[libc++] Optimize ctype::to{lower,upper} (#145344)

philnik777 · web-flow · commit 44d37695a5e1 · 2025-07-09T16:32:01.000+02:00
```
----------------------------------------------
Benchmark                       old        new
--------------------------- ------------------
BM_tolower_char&lt;char&gt;       1.64 ns    1.41 ns
BM_tolower_char&lt;wchar_t&gt;    1.64 ns    1.41 ns
BM_tolower_string&lt;char&gt;     32.4 ns    12.8 ns
BM_tolower_string&lt;wchar_t&gt;  32.9 ns    15.1 ns
BM_toupper_char&lt;char&gt;       1.63 ns    1.64 ns
BM_toupper_char&lt;wchar_t&gt;    1.63 ns    1.41 ns
BM_toupper_string&lt;char&gt;     32.2 ns    12.7 ns
BM_toupper_string&lt;wchar_t&gt;  33.0 ns    15.1 ns
```
diff --git a/libcxx/docs/ReleaseNotes/21.rst b/libcxx/docs/ReleaseNotes/21.rst
@@ -85,6 +85,8 @@ Improvements and New Features
 - The ``flat_map::insert`` and ``flat_set::insert_range`` have been optimized, resulting in a performance improvement of up
   to 10x for inserting elements into a ``flat_map`` when the input range is a ``flat_map`` or a ``zip_view``.
 
+- ``ctype::tolower`` and ``ctype::toupper`` have been optimized, resulting in a 2x performane improvement.
+
 Deprecations and Removals
 -------------------------
 
diff --git a/libcxx/include/__config b/libcxx/include/__config
@@ -639,10 +639,6 @@ typedef __char32_t char32_t;
 #    define _LIBCPP_HAS_C11_ALIGNED_ALLOC 1
 #  endif
 
-#  if defined(__APPLE__) || defined(__FreeBSD__)
-#    define _LIBCPP_HAS_DEFAULTRUNELOCALE
-#  endif
-
 #  if defined(__APPLE__) || defined(__FreeBSD__)
 #    define _LIBCPP_WCTYPE_IS_MASK
 #  endif
diff --git a/libcxx/include/__locale b/libcxx/include/__locale
@@ -589,18 +589,6 @@ public:
 #  endif
   _LIBCPP_HIDE_FROM_ABI const mask* table() const _NOEXCEPT { return __tab_; }
   static const mask* classic_table() _NOEXCEPT;
-#  if defined(__GLIBC__) || defined(__EMSCRIPTEN__)
-  static const int* __classic_upper_table() _NOEXCEPT;
-  static const int* __classic_lower_table() _NOEXCEPT;
-#  endif
-#  if defined(__NetBSD__)
-  static const short* __classic_upper_table() _NOEXCEPT;
-  static const short* __classic_lower_table() _NOEXCEPT;
-#  endif
-#  if defined(__MVS__)
-  static const unsigned short* __classic_upper_table() _NOEXCEPT;
-  static const unsigned short* __classic_lower_table() _NOEXCEPT;
-#  endif
 
 protected:
   ~ctype() override;
diff --git a/libcxx/include/__locale_dir/locale_base_api.h b/libcxx/include/__locale_dir/locale_base_api.h
@@ -64,8 +64,6 @@
 // Character manipulation functions
 // --------------------------------
 // namespace __locale {
-//  int     __islower(int, __locale_t);
-//  int     __isupper(int, __locale_t);
 //  int     __isdigit(int, __locale_t);  // required by the headers
 //  int     __isxdigit(int, __locale_t); // required by the headers
 //  int     __toupper(int, __locale_t);
@@ -208,11 +206,6 @@ __strtoull(const char* __nptr, char** __endptr, int __base, __locale_t __loc) {
 //
 // Character manipulation functions
 //
-#    if defined(_LIBCPP_BUILDING_LIBRARY)
-inline _LIBCPP_HIDE_FROM_ABI int __islower(int __ch, __locale_t __loc) { return islower_l(__ch, __loc); }
-inline _LIBCPP_HIDE_FROM_ABI int __isupper(int __ch, __locale_t __loc) { return isupper_l(__ch, __loc); }
-#    endif
-
 inline _LIBCPP_HIDE_FROM_ABI int __isdigit(int __ch, __locale_t __loc) { return isdigit_l(__ch, __loc); }
 inline _LIBCPP_HIDE_FROM_ABI int __isxdigit(int __ch, __locale_t __loc) { return isxdigit_l(__ch, __loc); }
 
