Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 1 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" |
| 2 | "http://www.w3.org/TR/html4/strict.dtd"> |
| 3 | <!-- Material used from: HTML 4.01 specs: http://www.w3.org/TR/html401/ --> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 4 | <html> |
| 5 | <head> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 6 | <META http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"> |
| 7 | <title>Clang LanguageExtensions</title> |
Sean Hunt | 64f857b | 2011-06-23 01:22:53 +0000 | [diff] [blame] | 8 | <link type="text/css" rel="stylesheet" href="../menu.css"> |
| 9 | <link type="text/css" rel="stylesheet" href="../content.css"> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 10 | <style type="text/css"> |
| 11 | td { |
| 12 | vertical-align: top; |
| 13 | } |
| 14 | </style> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 15 | </head> |
| 16 | <body> |
| 17 | |
| 18 | <!--#include virtual="../menu.html.incl"--> |
| 19 | |
| 20 | <div id="content"> |
| 21 | |
| 22 | <h1>Clang Language Extensions</h1> |
| 23 | |
| 24 | <ul> |
| 25 | <li><a href="#intro">Introduction</a></li> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 26 | <li><a href="#feature_check">Feature Checking Macros</a></li> |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 27 | <li><a href="#has_include">Include File Checking Macros</a></li> |
Chris Lattner | 81edc9f | 2009-04-13 02:45:46 +0000 | [diff] [blame] | 28 | <li><a href="#builtinmacros">Builtin Macros</a></li> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 29 | <li><a href="#vectors">Vectors and Extended Vectors</a></li> |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 30 | <li><a href="#deprecated">Messages on <tt>deprecated</tt> and <tt>unavailable</tt> attributes</a></li> |
| 31 | <li><a href="#attributes-on-enumerators">Attributes on enumerators</a></li> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 32 | <li><a href="#checking_language_features">Checks for Standard Language Features</a> |
Ted Kremenek | 22c3410 | 2009-12-03 02:05:57 +0000 | [diff] [blame] | 33 | <ul> |
| 34 | <li><a href="#cxx_exceptions">C++ exceptions</a></li> |
| 35 | <li><a href="#cxx_rtti">C++ RTTI</a></li> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 36 | </ul></li> |
| 37 | <li><a href="#checking_upcoming_features">Checks for Upcoming Standard Language Features</a> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 38 | <ul> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 39 | <li><a href="#cxx0x">C++0x</a> |
| 40 | <ul> |
| 41 | <li><a href="#cxx_decltype">C++0x <tt>decltype()</tt></a></li> |
| 42 | <li><a href="#cxx_access_control_sfinae">C++0x SFINAE includes access control</a></li> |
| 43 | <li><a href="#cxx_alias_templates">C++0x alias templates</a></li> |
| 44 | <li><a href="#cxx_attributes">C++0x attributes</a></li> |
| 45 | <li><a href="#cxx_default_function_template_args">C++0x default template arguments in function templates</a></li> |
Sean Hunt | d962499 | 2011-06-23 06:11:37 +0000 | [diff] [blame] | 46 | <li><a href="#cxx_delegating_constructor">C++0x delegating constructors</a></li> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 47 | <li><a href="#cxx_deleted_functions">C++0x deleted functions</a></li> |
| 48 | <li><a href="#cxx_lambdas">C++0x lambdas</a></li> |
| 49 | <li><a href="#cxx_nullptr">C++0x nullptr</a></li> |
| 50 | <li><a href="#cxx_override_control">C++0x override control</a></li> |
| 51 | <li><a href="#cxx_range_for">C++0x range-based for loop</a></li> |
| 52 | <li><a href="#cxx_rvalue_references">C++0x rvalue references</a></li> |
| 53 | <li><a href="#cxx_reference_qualified_functions">C++0x reference-qualified functions</a></li> |
| 54 | <li><a href="#cxx_static_assert">C++0x <tt>static_assert()</tt></a></li> |
| 55 | <li><a href="#cxx_auto_type">C++0x type inference</a></li> |
| 56 | <li><a href="#cxx_variadic_templates">C++0x variadic templates</a></li> |
| 57 | <li><a href="#cxx_inline_namespaces">C++0x inline namespaces</a></li> |
| 58 | <li><a href="#cxx_strong_enums">C++0x strongly-typed enumerations</a></li> |
| 59 | <li><a href="#cxx_trailing_return">C++0x trailing return type</a></li> |
| 60 | <li><a href="#cxx_noexcept">C++0x noexcept specification</a></li> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 61 | </ul></li> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 62 | <li><a href="#c1x">C1X</a> |
| 63 | <ul> |
| 64 | <li><a href="#c_generic_selections">C1X generic selections</a></li> |
| 65 | <li><a href="#c_static_assert">C1X <tt>_Static_assert()</tt></a></li> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 66 | </ul></li> |
| 67 | </ul> </li> |
Douglas Gregor | afdf137 | 2011-02-03 21:57:35 +0000 | [diff] [blame] | 68 | <li><a href="#checking_type_traits">Checks for Type Traits</a></li> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 69 | <li><a href="#blocks">Blocks</a></li> |
Douglas Gregor | 926df6c | 2011-06-11 01:09:30 +0000 | [diff] [blame] | 70 | <li><a href="#objc_features">Objective-C Features</a> |
| 71 | <ul> |
| 72 | <li><a href="#objc_instancetype">Related result types</a></li> |
John McCall | f85e193 | 2011-06-15 23:02:42 +0000 | [diff] [blame] | 73 | <li><a href="#objc_arc">Automatic reference counting</a></li> |
Douglas Gregor | 926df6c | 2011-06-11 01:09:30 +0000 | [diff] [blame] | 74 | </ul> |
| 75 | </li> |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 76 | <li><a href="#overloading-in-c">Function Overloading in C</a></li> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 77 | <li><a href="#builtins">Builtin Functions</a> |
| 78 | <ul> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 79 | <li><a href="#__builtin_shufflevector">__builtin_shufflevector</a></li> |
Chris Lattner | 21190d5 | 2009-09-21 03:09:59 +0000 | [diff] [blame] | 80 | <li><a href="#__builtin_unreachable">__builtin_unreachable</a></li> |
Chris Lattner | 23aa9c8 | 2011-04-09 03:57:26 +0000 | [diff] [blame] | 81 | <li><a href="#__sync_swap">__sync_swap</a></li> |
Douglas Gregor | afdf137 | 2011-02-03 21:57:35 +0000 | [diff] [blame] | 82 | </ul> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 83 | </li> |
Chris Lattner | 1177f91 | 2009-04-09 19:58:15 +0000 | [diff] [blame] | 84 | <li><a href="#targetspecific">Target-Specific Extensions</a> |
| 85 | <ul> |
| 86 | <li><a href="#x86-specific">X86/X86-64 Language Extensions</a></li> |
| 87 | </ul> |
| 88 | </li> |
John McCall | 8749401 | 2011-03-18 03:51:49 +0000 | [diff] [blame] | 89 | <li><a href="#analyzerspecific">Static Analysis-Specific Extensions</a></li> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 90 | </ul> |
| 91 | |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 92 | <!-- ======================================================================= --> |
| 93 | <h2 id="intro">Introduction</h2> |
| 94 | <!-- ======================================================================= --> |
| 95 | |
| 96 | <p>This document describes the language extensions provided by Clang. In |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 97 | addition to the language extensions listed here, Clang aims to support a broad |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 98 | range of GCC extensions. Please see the <a |
| 99 | href="http://gcc.gnu.org/onlinedocs/gcc/C-Extensions.html">GCC manual</a> for |
| 100 | more information on these extensions.</p> |
| 101 | |
| 102 | <!-- ======================================================================= --> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 103 | <h2 id="feature_check">Feature Checking Macros</h2> |
| 104 | <!-- ======================================================================= --> |
| 105 | |
| 106 | <p>Language extensions can be very useful, but only if you know you can depend |
Chris Lattner | c70e193 | 2011-03-21 16:25:11 +0000 | [diff] [blame] | 107 | on them. In order to allow fine-grain features checks, we support three builtin |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 108 | function-like macros. This allows you to directly test for a feature in your |
| 109 | code without having to resort to something like autoconf or fragile "compiler |
| 110 | version checks".</p> |
| 111 | |
| 112 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 113 | <h3><a name="__has_builtin">__has_builtin</a></h3> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 114 | <!-- ======================================================================= --> |
