+  }
+
+  const_child_range children() const {
+    return const_child_range(child_iterator(), child_iterator());
+  }
+
+  static bool classof(const Stmt *T) {
+    return T->getStmtClass() == PPEmbedExprClass;
+  }
+
+private:
+  friend class ASTStmtReader;
+};
+
 /// Describes an C or C++ initializer list.
 ///
 /// InitListExpr describes an initializer list, which can be used to

clang/include/clang/AST/RecursiveASTVisitor.h

@@ -2809,6 +2809,7 @@ DEF_TRAVERSE_STMT(ShuffleVectorExpr, {})
 DEF_TRAVERSE_STMT(ConvertVectorExpr, {})
 DEF_TRAVERSE_STMT(StmtExpr, {})
 DEF_TRAVERSE_STMT(SourceLocExpr, {})
+DEF_TRAVERSE_STMT(PPEmbedExpr, {})
 
 DEF_TRAVERSE_STMT(UnresolvedLookupExpr, {
   TRY_TO(TraverseNestedNameSpecifierLoc(S->getQualifierLoc()));

clang/include/clang/Basic/DiagnosticCommonKinds.td

@@ -59,6 +59,9 @@ def err_expected_string_literal : Error<"expected string literal "
           "'external_source_symbol' attribute|"
           "as argument of '%1' attribute}0">;
 
+def err_builtin_pp_embed_invalid_argument : Error<
+  "invalid argument to '__builtin_pp_embed': %0">;
+
 def err_invalid_string_udl : Error<
   "string literal with user-defined suffix cannot be used here">;
 def err_invalid_character_udl : Error<
@@ -80,6 +83,9 @@ def err_expected : Error<"expected %0">;
 def err_expected_either : Error<"expected %0 or %1">;
 def err_expected_after : Error<"expected %1 after %0">;
 
+def err_builtin_pp_embed_invalid_location : Error<
+  "'__builtin_pp_embed' in invalid location: %0%select{|%2}1">;
+
 def err_param_redefinition : Error<"redefinition of parameter %0">;
 def warn_method_param_redefinition : Warning<"redefinition of method parameter %0">;
 def warn_method_param_declaration : Warning<"redeclaration of method parameter %0">,

clang/include/clang/Basic/FileManager.h

@@ -282,8 +282,9 @@ class FileManager : public RefCountedBase<FileManager> {
   getBufferForFile(StringRef Filename, bool isVolatile = false,
                    bool RequiresNullTerminator = true,
                    std::optional<int64_t> MaybeLimit = std::nullopt) {
-    return getBufferForFileImpl(Filename, /*FileSize=*/(MaybeLimit ? *MaybeLimit : -1), isVolatile,
-                                RequiresNullTerminator);
+    return getBufferForFileImpl(Filename,
+                                /*FileSize=*/(MaybeLimit ? *MaybeLimit : -1),
+                                isVolatile, RequiresNullTerminator);
   }
 
 private:

clang/include/clang/Basic/StmtNodes.td

@@ -203,6 +203,7 @@ def OpaqueValueExpr : StmtNode<Expr>;
 def TypoExpr : StmtNode<Expr>;
 def RecoveryExpr : StmtNode<Expr>;
 def BuiltinBitCastExpr : StmtNode<ExplicitCastExpr>;
+def PPEmbedExpr : StmtNode<Expr>;
 
 // Microsoft Extensions.
 def MSPropertyRefExpr : StmtNode<Expr>;

clang/include/clang/Basic/TokenKinds.def

@@ -154,10 +154,6 @@ TOK(eod)                 // End of preprocessing directive (end of line inside a
                          // directive).
 TOK(code_completion)     // Code completion marker
 
-// #embed speed support
-TOK(builtin_embed)
-
-
 // C99 6.4.9: Comments.
 TOK(comment)             // Comment (only in -E -C[C] mode)
 
@@ -758,6 +754,7 @@ ALIAS("__char32_t"   , char32_t          , KEYCXX)
 KEYWORD(__builtin_bit_cast               , KEYALL)
 KEYWORD(__builtin_available              , KEYALL)
 KEYWORD(__builtin_sycl_unique_stable_name, KEYSYCL)
+KEYWORD(__builtin_pp_embed               , KEYALL)
 