diff --git a/libcxx/include/__locale_dir/support/bsd_like.h b/libcxx/include/__locale_dir/support/bsd_like.h
@@ -89,12 +89,6 @@ __strtoull(const char* __nptr, char** __endptr, int __base, __locale_t __loc) {
 //
 // Character manipulation functions
 //
-#if defined(_LIBCPP_BUILDING_LIBRARY)
-inline _LIBCPP_HIDE_FROM_ABI int __islower(int __c, __locale_t __loc) { return ::islower_l(__c, __loc); }
-
-inline _LIBCPP_HIDE_FROM_ABI int __isupper(int __c, __locale_t __loc) { return ::isupper_l(__c, __loc); }
-#endif
-
 inline _LIBCPP_HIDE_FROM_ABI int __isdigit(int __c, __locale_t __loc) { return ::isdigit_l(__c, __loc); }
 
 inline _LIBCPP_HIDE_FROM_ABI int __isxdigit(int __c, __locale_t __loc) { return ::isxdigit_l(__c, __loc); }
diff --git a/libcxx/include/__locale_dir/support/linux.h b/libcxx/include/__locale_dir/support/linux.h
@@ -116,12 +116,6 @@ __strtoull(const char* __nptr, char** __endptr, int __base, __locale_t __loc) {
 //
 // Character manipulation functions
 //
-#if defined(_LIBCPP_BUILDING_LIBRARY)
-inline _LIBCPP_HIDE_FROM_ABI int __islower(int __c, __locale_t __loc) { return islower_l(__c, __loc); }
-
-inline _LIBCPP_HIDE_FROM_ABI int __isupper(int __c, __locale_t __loc) { return isupper_l(__c, __loc); }
-#endif
-
 inline _LIBCPP_HIDE_FROM_ABI int __isdigit(int __c, __locale_t __loc) { return isdigit_l(__c, __loc); }
 
 inline _LIBCPP_HIDE_FROM_ABI int __isxdigit(int __c, __locale_t __loc) { return isxdigit_l(__c, __loc); }
diff --git a/libcxx/include/__locale_dir/support/no_locale/characters.h b/libcxx/include/__locale_dir/support/no_locale/characters.h
@@ -29,12 +29,6 @@ namespace __locale {
 //
 // Character manipulation functions
 //
-#if defined(_LIBCPP_BUILDING_LIBRARY)
-inline _LIBCPP_HIDE_FROM_ABI int __islower(int __c, __locale_t) { return std::islower(__c); }
-
-inline _LIBCPP_HIDE_FROM_ABI int __isupper(int __c, __locale_t) { return std::isupper(__c); }
-#endif
-
 inline _LIBCPP_HIDE_FROM_ABI int __isdigit(int __c, __locale_t) { return std::isdigit(__c); }
 
 inline _LIBCPP_HIDE_FROM_ABI int __isxdigit(int __c, __locale_t) { return std::isxdigit(__c); }
diff --git a/libcxx/include/__locale_dir/support/windows.h b/libcxx/include/__locale_dir/support/windows.h
@@ -197,12 +197,6 @@ __strtoull(const char* __nptr, char** __endptr, int __base, __locale_t __loc) {
 //
 // Character manipulation functions
 //
-#if defined(_LIBCPP_BUILDING_LIBRARY)
-inline _LIBCPP_HIDE_FROM_ABI int __islower(int __c, __locale_t __loc) { return _islower_l(__c, __loc); }
-
-inline _LIBCPP_HIDE_FROM_ABI int __isupper(int __c, __locale_t __loc) { return _isupper_l(__c, __loc); }
-#endif
-
 inline _LIBCPP_HIDE_FROM_ABI int __isdigit(int __c, __locale_t __loc) { return _isdigit_l(__c, __loc); }
 
 inline _LIBCPP_HIDE_FROM_ABI int __isxdigit(int __c, __locale_t __loc) { return _isxdigit_l(__c, __loc); }
diff --git a/libcxx/lib/abi/CHANGELOG.TXT b/libcxx/lib/abi/CHANGELOG.TXT
@@ -16,6 +16,15 @@ New entries should be added directly below the "Version" header.
 Version 21.0
 ------------
 