| 115 | |
| 116 | <p>This function-like macro takes a single identifier argument that is the name |
| 117 | of a builtin function. It evaluates to 1 if the builtin is supported or 0 if |
| 118 | not. It can be used like this:</p> |
| 119 | |
| 120 | <blockquote> |
| 121 | <pre> |
| 122 | #ifndef __has_builtin // Optional of course. |
| 123 | #define __has_builtin(x) 0 // Compatibility with non-clang compilers. |
| 124 | #endif |
| 125 | |
| 126 | ... |
| 127 | #if __has_builtin(__builtin_trap) |
| 128 | __builtin_trap(); |
| 129 | #else |
| 130 | abort(); |
| 131 | #endif |
| 132 | ... |
| 133 | </pre> |
| 134 | </blockquote> |
| 135 | |
| 136 | |
| 137 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 138 | <h3><a name="__has_feature_extension"> __has_feature and __has_extension</a></h3> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 139 | <!-- ======================================================================= --> |
| 140 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 141 | <p>These function-like macros take a single identifier argument that is the |
| 142 | name of a feature. <code>__has_feature</code> evaluates to 1 if the feature |
| 143 | is both supported by Clang and standardized in the current language standard |
| 144 | or 0 if not (but see <a href="#has_feature_back_compat">below</a>), while |
| 145 | <code>__has_extension</code> evaluates to 1 if the feature is supported by |
| 146 | Clang in the current language (either as a language extension or a standard |
| 147 | language feature) or 0 if not. They can be used like this:</p> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 148 | |
| 149 | <blockquote> |
| 150 | <pre> |
| 151 | #ifndef __has_feature // Optional of course. |
| 152 | #define __has_feature(x) 0 // Compatibility with non-clang compilers. |
| 153 | #endif |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 154 | #ifndef __has_extension |
| 155 | #define __has_extension __has_feature // Compatibility with pre-3.0 compilers. |
| 156 | #endif |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 157 | |
| 158 | ... |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 159 | #if __has_feature(cxx_rvalue_references) |
| 160 | // This code will only be compiled with the -std=c++0x and -std=gnu++0x |
| 161 | // options, because rvalue references are only standardized in C++0x. |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 162 | #endif |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 163 | |
| 164 | #if __has_extension(cxx_rvalue_references) |
| 165 | // This code will be compiled with the -std=c++0x, -std=gnu++0x, -std=c++98 |
| 166 | // and -std=gnu++98 options, because rvalue references are supported as a |
| 167 | // language extension in C++98. |
| 168 | #endif |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 169 | </pre> |
| 170 | </blockquote> |
| 171 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 172 | <p id="has_feature_back_compat">For backwards compatibility reasons, |
| 173 | <code>__has_feature</code> can also be used to test for support for |
| 174 | non-standardized features, i.e. features not prefixed <code>c_</code>, |
| 175 | <code>cxx_</code> or <code>objc_</code>.</p> |
| 176 | |
| 177 | <p>If the <code>-pedantic-errors</code> option is given, |
| 178 | <code>__has_extension</code> is equivalent to <code>__has_feature</code>.</p> |
| 179 | |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 180 | <p>The feature tag is described along with the language feature below.</p> |
| 181 | |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 182 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 183 | <h3><a name="__has_attribute">__has_attribute</a></h3> |
Anders Carlsson | cae5095 | 2010-10-20 02:31:43 +0000 | [diff] [blame] | 184 | <!-- ======================================================================= --> |
| 185 | |
| 186 | <p>This function-like macro takes a single identifier argument that is the name |
| 187 | of an attribute. It evaluates to 1 if the attribute is supported or 0 if not. It |
| 188 | can be used like this:</p> |
| 189 | |
| 190 | <blockquote> |
| 191 | <pre> |
| 192 | #ifndef __has_attribute // Optional of course. |
| 193 | #define __has_attribute(x) 0 // Compatibility with non-clang compilers. |
| 194 | #endif |
| 195 | |
| 196 | ... |
Anders Carlsson | 961003d | 2011-01-24 03:54:51 +0000 | [diff] [blame] | 197 | #if __has_attribute(always_inline) |
| 198 | #define ALWAYS_INLINE __attribute__((always_inline)) |
Anders Carlsson | cae5095 | 2010-10-20 02:31:43 +0000 | [diff] [blame] | 199 | #else |
Anders Carlsson | 961003d | 2011-01-24 03:54:51 +0000 | [diff] [blame] | 200 | #define ALWAYS_INLINE |
Anders Carlsson | cae5095 | 2010-10-20 02:31:43 +0000 | [diff] [blame] | 201 | #endif |
| 202 | ... |
| 203 | </pre> |
| 204 | </blockquote> |
| 205 | |
| 206 | <!-- ======================================================================= --> |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 207 | <h2 id="has_include">Include File Checking Macros</h2> |
| 208 | <!-- ======================================================================= --> |
| 209 | |
| 210 | <p>Not all developments systems have the same include files. |
| 211 | The <a href="#__has_include">__has_include</a> and |
| 212 | <a href="#__has_include_next">__has_include_next</a> macros allow you to |
| 213 | check for the existence of an include file before doing |
| 214 | a possibly failing #include directive.</p> |
| 215 | |
| 216 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 217 | <h3><a name="__has_include">__has_include</a></h3> |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 218 | <!-- ======================================================================= --> |
| 219 | |
| 220 | <p>This function-like macro takes a single file name string argument that |
| 221 | is the name of an include file. It evaluates to 1 if the file can |
| 222 | be found using the include paths, or 0 otherwise:</p> |
| 223 | |
| 224 | <blockquote> |
| 225 | <pre> |
| 226 | // Note the two possible file name string formats. |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 227 | #if __has_include("myinclude.h") && __has_include(<stdint.h>) |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 228 | # include "myinclude.h" |
| 229 | #endif |
| 230 | |
| 231 | // To avoid problem with non-clang compilers not having this macro. |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 232 | #if defined(__has_include) && __has_include("myinclude.h") |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 233 | # include "myinclude.h" |
| 234 | #endif |
| 235 | </pre> |
| 236 | </blockquote> |
| 237 | |
| 238 | <p>To test for this feature, use #if defined(__has_include).</p> |
| 239 | |
| 240 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 241 | <h3><a name="__has_include_next">__has_include_next</a></h3> |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 242 | <!-- ======================================================================= --> |
| 243 | |
| 244 | <p>This function-like macro takes a single file name string argument that |
| 245 | is the name of an include file. It is like __has_include except that it |
| 246 | looks for the second instance of the given file found in the include |
| 247 | paths. It evaluates to 1 if the second instance of the file can |
| 248 | be found using the include paths, or 0 otherwise:</p> |
| 249 | |
| 250 | <blockquote> |
| 251 | <pre> |
| 252 | // Note the two possible file name string formats. |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 253 | #if __has_include_next("myinclude.h") && __has_include_next(<stdint.h>) |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 254 | # include_next "myinclude.h" |
| 255 | #endif |
| 256 | |
| 257 | // To avoid problem with non-clang compilers not having this macro. |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 258 | #if defined(__has_include_next) && __has_include_next("myinclude.h") |
John Thompson | 92bd8c7 | 2009-11-02 22:28:12 +0000 | [diff] [blame] | 259 | # include_next "myinclude.h" |
| 260 | #endif |
| 261 | </pre> |
| 262 | </blockquote> |
| 263 | |
| 264 | <p>Note that __has_include_next, like the GNU extension |
| 265 | #include_next directive, is intended for use in headers only, |
| 266 | and will issue a warning if used in the top-level compilation |