 // Keywords defined by Attr.td.
 #ifndef KEYWORD_ATTRIBUTE
@@ -993,6 +990,7 @@ ANNOTATION(repl_input_end)
 #undef CXX11_KEYWORD
 #undef KEYWORD
 #undef PUNCTUATOR
+#undef BUILTINOK
 #undef TOK
 #undef C99_KEYWORD
 #undef C23_KEYWORD

clang/include/clang/Frontend/PreprocessorOutputOptions.h

@@ -22,7 +22,8 @@ class PreprocessorOutputOptions {
   unsigned ShowMacroComments : 1;  ///< Show comments, even in macros.
   unsigned ShowMacros : 1;         ///< Print macro definitions.
   unsigned ShowIncludeDirectives : 1;  ///< Print includes, imports etc. within preprocessed output.
-  unsigned ShowEmbedDirectives : 1;  ///< Print embeds, etc. within preprocessed output.
+  unsigned ShowEmbedDirectives : 1; ///< Print embeds, etc. within preprocessed
+                                    ///< output.
   unsigned RewriteIncludes : 1;    ///< Preprocess include directives only.
   unsigned RewriteImports  : 1;    ///< Include contents of transitively-imported modules.
   unsigned MinimizeWhitespace : 1; ///< Ignore whitespace from input.

clang/include/clang/Lex/PPDirectiveParameter.h

@@ -0,0 +1,32 @@
+//===--- MacroArgs.h - Formal argument info for Macros ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the MacroArgs interface.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_LEX_PPDIRECTIVEPARAMETER_H
+#define LLVM_CLANG_LEX_PPDIRECTIVEPARAMETER_H
+
+#include "clang/Basic/SourceLocation.h"
+
+namespace clang {
+
+/// Captures basic information about a preprocessor directive parameter.
+class PPDirectiveParameter {
+public:
+  SourceLocation Start;
+  SourceLocation End;
+
+  PPDirectiveParameter(SourceLocation Start, SourceLocation End)
+      : Start(Start), End(End) {}
+};
+
+} // end namespace clang
+
+#endif

clang/include/clang/Lex/PPEmbedParameters.h

@@ -0,0 +1,78 @@
+//===--- MacroArgs.h - Formal argument info for Macros ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the MacroArgs interface.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_LEX_PPEMBEDPARAMETERS_H
+#define LLVM_CLANG_LEX_PPEMBEDPARAMETERS_H
+
+#include "clang/Lex/PPDirectiveParameter.h"
+#include "clang/Lex/Token.h"
+#include "llvm/ADT/SmallVector.h"
+
+namespace clang {
+
+/// Preprocessor extension embed parameter "clang::offset"
+/// `clang::offset( constant-expression )`
+class PPEmbedParameterOffset : public PPDirectiveParameter {
+public:
+  size_t Offset;
+
+  PPEmbedParameterOffset(size_t Offset, SourceLocation Start,
+                         SourceLocation End)
+      : Offset(Offset), PPDirectiveParameter(Start, End) {}
+};
+
+/// Preprocessor standard embed parameter "limit"
+/// `limit( constant-expression )`
+class PPEmbedParameterLimit : public PPDirectiveParameter {
+public:
+  size_t Limit;
+
+  PPEmbedParameterLimit(size_t Limit, SourceLocation Start, SourceLocation End)
+      : Limit(Limit), PPDirectiveParameter(Start, End) {}
+};
+
+/// Preprocessor standard embed parameter "prefix"
+/// `prefix( balanced-token-seq )`
+class PPEmbedParameterPrefix : public PPDirectiveParameter {
+public:
+  SmallVector<Token, 2> Tokens;
+
+  PPEmbedParameterPrefix(SmallVector<Token, 2> Tokens, SourceLocation Start,
+                         SourceLocation End)
+      : Tokens(std::move(Tokens)), PPDirectiveParameter(Start, End) {}
+};
+
+/// Preprocessor standard embed parameter "suffix"
+/// `suffix( balanced-token-seq )`
+class PPEmbedParameterSuffix : public PPDirectiveParameter {
+public:
+  SmallVector<Token, 2> Tokens;
+
+  PPEmbedParameterSuffix(SmallVector<Token, 2> Tokens, SourceLocation Start,
+                         SourceLocation End)
+      : Tokens(std::move(Tokens)), PPDirectiveParameter(Start, End) {}
+};
+
+/// Preprocessor standard embed parameter "if_empty"
+/// `if_empty( balanced-token-seq )`
+class PPEmbedParameterIfEmpty : public PPDirectiveParameter {
+public:
+  SmallVector<Token, 2> Tokens;
+
+  PPEmbedParameterIfEmpty(SmallVector<Token, 2> Tokens, SourceLocation Start,
+                          SourceLocation End)
+      : Tokens(std::move(Tokens)), PPDirectiveParameter(Start, End) {}
+};
+
+} // end namespace clang
+
+#endif

