Make -fno-char8_t disable the char8_t keyword, even in C++20.

This fixes a regression introduced in r354736, and makes our behavior
compatible with that of Clang 8 and GCC.

diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 9c22117..a35c966 100644 (file)
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -127,6 +127,9 @@ Modified Compiler Flags
   ``-f[no-]sanitize-recover=undefined,integer`` and is no longer deprecated.
 - The argument to ``-f[no-]sanitize-trap=...`` is now optional and defaults to
   ``all``.
+- ``-fno-char8_t`` now disables the ``char8_t`` keyword, not just the use of
+  ``char8_t`` as the character type of ``u8`` literals. This restores the
+  Clang 8 behavior that regressed in Clang 9 and 10.
 
 New Pragmas in Clang
 --------------------

diff --git a/clang/include/clang/Basic/TokenKinds.def b/clang/include/clang/Basic/TokenKinds.def
index 462e160..0705896 100644 (file)
--- a/clang/include/clang/Basic/TokenKinds.def
+++ b/clang/include/clang/Basic/TokenKinds.def
@@ -388,9 +388,10 @@ MODULES_KEYWORD(module)
 MODULES_KEYWORD(import)
 
 // C++20 keywords.
-CXX20_KEYWORD(char8_t               , CHAR8SUPPORT)
 CXX20_KEYWORD(consteval             , 0)
 CXX20_KEYWORD(constinit             , 0)
+// Not a CXX20_KEYWORD because it is disabled by -fno-char8_t.
+KEYWORD(char8_t                     , CHAR8SUPPORT)
 
 // C11 Extension
 KEYWORD(_Float16                    , KEYALL)

diff --git a/clang/lib/Basic/IdentifierTable.cpp b/clang/lib/Basic/IdentifierTable.cpp
index e88d939..d7ef159 100644 (file)
--- a/clang/lib/Basic/IdentifierTable.cpp
+++ b/clang/lib/Basic/IdentifierTable.cpp
@@ -146,6 +146,8 @@ static KeywordStatus getKeywordStatus(const LangOptions &LangOpts,
   if (LangOpts.Coroutines && (Flags & KEYCOROUTINES)) return KS_Enabled;
   if (LangOpts.ModulesTS && (Flags & KEYMODULES)) return KS_Enabled;
   if (LangOpts.CPlusPlus && (Flags & KEYALLCXX)) return KS_Future;
+  if (LangOpts.CPlusPlus && !LangOpts.CPlusPlus20 && (Flags & CHAR8SUPPORT))
+    return KS_Future;
   return KS_Disabled;
 }
 

diff --git a/clang/lib/Format/Format.cpp b/clang/lib/Format/Format.cpp
index eead2b4..737e502 100644 (file)
--- a/clang/lib/Format/Format.cpp
+++ b/clang/lib/Format/Format.cpp
@@ -2622,6 +2622,7 @@ LangOptions getFormattingLangOpts(const FormatStyle &Style) {
   LangOpts.CPlusPlus14 = LexingStd >= FormatStyle::LS_Cpp14;
   LangOpts.CPlusPlus17 = LexingStd >= FormatStyle::LS_Cpp17;
   LangOpts.CPlusPlus20 = LexingStd >= FormatStyle::LS_Cpp20;
+  LangOpts.Char8 = LexingStd >= FormatStyle::LS_Cpp20;
 
   LangOpts.LineComment = 1;
   bool AlternativeOperators = Style.isCpp();

diff --git a/clang/lib/Lex/Preprocessor.cpp b/clang/lib/Lex/Preprocessor.cpp
index 961b55c..105aa66 100644 (file)
--- a/clang/lib/Lex/Preprocessor.cpp
+++ b/clang/lib/Lex/Preprocessor.cpp
@@ -774,6 +774,10 @@ static diag::kind getFutureCompatDiagKind(const IdentifierInfo &II,
 #define CXX20_KEYWORD(NAME, FLAGS)                                             \
         .Case(#NAME, diag::warn_cxx20_keyword)
 #include "clang/Basic/TokenKinds.def"
+        // char8_t is not modeled as a CXX20_KEYWORD because it's not
+        // unconditionally enabled in C++20 mode. (It can be disabled
+        // by -fno-char8_t.)
+        .Case("char8_t", diag::warn_cxx20_keyword)
         ;
 
   llvm_unreachable(

diff --git a/clang/test/Lexer/char8_t.cpp b/clang/test/Lexer/char8_t.cpp
index 20f820e..d65597c 100644 (file)
--- a/clang/test/Lexer/char8_t.cpp
+++ b/clang/test/Lexer/char8_t.cpp
@@ -1,5 +1,14 @@
-// RUN: %clang_cc1 -std=c++2a -verify %s
-// RUN: %clang_cc1 -std=c++2a -verify %s -fchar8_t
+// RUN: %clang_cc1 -std=c++20 -verify %s -DCHAR8_T
+// RUN: %clang_cc1 -std=c++20 -verify %s -fchar8_t -DCHAR8_T
+// RUN: %clang_cc1 -std=c++17 -verify %s -fchar8_t -DCHAR8_T
+
+// RUN: %clang_cc1 -std=c++17 -verify %s
+// RUN: %clang_cc1 -std=c++17 -verify %s -fno-char8_t
+// RUN: %clang_cc1 -std=c++20 -verify %s -fno-char8_t
+
+#if defined(__cpp_char8_t) != defined(CHAR8_T)
+#error wrong setting for __cpp_char_t
+#endif
 
 #if defined(__cpp_char8_t) && __is_identifier(char8_t)
 #error char8_t is an identifier under -fchar8_t