| 267 | file. A warning will also be issued if an absolute path |
| 268 | is used in the file argument.</p> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 269 | |
| 270 | <!-- ======================================================================= --> |
Chris Lattner | 81edc9f | 2009-04-13 02:45:46 +0000 | [diff] [blame] | 271 | <h2 id="builtinmacros">Builtin Macros</h2> |
| 272 | <!-- ======================================================================= --> |
| 273 | |
Douglas Gregor | 4290fbd | 2010-04-30 02:51:06 +0000 | [diff] [blame] | 274 | <dl> |
| 275 | <dt><code>__BASE_FILE__</code></dt> |
| 276 | <dd>Defined to a string that contains the name of the main input |
| 277 | file passed to Clang.</dd> |
| 278 | |
| 279 | <dt><code>__COUNTER__</code></dt> |
| 280 | <dd>Defined to an integer value that starts at zero and is |
| 281 | incremented each time the <code>__COUNTER__</code> macro is |
| 282 | expanded.</dd> |
| 283 | |
| 284 | <dt><code>__INCLUDE_LEVEL__</code></dt> |
| 285 | <dd>Defined to an integral value that is the include depth of the |
| 286 | file currently being translated. For the main file, this value is |
| 287 | zero.</dd> |
| 288 | |
| 289 | <dt><code>__TIMESTAMP__</code></dt> |
| 290 | <dd>Defined to the date and time of the last modification of the |
| 291 | current source file.</dd> |
| 292 | |
| 293 | <dt><code>__clang__</code></dt> |
| 294 | <dd>Defined when compiling with Clang</dd> |
| 295 | |
| 296 | <dt><code>__clang_major__</code></dt> |
| 297 | <dd>Defined to the major version number of Clang (e.g., the 2 in |
| 298 | 2.0.1).</dd> |
| 299 | |
| 300 | <dt><code>__clang_minor__</code></dt> |
| 301 | <dd>Defined to the minor version number of Clang (e.g., the 0 in |
| 302 | 2.0.1).</dd> |
| 303 | |
| 304 | <dt><code>__clang_patchlevel__</code></dt> |
| 305 | <dd>Defined to the patch level of Clang (e.g., the 1 in 2.0.1).</dd> |
| 306 | |
| 307 | <dt><code>__clang_version__</code></dt> |
| 308 | <dd>Defined to a string that captures the Clang version, including |
| 309 | the Subversion tag or revision number, e.g., "1.5 (trunk |
| 310 | 102332)".</dd> |
| 311 | </dl> |
Chris Lattner | 81edc9f | 2009-04-13 02:45:46 +0000 | [diff] [blame] | 312 | |
| 313 | <!-- ======================================================================= --> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 314 | <h2 id="vectors">Vectors and Extended Vectors</h2> |
| 315 | <!-- ======================================================================= --> |
| 316 | |
Owen Anderson | d2bf0cd | 2010-01-27 01:22:36 +0000 | [diff] [blame] | 317 | <p>Supports the GCC vector extensions, plus some stuff like V[1].</p> |
| 318 | |
| 319 | <p>Also supports <tt>ext_vector</tt>, which additionally support for V.xyzw |
| 320 | syntax and other tidbits as seen in OpenCL. An example is:</p> |
| 321 | |
| 322 | <blockquote> |
| 323 | <pre> |
| 324 | typedef float float4 <b>__attribute__((ext_vector_type(4)))</b>; |
| 325 | typedef float float2 <b>__attribute__((ext_vector_type(2)))</b>; |
| 326 | |
| 327 | float4 foo(float2 a, float2 b) { |
| 328 | float4 c; |
| 329 | c.xz = a; |
| 330 | c.yw = b; |
| 331 | return c; |
| 332 | } |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 333 | </pre> |
Owen Anderson | d2bf0cd | 2010-01-27 01:22:36 +0000 | [diff] [blame] | 334 | </blockquote> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 335 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 336 | <p>Query for this feature with __has_extension(attribute_ext_vector_type).</p> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 337 | |
Owen Anderson | d2bf0cd | 2010-01-27 01:22:36 +0000 | [diff] [blame] | 338 | <p>See also <a href="#__builtin_shufflevector">__builtin_shufflevector</a>.</p> |
| 339 | |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 340 | <!-- ======================================================================= --> |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 341 | <h2 id="deprecated">Messages on <tt>deprecated</tt> and <tt>unavailable</tt> Attributes</h2> |
Fariborz Jahanian | c784dc1 | 2010-10-06 23:12:32 +0000 | [diff] [blame] | 342 | <!-- ======================================================================= --> |
| 343 | |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 344 | <p>An optional string message can be added to the <tt>deprecated</tt> |
| 345 | and <tt>unavailable</tt> attributes. For example:</p> |
Fariborz Jahanian | c784dc1 | 2010-10-06 23:12:32 +0000 | [diff] [blame] | 346 | |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 347 | <blockquote> |
Chris Lattner | 4836d6a | 2010-11-09 19:43:35 +0000 | [diff] [blame] | 348 | <pre>void explode(void) __attribute__((deprecated("extremely unsafe, use 'combust' instead!!!")));</pre> |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 349 | </blockquote> |
| 350 | |
| 351 | <p>If the deprecated or unavailable declaration is used, the message |
| 352 | will be incorporated into the appropriate diagnostic:</p> |
| 353 | |
| 354 | <blockquote> |
Chris Lattner | 4836d6a | 2010-11-09 19:43:35 +0000 | [diff] [blame] | 355 | <pre>harmless.c:4:3: warning: 'explode' is deprecated: extremely unsafe, use 'combust' instead!!! [-Wdeprecated-declarations] |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 356 | explode(); |
| 357 | ^</pre> |
| 358 | </blockquote> |
| 359 | |
| 360 | <p>Query for this feature |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 361 | with <tt>__has_extension(attribute_deprecated_with_message)</tt> |
| 362 | and <tt>__has_extension(attribute_unavailable_with_message)</tt>.</p> |
John McCall | 4820908 | 2010-11-08 19:48:17 +0000 | [diff] [blame] | 363 | |
| 364 | <!-- ======================================================================= --> |
| 365 | <h2 id="attributes-on-enumerators">Attributes on Enumerators</h2> |
| 366 | <!-- ======================================================================= --> |
| 367 | |
| 368 | <p>Clang allows attributes to be written on individual enumerators. |
| 369 | This allows enumerators to be deprecated, made unavailable, etc. The |
| 370 | attribute must appear after the enumerator name and before any |
| 371 | initializer, like so:</p> |
| 372 | |
| 373 | <blockquote> |
| 374 | <pre>enum OperationMode { |
| 375 | OM_Invalid, |
| 376 | OM_Normal, |
| 377 | OM_Terrified __attribute__((deprecated)), |
| 378 | OM_AbortOnError __attribute__((deprecated)) = 4 |
| 379 | };</pre> |
| 380 | </blockquote> |
| 381 | |
| 382 | <p>Attributes on the <tt>enum</tt> declaration do not apply to |
| 383 | individual enumerators.</p> |
| 384 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 385 | <p>Query for this feature with <tt>__has_extension(enumerator_attributes)</tt>.</p> |
Fariborz Jahanian | c784dc1 | 2010-10-06 23:12:32 +0000 | [diff] [blame] | 386 | |
| 387 | <!-- ======================================================================= --> |
Ted Kremenek | 87774fd | 2009-12-03 02:04:01 +0000 | [diff] [blame] | 388 | <h2 id="checking_language_features">Checks for Standard Language Features</h2> |
| 389 | <!-- ======================================================================= --> |
| 390 | |
| 391 | <p>The <tt>__has_feature</tt> macro can be used to query if certain standard language features are |
| 392 | enabled. Those features are listed here.</p> |
| 393 | |
Ted Kremenek | 22c3410 | 2009-12-03 02:05:57 +0000 | [diff] [blame] | 394 | <h3 id="cxx_exceptions">C++ exceptions</h3> |
Ted Kremenek | 87774fd | 2009-12-03 02:04:01 +0000 | [diff] [blame] | 395 | |
Ted Kremenek | 22c3410 | 2009-12-03 02:05:57 +0000 | [diff] [blame] | 396 | <p>Use <tt>__has_feature(cxx_exceptions)</tt> to determine if C++ exceptions have been enabled. For |
Sean Hunt | 647ba1b | 2011-06-23 00:42:53 +0000 | [diff] [blame] | 397 | example, compiling code with <tt>-fexceptions</tt> enables C++ exceptions.</p> |
Ted Kremenek | 87774fd | 2009-12-03 02:04:01 +0000 | [diff] [blame] | 398 | |
Ted Kremenek | 22c3410 | 2009-12-03 02:05:57 +0000 | [diff] [blame] | 399 | <h3 id="cxx_rtti">C++ RTTI</h3> |
Ted Kremenek | 87774fd | 2009-12-03 02:04:01 +0000 | [diff] [blame] | 400 | |
Ted Kremenek | 0eb9560 | 2009-12-03 02:06:43 +0000 | [diff] [blame] | 401 | <p>Use <tt>__has_feature(cxx_rtti)</tt> to determine if C++ RTTI has been enabled. For example, |
Ted Kremenek | 22c3410 | 2009-12-03 02:05:57 +0000 | [diff] [blame] | 402 | compiling code with <tt>-fno-rtti</tt> disables the use of RTTI.</p> |