clang/include/clang/Lex/Preprocessor.h

@@ -29,6 +29,7 @@
 #include "clang/Lex/ModuleLoader.h"
 #include "clang/Lex/ModuleMap.h"
 #include "clang/Lex/PPCallbacks.h"
+#include "clang/Lex/PPEmbedParameters.h"
 #include "clang/Lex/Token.h"
 #include "clang/Lex/TokenLexer.h"
 #include "llvm/ADT/APSInt.h"
@@ -1165,6 +1166,9 @@ class Preprocessor {
 
   void updateOutOfDateIdentifier(IdentifierInfo &II) const;
 
+  /// Buffers for used #embed directives
+  std::vector<std::string> EmbedBuffers;
+
 public:
   Preprocessor(std::shared_ptr<PreprocessorOptions> PPOpts,
                DiagnosticsEngine &diags, const LangOptions &LangOpts,
@@ -1735,15 +1739,15 @@ class Preprocessor {
   bool LexHeaderName(Token &Result, bool AllowMacroExpansion = true);
 
   struct LexEmbedParametersResult {
-    bool Successful;
-    std::optional<size_t> MaybeLimitParam;
-    std::optional<size_t> MaybeOffsetParam;
-    std::optional<SmallVector<Token, 2>> MaybeIfEmptyParam;
-    std::optional<SmallVector<Token, 2>> MaybePrefixParam;
-    std::optional<SmallVector<Token, 2>> MaybeSuffixParam;
-    int UnrecognizedParams;
+    std::optional<PPEmbedParameterLimit> MaybeLimitParam;
+    std::optional<PPEmbedParameterOffset> MaybeOffsetParam;
+    std::optional<PPEmbedParameterIfEmpty> MaybeIfEmptyParam;
+    std::optional<PPEmbedParameterPrefix> MaybePrefixParam;
+    std::optional<PPEmbedParameterSuffix> MaybeSuffixParam;
     SourceLocation StartLoc;
     SourceLocation EndLoc;
+    int UnrecognizedParams;
+    bool Successful;
   };
 
   LexEmbedParametersResult LexEmbedParameters(Token &Current,
@@ -1812,7 +1816,8 @@ class Preprocessor {
   /// Parses a simple integer literal to get its numeric value.  Floating
   /// point literals and user defined literals are rejected.  Used primarily to
   /// handle pragmas that accept integer arguments.
-  bool parseSimpleIntegerLiteral(Token &Tok, uint64_t &Value);
+  bool parseSimpleIntegerLiteral(Token &Tok, uint64_t &Value,
+                                 bool WithLex = true);
 
   /// Disables macro expansion everywhere except for preprocessor directives.
   void SetMacroExpansionOnlyInDirectives() {
@@ -2735,12 +2740,18 @@ class Preprocessor {
   // Binary data inclusion
   void HandleEmbedDirective(SourceLocation HashLoc, Token &Tok,
                             const FileEntry *LookupFromFile = nullptr);
-  void HandleEmbedDirectiveNaive(
-      SourceLocation FilenameTok, LexEmbedParametersResult &Params,
-      StringRef BinaryContents, const size_t TargetCharWidth);
-  void HandleEmbedDirectiveBuiltin(
-      SourceLocation FilenameTok, LexEmbedParametersResult &Params,
-      StringRef BinaryContents, const size_t TargetCharWidth);
+  void HandleEmbedDirectiveNaive(SourceLocation HashLoc,
+                                 SourceLocation FilenameTok,
+                                 const LexEmbedParametersResult &Params,
+                                 StringRef BinaryContents,
+                                 const size_t TargetCharWidth);
+  void HandleEmbedDirectiveBuiltin(SourceLocation HashLoc,
+                                   const Token &FilenameTok,
+                                   StringRef ResolvedFilename,
+                                   StringRef SearchPath, StringRef RelativePath,
+                                   const LexEmbedParametersResult &Params,
+                                   StringRef BinaryContents,
+                                   const size_t TargetCharWidth);
 