+* [libc++] Optimize ctype::to{lower,upper}
+
+  This patch removed __classic_upper_table() and __classic_lower_table(), which were only ever accessed from the dylib.
+
+  x86_64-unknown-gnu-linux
+  ------------------------
+  Symbol removed: _ZNSt3__15ctypeIcE21__classic_lower_tableEv
+  Symbol removed: _ZNSt3__15ctypeIcE21__classic_upper_tableEv
+
 * [libc++] Instantiate hash function externally
 
   This has multiple benefits:
diff --git a/libcxx/lib/abi/x86_64-unknown-linux-gnu.libcxxabi.v1.stable.exceptions.nonew.abilist b/libcxx/lib/abi/x86_64-unknown-linux-gnu.libcxxabi.v1.stable.exceptions.nonew.abilist
@@ -1318,8 +1318,6 @@
 {'is_defined': True, 'name': '_ZNSt3__15alignEmmRPvRm', 'type': 'FUNC'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcE10table_sizeE', 'size': 8, 'type': 'OBJECT'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcE13classic_tableEv', 'type': 'FUNC'}
-{'is_defined': True, 'name': '_ZNSt3__15ctypeIcE21__classic_lower_tableEv', 'type': 'FUNC'}
-{'is_defined': True, 'name': '_ZNSt3__15ctypeIcE21__classic_upper_tableEv', 'type': 'FUNC'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcE2idE', 'size': 16, 'type': 'OBJECT'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcEC1EPKtbm', 'type': 'FUNC'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcEC2EPKtbm', 'type': 'FUNC'}
diff --git a/libcxx/lib/abi/x86_64-unknown-linux-gnu.libcxxabi.v1.stable.noexceptions.nonew.abilist b/libcxx/lib/abi/x86_64-unknown-linux-gnu.libcxxabi.v1.stable.noexceptions.nonew.abilist
@@ -1289,8 +1289,6 @@
 {'is_defined': True, 'name': '_ZNSt3__15alignEmmRPvRm', 'type': 'FUNC'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcE10table_sizeE', 'size': 8, 'type': 'OBJECT'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcE13classic_tableEv', 'type': 'FUNC'}
-{'is_defined': True, 'name': '_ZNSt3__15ctypeIcE21__classic_lower_tableEv', 'type': 'FUNC'}
-{'is_defined': True, 'name': '_ZNSt3__15ctypeIcE21__classic_upper_tableEv', 'type': 'FUNC'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcE2idE', 'size': 16, 'type': 'OBJECT'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcEC1EPKtbm', 'type': 'FUNC'}
 {'is_defined': True, 'name': '_ZNSt3__15ctypeIcEC2EPKtbm', 'type': 'FUNC'}
diff --git a/libcxx/src/locale.cpp b/libcxx/src/locale.cpp
@@ -697,6 +697,20 @@ const ctype_base::mask ctype_base::graph;
 
 // template <> class ctype<wchar_t>;
 
+template <class CharT>
+static CharT to_upper_impl(CharT c) {
+  if (c < 'a' || c > 'z')
+    return c;
+  return c & ~0x20;
+}
+
+template <class CharT>
+static CharT to_lower_impl(CharT c) {
+  if (c < 'A' || c > 'Z')
+    return c;
+  return c | 0x20;
+}
+
 #if _LIBCPP_HAS_WIDE_CHARACTERS
 constinit locale::id ctype<wchar_t>::id;
 
@@ -726,48 +740,19 @@ const wchar_t* ctype<wchar_t>::do_scan_not(mask m, const char_type* low, const c
   return low;
 }
 