Ted Kremenek | 87774fd | 2009-12-03 02:04:01 +0000 | [diff] [blame] | 403 | |
| 404 | <!-- ======================================================================= --> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 405 | <h2 id="checking_upcoming_features">Checks for Upcoming Standard Language Features</h2> |
| 406 | <!-- ======================================================================= --> |
| 407 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 408 | <p>The <tt>__has_feature</tt> or <tt>__has_extension</tt> macros can be used |
| 409 | to query if certain upcoming standard language features are enabled. Those |
| 410 | features are listed here. Features that are not yet implemented will be |
| 411 | noted.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 412 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 413 | <h3 id="cxx0x">C++0x</h3> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 414 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 415 | <p>The features listed below are slated for inclusion in the upcoming |
| 416 | C++0x standard. As a result, all these features are enabled |
| 417 | with the <tt>-std=c++0x</tt> option when compiling C++ code.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 418 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 419 | <h4 id="cxx_decltype">C++0x <tt>decltype()</tt></h4> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 420 | |
| 421 | <p>Use <tt>__has_feature(cxx_decltype)</tt> or |
| 422 | <tt>__has_extension(cxx_decltype)</tt> to determine if support for the |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 423 | <tt>decltype()</tt> specifier is enabled.</p> |
| 424 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 425 | <h4 id="cxx_access_control_sfinae">C++0x SFINAE includes access control</h4> |
Douglas Gregor | 7822ee3 | 2011-05-11 23:45:11 +0000 | [diff] [blame] | 426 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 427 | <p>Use <tt>__has_feature(cxx_access_control_sfinae)</tt> or <tt>__has_extension(cxx_access_control_sfinae)</tt> to determine whether access-control errors (e.g., calling a private constructor) are considered to be template argument deduction errors (aka SFINAE errors), per <a href="http://www.open-std.org/jtc1/sc22/wg21/docs/cwg_defects.html#1170">C++ DR1170</a>.</p> |
Douglas Gregor | 7822ee3 | 2011-05-11 23:45:11 +0000 | [diff] [blame] | 428 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 429 | <h4 id="cxx_alias_templates">C++0x alias templates</h4> |
Richard Smith | 3e4c6c4 | 2011-05-05 21:57:07 +0000 | [diff] [blame] | 430 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 431 | <p>Use <tt>__has_feature(cxx_alias_templates)</tt> or |
| 432 | <tt>__has_extension(cxx_alias_templates)</tt> to determine if support for |
Richard Smith | 3e4c6c4 | 2011-05-05 21:57:07 +0000 | [diff] [blame] | 433 | C++0x's alias declarations and alias templates is enabled.</p> |
| 434 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 435 | <h4 id="cxx_attributes">C++0x attributes</h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 436 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 437 | <p>Use <tt>__has_feature(cxx_attributes)</tt> or |
| 438 | <tt>__has_extension(cxx_attributes)</tt> to determine if support for attribute |
| 439 | parsing with C++0x's square bracket notation is enabled.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 440 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 441 | <h4 id="cxx_default_function_template_args">C++0x default template arguments in function templates</h4> |
Douglas Gregor | 0750800 | 2011-02-05 20:35:30 +0000 | [diff] [blame] | 442 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 443 | <p>Use <tt>__has_feature(cxx_default_function_template_args)</tt> or |
| 444 | <tt>__has_extension(cxx_default_function_template_args)</tt> to determine |
| 445 | if support for default template arguments in function templates is enabled.</p> |
Douglas Gregor | 0750800 | 2011-02-05 20:35:30 +0000 | [diff] [blame] | 446 | |
Sean Hunt | d962499 | 2011-06-23 06:11:37 +0000 | [diff] [blame] | 447 | <h4 id="cxx_delegating_constructors">C++0x delegating constructors</h4> |
| 448 | |
| 449 | <p>Use <tt>__has_feature(cxx_delegating_constructors)</tt> to determine if |
| 450 | support for delegating constructors is enabled.</p> |
| 451 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 452 | <h4 id="cxx_deleted_functions">C++0x <tt>delete</tt>d functions</h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 453 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 454 | <p>Use <tt>__has_feature(cxx_deleted_functions)</tt> or |
| 455 | <tt>__has_extension(cxx_deleted_functions)</tt> to determine if support for |
Sebastian Redl | f6c0977 | 2010-08-31 23:28:47 +0000 | [diff] [blame] | 456 | deleted function definitions (with <tt>= delete</tt>) is enabled.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 457 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 458 | <h4 id="cxx_lambdas">C++0x lambdas</h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 459 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 460 | <p>Use <tt>__has_feature(cxx_lambdas)</tt> or |
| 461 | <tt>__has_extension(cxx_lambdas)</tt> to determine if support for lambdas |
| 462 | is enabled. clang does not currently implement this feature.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 463 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 464 | <h4 id="cxx_nullptr">C++0x <tt>nullptr</tt></h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 465 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 466 | <p>Use <tt>__has_feature(cxx_nullptr)</tt> or |
| 467 | <tt>__has_extension(cxx_nullptr)</tt> to determine if support for |
Douglas Gregor | 84ee2ee | 2011-05-21 23:15:46 +0000 | [diff] [blame] | 468 | <tt>nullptr</tt> is enabled.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 469 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 470 | <h4 id="cxx_override_control">C++0x <tt>override control</tt></h4> |
Anders Carlsson | c8b9f79 | 2011-03-25 15:04:23 +0000 | [diff] [blame] | 471 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 472 | <p>Use <tt>__has_feature(cxx_override_control)</tt> or |
| 473 | <tt>__has_extension(cxx_override_control)</tt> to determine if support for |
Anders Carlsson | c8b9f79 | 2011-03-25 15:04:23 +0000 | [diff] [blame] | 474 | the override control keywords is enabled.</p> |
| 475 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 476 | <h4 id="cxx_reference_qualified_functions">C++0x reference-qualified functions</h4> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 477 | <p>Use <tt>__has_feature(cxx_reference_qualified_functions)</tt> or |
| 478 | <tt>__has_extension(cxx_reference_qualified_functions)</tt> to determine |
| 479 | if support for reference-qualified functions (e.g., member functions with |
| 480 | <code>&</code> or <code>&&</code> applied to <code>*this</code>) |
| 481 | is enabled.</p> |
Douglas Gregor | 56209ff | 2011-01-26 21:25:54 +0000 | [diff] [blame] | 482 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 483 | <h4 id="cxx_range_for">C++0x range-based <tt>for</tt> loop</h4> |
Richard Smith | a391a46 | 2011-04-15 15:14:40 +0000 | [diff] [blame] | 484 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 485 | <p>Use <tt>__has_feature(cxx_range_for)</tt> or |
| 486 | <tt>__has_extension(cxx_range_for)</tt> to determine if support for the |
| 487 | range-based for loop is enabled. </p> |
Richard Smith | a391a46 | 2011-04-15 15:14:40 +0000 | [diff] [blame] | 488 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 489 | <h4 id="cxx_rvalue_references">C++0x rvalue references</h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 490 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 491 | <p>Use <tt>__has_feature(cxx_rvalue_references)</tt> or |
| 492 | <tt>__has_extension(cxx_rvalue_references)</tt> to determine if support for |
Douglas Gregor | 56209ff | 2011-01-26 21:25:54 +0000 | [diff] [blame] | 493 | rvalue references is enabled. </p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 494 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 495 | <h4 id="cxx_static_assert">C++0x <tt>static_assert()</tt></h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 496 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 497 | <p>Use <tt>__has_feature(cxx_static_assert)</tt> or |