   // File inclusion.
   void HandleIncludeDirective(SourceLocation HashLoc, Token &Tok,

clang/include/clang/Sema/Sema.h

@@ -5983,6 +5983,10 @@ class Sema final {
                         ArrayRef<Expr *> Arg, SourceLocation RParenLoc,
                         Expr *Config = nullptr, bool IsExecConfig = false,
                         ADLCallKind UsesADL = ADLCallKind::NotADL);
+  /// `Fn` may be a null pointer.
+  void ModifyCallExprArguments(Expr *Fn, SourceLocation LParenLoc,
+                               SmallVectorImpl<Expr *> &ArgExprs,
+                               SourceLocation RParenLoc);
 
   ExprResult ActOnCUDAExecConfigExpr(Scope *S, SourceLocation LLLLoc,
                                      MultiExprArg ExecConfig,
@@ -6100,6 +6104,35 @@ class Sema final {
                                 SourceLocation BuiltinLoc,
                                 SourceLocation RPLoc);
 
+  // __builtin_pp_embed()
+  ExprResult ActOnPPEmbedExpr(SourceLocation BuiltinLoc,
+                              SourceLocation Base64DataLocation,
+                              SourceLocation RPLoc, StringLiteral *Filename,
+                              QualType DataTy, std::vector<char> BinaryData);
+
+  IntegerLiteral *ExpandSinglePPEmbedExpr(PPEmbedExpr *PPEmbed);
+
+  PPEmbedExpr::Action
+  CheckExprListForPPEmbedExpr(ArrayRef<Expr *> ExprList,
+                              std::optional<QualType> MaybeInitType);
+  PPEmbedExpr::Action
+  ExpandPPEmbedExprInExprList(ArrayRef<Expr *> ExprList,
+                              SmallVectorImpl<Expr *> &OutputExprList,
+                              bool ClearOutputFirst = true);
+  PPEmbedExpr::Action
+  ExpandPPEmbedExprInExprList(SmallVectorImpl<Expr *> &OutputList);
+
+  enum PPEmbedExprContext {
+    PPEEC__StaticAssert,
+    PPEEC_StaticAssert,
+  };
+
+  StringRef GetLocationName(PPEmbedExprContext Context) const;
+
+  bool DiagnosePPEmbedExpr(Expr *&E, SourceLocation ContextLocation,
+                           PPEmbedExprContext Context,
+                           bool SingleAllowed = true);
+
   // Build a potentially resolved SourceLocExpr.
   ExprResult BuildSourceLocExpr(SourceLocExpr::IdentKind Kind,
                                 QualType ResultTy, SourceLocation BuiltinLoc,
@@ -8292,6 +8325,10 @@ class Sema final {
                                        SourceLocation EqualLoc,
                                        ParsedTemplateArgument DefaultArg);
 
+  void ModifyTemplateArguments(
+      const TemplateTy &Template,
+      SmallVectorImpl<ParsedTemplateArgument> &TemplateArgs);
+
   TemplateParameterList *
   ActOnTemplateParameterList(unsigned Depth,
                              SourceLocation ExportLoc,

clang/include/clang/Serialization/ASTBitCodes.h

@@ -1715,6 +1715,9 @@ enum StmtCode {
   /// A SourceLocExpr record.
   EXPR_SOURCE_LOC,
 
+  /// A PPEmbedExpr record.
+  EXPR_BUILTIN_PP_EMBED,
+
   /// A ShuffleVectorExpr record.
   EXPR_SHUFFLE_VECTOR,