-wchar_t ctype<wchar_t>::do_toupper(char_type c) const {
-#  ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-  return std::__libcpp_isascii(c) ? _DefaultRuneLocale.__mapupper[c] : c;
-#  elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) || defined(__MVS__)
-  return std::__libcpp_isascii(c) ? ctype<char>::__classic_upper_table()[c] : c;
-#  else
-  return (std::__libcpp_isascii(c) && __locale::__iswlower(c, _LIBCPP_GET_C_LOCALE)) ? c - L'a' + L'A' : c;
-#  endif
-}
+wchar_t ctype<wchar_t>::do_toupper(char_type c) const { return to_upper_impl(c); }
 
 const wchar_t* ctype<wchar_t>::do_toupper(char_type* low, const char_type* high) const {
   for (; low != high; ++low)
-#  ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-    *low = std::__libcpp_isascii(*low) ? _DefaultRuneLocale.__mapupper[*low] : *low;
-#  elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) || defined(__MVS__)
-    *low = std::__libcpp_isascii(*low) ? ctype<char>::__classic_upper_table()[*low] : *low;
-#  else
-    *low =
-        (std::__libcpp_isascii(*low) && __locale::__islower(*low, _LIBCPP_GET_C_LOCALE)) ? (*low - L'a' + L'A') : *low;
-#  endif
+    *low = to_upper_impl(*low);
   return low;
 }
 
-wchar_t ctype<wchar_t>::do_tolower(char_type c) const {
-#  ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-  return std::__libcpp_isascii(c) ? _DefaultRuneLocale.__maplower[c] : c;
-#  elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) || defined(__MVS__)
-  return std::__libcpp_isascii(c) ? ctype<char>::__classic_lower_table()[c] : c;
-#  else
-  return (std::__libcpp_isascii(c) && __locale::__isupper(c, _LIBCPP_GET_C_LOCALE)) ? c - L'A' + 'a' : c;
-#  endif
-}
+wchar_t ctype<wchar_t>::do_tolower(char_type c) const { return to_lower_impl(c); }
 
 const wchar_t* ctype<wchar_t>::do_tolower(char_type* low, const char_type* high) const {
   for (; low != high; ++low)
-#  ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-    *low = std::__libcpp_isascii(*low) ? _DefaultRuneLocale.__maplower[*low] : *low;
-#  elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) || defined(__MVS__)
-    *low = std::__libcpp_isascii(*low) ? ctype<char>::__classic_lower_table()[*low] : *low;
-#  else
-    *low = (std::__libcpp_isascii(*low) && __locale::__isupper(*low, _LIBCPP_GET_C_LOCALE)) ? *low - L'A' + L'a' : *low;
-#  endif
+    *low = to_lower_impl(*low);
   return low;
 }
 
@@ -811,59 +796,19 @@ ctype<char>::~ctype() {
     delete[] __tab_;
 }
 
-char ctype<char>::do_toupper(char_type c) const {
-#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-  return std::__libcpp_isascii(c) ? static_cast<char>(_DefaultRuneLocale.__mapupper[static_cast<ptrdiff_t>(c)]) : c;
-#elif defined(__NetBSD__)
-  return static_cast<char>(__classic_upper_table()[static_cast<unsigned char>(c)]);
-#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__MVS__)
-  return std::__libcpp_isascii(c) ? static_cast<char>(__classic_upper_table()[static_cast<unsigned char>(c)]) : c;
-#else
-  return (std::__libcpp_isascii(c) && __locale::__islower(c, _LIBCPP_GET_C_LOCALE)) ? c - 'a' + 'A' : c;
-#endif
-}
+char ctype<char>::do_toupper(char_type c) const { return to_upper_impl(c); }
 
 const char* ctype<char>::do_toupper(char_type* low, const char_type* high) const {
   for (; low != high; ++low)
-#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-    *low = std::__libcpp_isascii(*low)
-             ? static_cast<char>(_DefaultRuneLocale.__mapupper[static_cast<ptrdiff_t>(*low)])
-             : *low;
-#elif defined(__NetBSD__)
-    *low = static_cast<char>(__classic_upper_table()[static_cast<unsigned char>(*low)]);
-#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__MVS__)
-    *low = std::__libcpp_isascii(*low) ? static_cast<char>(__classic_upper_table()[static_cast<size_t>(*low)]) : *low;
-#else
-    *low = (std::__libcpp_isascii(*low) && __locale::__islower(*low, _LIBCPP_GET_C_LOCALE)) ? *low - 'a' + 'A' : *low;
-#endif
+    *low = to_upper_impl(*low);
   return low;
 }
 