| 498 | <tt>__has_extension(cxx_static_assert)</tt> to determine if support for |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 499 | compile-time assertions using <tt>static_assert</tt> is enabled.</p> |
| 500 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 501 | <h4 id="cxx_auto_type">C++0x type inference</h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 502 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 503 | <p>Use <tt>__has_feature(cxx_auto_type)</tt> or |
| 504 | <tt>__has_extension(cxx_auto_type)</tt> to determine C++0x type inference is |
| 505 | supported using the <tt>auto</tt> specifier. If this is disabled, <tt>auto</tt> |
| 506 | will instead be a storage class specifier, as in C or C++98.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 507 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 508 | <h4 id="cxx_variadic_templates">C++0x variadic templates</h4> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 509 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 510 | <p>Use <tt>__has_feature(cxx_variadic_templates)</tt> or |
| 511 | <tt>__has_extension(cxx_variadic_templates)</tt> to determine if support |
Douglas Gregor | 83d7781 | 2011-01-19 23:15:20 +0000 | [diff] [blame] | 512 | for variadic templates is enabled.</p> |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 513 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 514 | <h4 id="cxx_inline_namespaces">C++0x inline namespaces</h4> |
Sebastian Redl | f6c0977 | 2010-08-31 23:28:47 +0000 | [diff] [blame] | 515 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 516 | <p>Use <tt>__has_feature(cxx_inline_namespaces)</tt> or |
| 517 | <tt>__has_extension(cxx_inline_namespaces)</tt> to determine if support for |
Sebastian Redl | f6c0977 | 2010-08-31 23:28:47 +0000 | [diff] [blame] | 518 | inline namespaces is enabled.</p> |
| 519 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 520 | <h4 id="cxx_trailing_return">C++0x trailing return type</h4> |
Douglas Gregor | dab60ad | 2010-10-01 18:44:50 +0000 | [diff] [blame] | 521 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 522 | <p>Use <tt>__has_feature(cxx_trailing_return)</tt> or |
| 523 | <tt>__has_extension(cxx_trailing_return)</tt> to determine if support for the |
| 524 | alternate function declaration syntax with trailing return type is enabled.</p> |
Douglas Gregor | dab60ad | 2010-10-01 18:44:50 +0000 | [diff] [blame] | 525 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 526 | <h4 id="cxx_noexcept">C++0x noexcept</h4> |
Sebastian Redl | 4561ecd | 2011-03-15 21:17:12 +0000 | [diff] [blame] | 527 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 528 | <p>Use <tt>__has_feature(cxx_noexcept)</tt> or |
| 529 | <tt>__has_extension(cxx_noexcept)</tt> to determine if support for noexcept |
| 530 | exception specifications is enabled.</p> |
Sebastian Redl | 4561ecd | 2011-03-15 21:17:12 +0000 | [diff] [blame] | 531 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 532 | <h4 id="cxx_strong_enums">C++0x strongly typed enumerations</h4> |
Douglas Gregor | 1274ccd | 2010-10-08 23:50:27 +0000 | [diff] [blame] | 533 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 534 | <p>Use <tt>__has_feature(cxx_strong_enums)</tt> or |
| 535 | <tt>__has_extension(cxx_strong_enums)</tt> to determine if support for |
Douglas Gregor | 1274ccd | 2010-10-08 23:50:27 +0000 | [diff] [blame] | 536 | strongly typed, scoped enumerations is enabled.</p> |
| 537 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 538 | <h3 id="c1x">C1X</h3> |
| 539 | |
| 540 | <p>The features listed below are slated for inclusion in the upcoming |
| 541 | C1X standard. As a result, all these features are enabled |
| 542 | with the <tt>-std=c1x</tt> option when compiling C code.</p> |
| 543 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 544 | <h4 id="c_generic_selections">C1X generic selections</h4> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 545 | |
| 546 | <p>Use <tt>__has_feature(c_generic_selections)</tt> or |
| 547 | <tt>__has_extension(c_generic_selections)</tt> to determine if support for |
| 548 | generic selections is enabled.</p> |
| 549 | |
| 550 | <p>As an extension, the C1X generic selection expression is available in all |
| 551 | languages supported by Clang. The syntax is the same as that given in the |
| 552 | C1X draft standard.</p> |
| 553 | |
| 554 | <p>In C, type compatibility is decided according to the rules given in the |
| 555 | appropriate standard, but in C++, which lacks the type compatibility rules |
| 556 | used in C, types are considered compatible only if they are equivalent.</p> |
| 557 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 558 | <h4 id="c_static_assert">C1X <tt>_Static_assert()</tt></h4> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 559 | |
| 560 | <p>Use <tt>__has_feature(c_static_assert)</tt> or |
| 561 | <tt>__has_extension(c_static_assert)</tt> to determine if support for |
| 562 | compile-time assertions using <tt>_Static_assert</tt> is enabled.</p> |
| 563 | |
Sean Hunt | 4ef4c6b | 2010-01-13 08:31:49 +0000 | [diff] [blame] | 564 | <!-- ======================================================================= --> |
Douglas Gregor | afdf137 | 2011-02-03 21:57:35 +0000 | [diff] [blame] | 565 | <h2 id="checking_type_traits">Checks for Type Traits</h2> |
| 566 | <!-- ======================================================================= --> |
| 567 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 568 | <p>Clang supports the <a href="http://gcc.gnu.org/onlinedocs/gcc/Type-Traits.html">GNU C++ type traits</a> and a subset of the <a href="http://msdn.microsoft.com/en-us/library/ms177194(v=VS.100).aspx">Microsoft Visual C++ Type traits</a>. For each supported type trait <code>__X</code>, <code>__has_extension(X)</code> indicates the presence of the type trait. For example: |
Douglas Gregor | afdf137 | 2011-02-03 21:57:35 +0000 | [diff] [blame] | 569 | <blockquote> |
| 570 | <pre> |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 571 | #if __has_extension(is_convertible_to) |
Douglas Gregor | afdf137 | 2011-02-03 21:57:35 +0000 | [diff] [blame] | 572 | template<typename From, typename To> |
| 573 | struct is_convertible_to { |
| 574 | static const bool value = __is_convertible_to(From, To); |
| 575 | }; |
| 576 | #else |
| 577 | // Emulate type trait |
| 578 | #endif |
| 579 | </pre> |
| 580 | </blockquote> |
| 581 | |
| 582 | <p>The following type traits are supported by Clang:</p> |
| 583 | <ul> |
| 584 | <li><code>__has_nothrow_assign</code> (GNU, Microsoft)</li> |
| 585 | <li><code>__has_nothrow_copy</code> (GNU, Microsoft)</li> |
| 586 | <li><code>__has_nothrow_constructor</code> (GNU, Microsoft)</li> |
| 587 | <li><code>__has_trivial_assign</code> (GNU, Microsoft)</li> |
| 588 | <li><code>__has_trivial_copy</code> (GNU, Microsoft)</li> |
| 589 | <li><code>__has_trivial_constructor</code> (GNU, Microsoft)</li> |
| 590 | <li><code>__has_trivial_destructor</code> (GNU, Microsoft)</li> |
| 591 | <li><code>__has_virtual_destructor</code> (GNU, Microsoft)</li> |
| 592 | <li><code>__is_abstract</code> (GNU, Microsoft)</li> |
| 593 | <li><code>__is_base_of</code> (GNU, Microsoft)</li> |
| 594 | <li><code>__is_class</code> (GNU, Microsoft)</li> |
| 595 | <li><code>__is_convertible_to</code> (Microsoft)</li> |
| 596 | <li><code>__is_empty</code> (GNU, Microsoft)</li> |
| 597 | <li><code>__is_enum</code> (GNU, Microsoft)</li> |
| 598 | <li><code>__is_pod</code> (GNU, Microsoft)</li> |
| 599 | <li><code>__is_polymorphic</code> (GNU, Microsoft)</li> |
| 600 | <li><code>__is_union</code> (GNU, Microsoft)</li> |
| 601 | <li><code>__is_literal(type)</code>: Determines whether the given type is a literal type</li> |
Sean Hunt | 1fba828 | 2011-07-18 17:22:33 +0000 | [diff] [blame] | 602 | <li><code>__underlying_type(type)</code>: Retrieves the underlying type for a given <code>enum</code> type. This trait is required to implement the C++0x standard library.</li> |
Douglas Gregor | afdf137 | 2011-02-03 21:57:35 +0000 | [diff] [blame] | 603 | </ul> |
| 604 | |
| 605 | <!-- ======================================================================= --> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 606 | <h2 id="blocks">Blocks</h2> |
| 607 | <!-- ======================================================================= --> |
| 608 | |
Chris Lattner | a7dbdf5 | 2009-03-09 07:03:22 +0000 | [diff] [blame] | 609 | <p>The syntax and high level language feature description is in <a |
| 610 | href="BlockLanguageSpec.txt">BlockLanguageSpec.txt</a>. Implementation and ABI |
| 611 | details for the clang implementation are in <a |
Chris Lattner | 5d7650b | 2010-03-16 21:43:03 +0000 | [diff] [blame] | 612 | href="Block-ABI-Apple.txt">Block-ABI-Apple.txt</a>.</p> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 613 | |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 614 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 615 | <p>Query for this feature with __has_extension(blocks).</p> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 616 | |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 617 | <!-- ======================================================================= --> |
Douglas Gregor | 926df6c | 2011-06-11 01:09:30 +0000 | [diff] [blame] | 618 | <h2 id="objc_features">Objective-C Features</h2> |
| 619 | <!-- ======================================================================= --> |
| 620 | |
| 621 | <h3 id="objc_instancetype">Related result types</h3> |
| 622 | |
| 623 | <p>According to Cocoa conventions, Objective-C methods with certain names ("init", "alloc", etc.) always return objects that are an instance of the receiving class's type. Such methods are said to have a "related result type", meaning that a message send to one of these methods will have the same static type as an instance of the receiver class. For example, given the following classes:</p> |
| 624 | |
| 625 | <blockquote> |
| 626 | <pre> |
| 627 | @interface NSObject |
| 628 | + (id)alloc; |
| 629 | - (id)init; |
| 630 | @end |
| 631 | |
| 632 | @interface NSArray : NSObject |
| 633 | @end |
| 634 | </pre> |
| 635 | </blockquote> |
| 636 | |
| 637 | <p>and this common initialization pattern</p> |
| 638 | |
| 639 | <blockquote> |
| 640 | <pre> |
| 641 | NSArray *array = [[NSArray alloc] init]; |
| 642 | </pre> |
| 643 | </blockquote> |
| 644 | |
| 645 | <p>the type of the expression <code>[NSArray alloc]</code> is |
| 646 | <code>NSArray*</code> because <code>alloc</code> implicitly has a |
| 647 | related result type. Similarly, the type of the expression |
| 648 | <code>[[NSArray alloc] init]</code> is <code>NSArray*</code>, since |
| 649 | <code>init</code> has a related result type and its receiver is known |
| 650 | to have the type <code>NSArray *</code>. If neither <code>alloc</code> nor <code>init</code> had a related result type, the expressions would have had type <code>id</code>, as declared in the method signature.</p> |
| 651 | |
| 652 | <p>To determine whether a method has a related result type, the first |
| 653 | word in the camel-case selector (e.g., "init" in "initWithObjects") is |
| 654 | considered, and the method will a related result type if its return |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 655 | type is compatible with the type of its class and if</p> |
Douglas Gregor | 926df6c | 2011-06-11 01:09:30 +0000 | [diff] [blame] | 656 | |
| 657 | <ul> |
| 658 | |
| 659 | <li>the first word is "alloc" or "new", and the method is a class |
| 660 | method, or</li> |
| 661 | |
| 662 | <li>the first word is "autorelease", "init", "retain", or "self", |
| 663 | and the method is an instance method.</li> |
| 664 | |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 665 | </ul> |
Douglas Gregor | 926df6c | 2011-06-11 01:09:30 +0000 | [diff] [blame] | 666 | |
| 667 | <p>If a method with a related result type is overridden by a subclass |
| 668 | method, the subclass method must also return a type that is compatible |
| 669 | with the subclass type. For example:</p> |
| 670 | |
| 671 | <blockquote> |
| 672 | <pre> |
| 673 | @interface NSString : NSObject |
| 674 | - (NSUnrelated *)init; // incorrect usage: NSUnrelated is not NSString or a superclass of NSString |
| 675 | @end |
| 676 | </pre> |
| 677 | </blockquote> |
| 678 | |
| 679 | <p>Related result types only affect the type of a message send or |
| 680 | property access via the given method. In all other respects, a method |
| 681 | with a related result type is treated the same way as method without a |
| 682 | related result type.</p> |
| 683 | |
| 684 | <!-- ======================================================================= --> |
John McCall | f85e193 | 2011-06-15 23:02:42 +0000 | [diff] [blame] | 685 | <h2 id="objc_arc">Automatic reference counting </h2> |
| 686 | <!-- ======================================================================= --> |
| 687 | |
| 688 | <p>Clang provides support for <a href="AutomaticReferenceCounting.html">automated reference counting</a> in Objective-C, which eliminates the need for manual retain/release/autorelease message sends. There are two feature macros associated with automatic reference counting: <code>__has_feature(objc_arc)</code> indicates the availability of automated reference counting in general, while <code>__has_feature(objc_arc_weak)</code> indicates that automated reference counting also includes support for <code>__weak</code> pointers to Objective-C objects.</p> |
| 689 | |
| 690 | <!-- ======================================================================= --> |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 691 | <h2 id="overloading-in-c">Function Overloading in C</h2> |
| 692 | <!-- ======================================================================= --> |
| 693 | |
Chris Lattner | f161d41 | 2009-02-13 21:51:45 +0000 | [diff] [blame] | 694 | <p>Clang provides support for C++ function overloading in C. Function |
| 695 | overloading in C is introduced using the <tt>overloadable</tt> attribute. For |
| 696 | example, one might provide several overloaded versions of a <tt>tgsin</tt> |
| 697 | function that invokes the appropriate standard function computing the sine of a |
| 698 | value with <tt>float</tt>, <tt>double</tt>, or <tt>long double</tt> |
| 699 | precision:</p> |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 700 | |
| 701 | <blockquote> |
| 702 | <pre> |
| 703 | #include <math.h> |
| 704 | float <b>__attribute__((overloadable))</b> tgsin(float x) { return sinf(x); } |
| 705 | double <b>__attribute__((overloadable))</b> tgsin(double x) { return sin(x); } |
| 706 | long double <b>__attribute__((overloadable))</b> tgsin(long double x) { return sinl(x); } |
| 707 | </pre> |
| 708 | </blockquote> |
| 709 | |
| 710 | <p>Given these declarations, one can call <tt>tgsin</tt> with a |
| 711 | <tt>float</tt> value to receive a <tt>float</tt> result, with a |
| 712 | <tt>double</tt> to receive a <tt>double</tt> result, etc. Function |
| 713 | overloading in C follows the rules of C++ function overloading to pick |
| 714 | the best overload given the call arguments, with a few C-specific |
| 715 | semantics:</p> |
| 716 | <ul> |
| 717 | <li>Conversion from <tt>float</tt> or <tt>double</tt> to <tt>long |
| 718 | double</tt> is ranked as a floating-point promotion (per C99) rather |
| 719 | than as a floating-point conversion (as in C++).</li> |
| 720 | |
| 721 | <li>A conversion from a pointer of type <tt>T*</tt> to a pointer of type |
| 722 | <tt>U*</tt> is considered a pointer conversion (with conversion |
| 723 | rank) if <tt>T</tt> and <tt>U</tt> are compatible types.</li> |
| 724 | |
| 725 | <li>A conversion from type <tt>T</tt> to a value of type <tt>U</tt> |
| 726 | is permitted if <tt>T</tt> and <tt>U</tt> are compatible types. This |
| 727 | conversion is given "conversion" rank.</li> |
| 728 | </ul> |
| 729 | |
| 730 | <p>The declaration of <tt>overloadable</tt> functions is restricted to |
| 731 | function declarations and definitions. Most importantly, if any |
| 732 | function with a given name is given the <tt>overloadable</tt> |
| 733 | attribute, then all function declarations and definitions with that |
| 734 | name (and in that scope) must have the <tt>overloadable</tt> |
Chris Lattner | f161d41 | 2009-02-13 21:51:45 +0000 | [diff] [blame] | 735 | attribute. This rule even applies to redeclarations of functions whose original |
| 736 | declaration had the <tt>overloadable</tt> attribute, e.g.,</p> |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 737 | |
| 738 | <blockquote> |
| 739 | <pre> |
| 740 | int f(int) __attribute__((overloadable)); |
| 741 | float f(float); <i>// error: declaration of "f" must have the "overloadable" attribute</i> |
| 742 | |
| 743 | int g(int) __attribute__((overloadable)); |
| 744 | int g(int) { } <i>// error: redeclaration of "g" must also have the "overloadable" attribute</i> |