-char ctype<char>::do_tolower(char_type c) const {
-#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-  return std::__libcpp_isascii(c) ? static_cast<char>(_DefaultRuneLocale.__maplower[static_cast<ptrdiff_t>(c)]) : c;
-#elif defined(__NetBSD__)
-  return static_cast<char>(__classic_lower_table()[static_cast<unsigned char>(c)]);
-#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__MVS__)
-  return std::__libcpp_isascii(c) ? static_cast<char>(__classic_lower_table()[static_cast<size_t>(c)]) : c;
-#else
-  return (std::__libcpp_isascii(c) && __locale::__isupper(c, _LIBCPP_GET_C_LOCALE)) ? c - 'A' + 'a' : c;
-#endif
-}
+char ctype<char>::do_tolower(char_type c) const { return to_lower_impl(c); }
 
 const char* ctype<char>::do_tolower(char_type* low, const char_type* high) const {
   for (; low != high; ++low)
-#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE
-    *low = std::__libcpp_isascii(*low)
-             ? static_cast<char>(_DefaultRuneLocale.__maplower[static_cast<ptrdiff_t>(*low)])
-             : *low;
-#elif defined(__NetBSD__)
-    *low = static_cast<char>(__classic_lower_table()[static_cast<unsigned char>(*low)]);
-#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__MVS__)
-    *low = std::__libcpp_isascii(*low) ? static_cast<char>(__classic_lower_table()[static_cast<size_t>(*low)]) : *low;
-#else
-    *low = (std::__libcpp_isascii(*low) && __locale::__isupper(*low, _LIBCPP_GET_C_LOCALE)) ? *low - 'A' + 'a' : *low;
-#endif
+    *low = to_lower_impl(*low);
   return low;
 }
 
@@ -1010,36 +955,6 @@ const ctype<char>::mask* ctype<char>::classic_table() noexcept {
 }
 #endif
 
-#if defined(__GLIBC__)
-const int* ctype<char>::__classic_lower_table() noexcept { return _LIBCPP_GET_C_LOCALE->__ctype_tolower; }
-
-const int* ctype<char>::__classic_upper_table() noexcept { return _LIBCPP_GET_C_LOCALE->__ctype_toupper; }
-#elif defined(__NetBSD__)
-const short* ctype<char>::__classic_lower_table() noexcept { return _C_tolower_tab_ + 1; }
-
-const short* ctype<char>::__classic_upper_table() noexcept { return _C_toupper_tab_ + 1; }
-
-#elif defined(__EMSCRIPTEN__)
-const int* ctype<char>::__classic_lower_table() noexcept { return *__ctype_tolower_loc(); }
-
-const int* ctype<char>::__classic_upper_table() noexcept { return *__ctype_toupper_loc(); }
-#elif defined(__MVS__)
-const unsigned short* ctype<char>::__classic_lower_table() _NOEXCEPT {
-#  if defined(__NATIVE_ASCII_F)
-  return const_cast<const unsigned short*>(__OBJ_DATA(__lc_ctype_a)->lower);
-#  else
-  return const_cast<const unsigned short*>(__ctype + __TOLOWER_INDEX);
-#  endif
-}
-const unsigned short* ctype<char>::__classic_upper_table() _NOEXCEPT {
-#  if defined(__NATIVE_ASCII_F)
-  return const_cast<const unsigned short*>(__OBJ_DATA(__lc_ctype_a)->upper);
-#  else
-  return const_cast<const unsigned short*>(__ctype + __TOUPPER_INDEX);
-#  endif
-}
-#endif // __GLIBC__ || __NETBSD__ || __EMSCRIPTEN__ || __MVS__
-
 // template <> class ctype_byname<char>
 
 ctype_byname<char>::ctype_byname(const char* name, size_t refs)
diff --git a/libcxx/test/benchmarks/locale/ctype.bench.cpp b/libcxx/test/benchmarks/locale/ctype.bench.cpp