| 745 | </pre> |
| 746 | </blockquote> |
| 747 | |
Douglas Gregor | 965acbb | 2009-02-18 07:07:28 +0000 | [diff] [blame] | 748 | <p>Functions marked <tt>overloadable</tt> must have |
| 749 | prototypes. Therefore, the following code is ill-formed:</p> |
| 750 | |
| 751 | <blockquote> |
| 752 | <pre> |
| 753 | int h() __attribute__((overloadable)); <i>// error: h does not have a prototype</i> |
| 754 | </pre> |
| 755 | </blockquote> |
| 756 | |
| 757 | <p>However, <tt>overloadable</tt> functions are allowed to use a |
| 758 | ellipsis even if there are no named parameters (as is permitted in C++). This feature is particularly useful when combined with the <tt>unavailable</tt> attribute:</p> |
| 759 | |
| 760 | <blockquote> |
| 761 | <pre> |
Chris Lattner | 0224680 | 2009-02-18 22:27:46 +0000 | [diff] [blame] | 762 | void honeypot(...) __attribute__((overloadable, unavailable)); <i>// calling me is an error</i> |
Douglas Gregor | 965acbb | 2009-02-18 07:07:28 +0000 | [diff] [blame] | 763 | </pre> |
| 764 | </blockquote> |
| 765 | |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 766 | <p>Functions declared with the <tt>overloadable</tt> attribute have |
| 767 | their names mangled according to the same rules as C++ function |
| 768 | names. For example, the three <tt>tgsin</tt> functions in our |
| 769 | motivating example get the mangled names <tt>_Z5tgsinf</tt>, |
Chris Lattner | 71b48d6 | 2010-11-28 18:19:13 +0000 | [diff] [blame] | 770 | <tt>_Z5tgsind</tt>, and <tt>_Z5tgsine</tt>, respectively. There are two |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 771 | caveats to this use of name mangling:</p> |
| 772 | |
| 773 | <ul> |
| 774 | |
| 775 | <li>Future versions of Clang may change the name mangling of |
| 776 | functions overloaded in C, so you should not depend on an specific |
| 777 | mangling. To be completely safe, we strongly urge the use of |
| 778 | <tt>static inline</tt> with <tt>overloadable</tt> functions.</li> |
| 779 | |
| 780 | <li>The <tt>overloadable</tt> attribute has almost no meaning when |
| 781 | used in C++, because names will already be mangled and functions are |
| 782 | already overloadable. However, when an <tt>overloadable</tt> |
| 783 | function occurs within an <tt>extern "C"</tt> linkage specification, |
| 784 | it's name <i>will</i> be mangled in the same way as it would in |
| 785 | C.</li> |
| 786 | </ul> |
| 787 | |
Peter Collingbourne | c1b5fa4 | 2011-05-13 20:54:45 +0000 | [diff] [blame] | 788 | <p>Query for this feature with __has_extension(attribute_overloadable).</p> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 789 | |
| 790 | |
Douglas Gregor | cb54d43 | 2009-02-13 00:57:04 +0000 | [diff] [blame] | 791 | <!-- ======================================================================= --> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 792 | <h2 id="builtins">Builtin Functions</h2> |
| 793 | <!-- ======================================================================= --> |
| 794 | |
| 795 | <p>Clang supports a number of builtin library functions with the same syntax as |
| 796 | GCC, including things like <tt>__builtin_nan</tt>, |
| 797 | <tt>__builtin_constant_p</tt>, <tt>__builtin_choose_expr</tt>, |
| 798 | <tt>__builtin_types_compatible_p</tt>, <tt>__sync_fetch_and_add</tt>, etc. In |
| 799 | addition to the GCC builtins, Clang supports a number of builtins that GCC does |
| 800 | not, which are listed here.</p> |
| 801 | |
| 802 | <p>Please note that Clang does not and will not support all of the GCC builtins |
| 803 | for vector operations. Instead of using builtins, you should use the functions |
| 804 | defined in target-specific header files like <tt><xmmintrin.h></tt>, which |
| 805 | define portable wrappers for these. Many of the Clang versions of these |
| 806 | functions are implemented directly in terms of <a href="#vectors">extended |
| 807 | vector support</a> instead of builtins, in order to reduce the number of |
| 808 | builtins that we need to implement.</p> |
| 809 | |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 810 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 811 | <h3><a name="__builtin_shufflevector">__builtin_shufflevector</a></h3> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 812 | <!-- ======================================================================= --> |
| 813 | |
Chris Lattner | aad826b | 2009-09-16 18:56:12 +0000 | [diff] [blame] | 814 | <p><tt>__builtin_shufflevector</tt> is used to express generic vector |
Chris Lattner | 6f72da5 | 2009-02-13 20:00:20 +0000 | [diff] [blame] | 815 | permutation/shuffle/swizzle operations. This builtin is also very important for |
| 816 | the implementation of various target-specific header files like |
| 817 | <tt><xmmintrin.h></tt>. |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 818 | </p> |
| 819 | |
| 820 | <p><b>Syntax:</b></p> |
| 821 | |
| 822 | <pre> |
Chris Lattner | 6f72da5 | 2009-02-13 20:00:20 +0000 | [diff] [blame] | 823 | __builtin_shufflevector(vec1, vec2, index1, index2, ...) |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 824 | </pre> |
| 825 | |
| 826 | <p><b>Examples:</b></p> |
| 827 | |
| 828 | <pre> |
Chris Lattner | 6f72da5 | 2009-02-13 20:00:20 +0000 | [diff] [blame] | 829 | // Identity operation - return 4-element vector V1. |
| 830 | __builtin_shufflevector(V1, V1, 0, 1, 2, 3) |
| 831 | |
| 832 | // "Splat" element 0 of V1 into a 4-element result. |
| 833 | __builtin_shufflevector(V1, V1, 0, 0, 0, 0) |
| 834 | |
| 835 | // Reverse 4-element vector V1. |
| 836 | __builtin_shufflevector(V1, V1, 3, 2, 1, 0) |
| 837 | |
| 838 | // Concatenate every other element of 4-element vectors V1 and V2. |
| 839 | __builtin_shufflevector(V1, V2, 0, 2, 4, 6) |
| 840 | |
| 841 | // Concatenate every other element of 8-element vectors V1 and V2. |
| 842 | __builtin_shufflevector(V1, V2, 0, 2, 4, 6, 8, 10, 12, 14) |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 843 | </pre> |
| 844 | |
| 845 | <p><b>Description:</b></p> |
| 846 | |
Chris Lattner | 6f72da5 | 2009-02-13 20:00:20 +0000 | [diff] [blame] | 847 | <p>The first two arguments to __builtin_shufflevector are vectors that have the |
| 848 | same element type. The remaining arguments are a list of integers that specify |
| 849 | the elements indices of the first two vectors that should be extracted and |
| 850 | returned in a new vector. These element indices are numbered sequentially |
| 851 | starting with the first vector, continuing into the second vector. Thus, if |
| 852 | vec1 is a 4-element vector, index 5 would refer to the second element of vec2. |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 853 | </p> |
| 854 | |
Chris Lattner | 6f72da5 | 2009-02-13 20:00:20 +0000 | [diff] [blame] | 855 | <p>The result of __builtin_shufflevector is a vector |
| 856 | with the same element type as vec1/vec2 but that has an element count equal to |
| 857 | the number of indices specified. |
| 858 | </p> |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 859 | |
Chris Lattner | 21190d5 | 2009-09-21 03:09:59 +0000 | [diff] [blame] | 860 | <p>Query for this feature with __has_builtin(__builtin_shufflevector).</p> |
| 861 | |
| 862 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 863 | <h3><a name="__builtin_unreachable">__builtin_unreachable</a></h3> |
Chris Lattner | 21190d5 | 2009-09-21 03:09:59 +0000 | [diff] [blame] | 864 | <!-- ======================================================================= --> |
| 865 | |
| 866 | <p><tt>__builtin_unreachable</tt> is used to indicate that a specific point in |
| 867 | the program cannot be reached, even if the compiler might otherwise think it |
| 868 | can. This is useful to improve optimization and eliminates certain warnings. |
| 869 | For example, without the <tt>__builtin_unreachable</tt> in the example below, |
| 870 | the compiler assumes that the inline asm can fall through and prints a "function |
| 871 | declared 'noreturn' should not return" warning. |
| 872 | </p> |
| 873 | |
| 874 | <p><b>Syntax:</b></p> |
| 875 | |
| 876 | <pre> |
| 877 | __builtin_unreachable() |
| 878 | </pre> |
| 879 | |
| 880 | <p><b>Example of Use:</b></p> |
| 881 | |
| 882 | <pre> |
| 883 | void myabort(void) __attribute__((noreturn)); |
| 884 | void myabort(void) { |
| 885 | asm("int3"); |
| 886 | __builtin_unreachable(); |
| 887 | } |
| 888 | </pre> |
| 889 | |
| 890 | <p><b>Description:</b></p> |
| 891 | |
| 892 | <p>The __builtin_unreachable() builtin has completely undefined behavior. Since |
| 893 | it has undefined behavior, it is a statement that it is never reached and the |
| 894 | optimizer can take advantage of this to produce better code. This builtin takes |
| 895 | no arguments and produces a void result. |
| 896 | </p> |
| 897 | |
| 898 | <p>Query for this feature with __has_builtin(__builtin_unreachable).</p> |
| 899 | |
Chris Lattner | 23aa9c8 | 2011-04-09 03:57:26 +0000 | [diff] [blame] | 900 | <!-- ======================================================================= --> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 901 | <h3><a name="__sync_swap">__sync_swap</a></h3> |
Chris Lattner | 23aa9c8 | 2011-04-09 03:57:26 +0000 | [diff] [blame] | 902 | <!-- ======================================================================= --> |
| 903 | |
| 904 | <p><tt>__sync_swap</tt> is used to atomically swap integers or pointers in |
| 905 | memory. |
| 906 | </p> |
| 907 | |
| 908 | <p><b>Syntax:</b></p> |
| 909 | |
| 910 | <pre> |
| 911 | <i>type</i> __sync_swap(<i>type</i> *ptr, <i>type</i> value, ...) |
| 912 | </pre> |
| 913 | |
| 914 | <p><b>Example of Use:</b></p> |
| 915 | |
| 916 | <pre> |
Sean Hunt | 7e98b47 | 2011-06-23 01:21:01 +0000 | [diff] [blame] | 917 | int old_value = __sync_swap(&value, new_value); |
Chris Lattner | 23aa9c8 | 2011-04-09 03:57:26 +0000 | [diff] [blame] | 918 | </pre> |
| 919 | |
| 920 | <p><b>Description:</b></p> |
| 921 | |
| 922 | <p>The __sync_swap() builtin extends the existing __sync_*() family of atomic |
| 923 | intrinsics to allow code to atomically swap the current value with the new |
| 924 | value. More importantly, it helps developers write more efficient and correct |
| 925 | code by avoiding expensive loops around __sync_bool_compare_and_swap() or |
| 926 | relying on the platform specific implementation details of |
| 927 | __sync_lock_test_and_set(). The __sync_swap() builtin is a full barrier. |
| 928 | </p> |
| 929 | |
Chris Lattner | 21190d5 | 2009-09-21 03:09:59 +0000 | [diff] [blame] | 930 | |
Chris Lattner | 1177f91 | 2009-04-09 19:58:15 +0000 | [diff] [blame] | 931 | <!-- ======================================================================= --> |
| 932 | <h2 id="targetspecific">Target-Specific Extensions</h2> |
| 933 | <!-- ======================================================================= --> |
| 934 | |
| 935 | <p>Clang supports some language features conditionally on some targets.</p> |
| 936 | |
| 937 | <!-- ======================================================================= --> |
| 938 | <h3 id="x86-specific">X86/X86-64 Language Extensions</h3> |
| 939 | <!-- ======================================================================= --> |
| 940 | |
| 941 | <p>The X86 backend has these language extensions:</p> |
| 942 | |
| 943 | <!-- ======================================================================= --> |
| 944 | <h4 id="x86-gs-segment">Memory references off the GS segment</h4> |
| 945 | <!-- ======================================================================= --> |
| 946 | |
| 947 | <p>Annotating a pointer with address space #256 causes it to be code generated |
Chris Lattner | a021e7c | 2009-05-05 18:54:47 +0000 | [diff] [blame] | 948 | relative to the X86 GS segment register, and address space #257 causes it to be |
| 949 | relative to the X86 FS segment. Note that this is a very very low-level |
| 950 | feature that should only be used if you know what you're doing (for example in |
| 951 | an OS kernel).</p> |
Chris Lattner | 1177f91 | 2009-04-09 19:58:15 +0000 | [diff] [blame] | 952 | |
| 953 | <p>Here is an example:</p> |
| 954 | |
| 955 | <pre> |
| 956 | #define GS_RELATIVE __attribute__((address_space(256))) |
| 957 | int foo(int GS_RELATIVE *P) { |
| 958 | return *P; |
| 959 | } |
| 960 | </pre> |
| 961 | |
| 962 | <p>Which compiles to (on X86-32):</p> |
| 963 | |
| 964 | <pre> |
| 965 | _foo: |
| 966 | movl 4(%esp), %eax |
| 967 | movl %gs:(%eax), %eax |
| 968 | ret |
| 969 | </pre> |
| 970 | |
Ted Kremenek | ed86931 | 2009-04-10 05:03:33 +0000 | [diff] [blame] | 971 | <!-- ======================================================================= --> |
| 972 | <h2 id="analyzerspecific">Static Analysis-Specific Extensions</h2> |
| 973 | <!-- ======================================================================= --> |
| 974 | |
| 975 | <p>Clang supports additional attributes that are useful for documenting program |
| 976 | invariants and rules for static analysis tools. The extensions documented here |
| 977 | are used by the <a |
| 978 | href="http://clang.llvm.org/StaticAnalysis.html">path-sensitive static analyzer |
| 979 | engine</a> that is part of Clang's Analysis library.</p> |
| 980 | |
John McCall | 8749401 | 2011-03-18 03:51:49 +0000 | [diff] [blame] | 981 | <h3 id="attr_analyzer_noreturn">The <tt>analyzer_noreturn</tt> attribute</h3> |
Ted Kremenek | ed86931 | 2009-04-10 05:03:33 +0000 | [diff] [blame] | 982 | |
| 983 | <p>Clang's static analysis engine understands the standard <tt>noreturn</tt> |
Ted Kremenek | 4df2114 | 2009-04-10 05:04:22 +0000 | [diff] [blame] | 984 | attribute. This attribute, which is typically affixed to a function prototype, |
| 985 | indicates that a call to a given function never returns. Function prototypes for |
| 986 | common functions like <tt>exit</tt> are typically annotated with this attribute, |
| 987 | as well as a variety of common assertion handlers. Users can educate the static |
| 988 | analyzer about their own custom assertion handles (thus cutting down on false |
| 989 | positives due to false paths) by marking their own "panic" functions |
| 990 | with this attribute.</p> |
Ted Kremenek | ed86931 | 2009-04-10 05:03:33 +0000 | [diff] [blame] | 991 | |
| 992 | <p>While useful, <tt>noreturn</tt> is not applicable in all cases. Sometimes |
Nick Lewycky | 625b586 | 2009-06-14 04:08:08 +0000 | [diff] [blame] | 993 | there are special functions that for all intents and purposes should be |
| 994 | considered panic functions (i.e., they are only called when an internal program |
| 995 | error occurs) but may actually return so that the program can fail gracefully. |
| 996 | The <tt>analyzer_noreturn</tt> attribute allows one to annotate such functions |
| 997 | as being interpreted as "no return" functions by the analyzer (thus |
Chris Lattner | 2893589 | 2009-04-10 05:54:56 +0000 | [diff] [blame] | 998 | pruning bogus paths) but will not affect compilation (as in the case of |
Ted Kremenek | ed86931 | 2009-04-10 05:03:33 +0000 | [diff] [blame] | 999 | <tt>noreturn</tt>).</p> |
| 1000 | |
| 1001 | <p><b>Usage</b>: The <tt>analyzer_noreturn</tt> attribute can be placed in the |
Chris Lattner | 2893589 | 2009-04-10 05:54:56 +0000 | [diff] [blame] | 1002 | same places where the <tt>noreturn</tt> attribute can be placed. It is commonly |
Ted Kremenek | ed86931 | 2009-04-10 05:03:33 +0000 | [diff] [blame] | 1003 | placed at the end of function prototypes:</p> |
| 1004 | |
| 1005 | <pre> |
| 1006 | void foo() <b>__attribute__((analyzer_noreturn))</b>; |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 1007 | </pre> |
| 1008 | |
John McCall | 8749401 | 2011-03-18 03:51:49 +0000 | [diff] [blame] | 1009 | <p>Query for this feature with |
| 1010 | <tt>__has_attribute(analyzer_noreturn)</tt>.</p> |
Chris Lattner | 148772a | 2009-06-13 07:13:28 +0000 | [diff] [blame] | 1011 | |
John McCall | 8749401 | 2011-03-18 03:51:49 +0000 | [diff] [blame] | 1012 | <h3 id="attr_method_family">The <tt>objc_method_family</tt> attribute</h3> |
| 1013 | |
| 1014 | <p>Many methods in Objective-C have conventional meanings determined |
| 1015 | by their selectors. For the purposes of static analysis, it is |
| 1016 | sometimes useful to be able to mark a method as having a particular |
| 1017 | conventional meaning despite not having the right selector, or as not |
| 1018 | having the conventional meaning that its selector would suggest. |
| 1019 | For these use cases, we provide an attribute to specifically describe |
| 1020 | the <q>method family</q> that a method belongs to.</p> |
| 1021 | |
| 1022 | <p><b>Usage</b>: <tt>__attribute__((objc_method_family(X)))</tt>, |
| 1023 | where <tt>X</tt> is one of <tt>none</tt>, <tt>alloc</tt>, <tt>copy</tt>, |
| 1024 | <tt>init</tt>, <tt>mutableCopy</tt>, or <tt>new</tt>. This attribute |
| 1025 | can only be placed at the end of a method declaration:</p> |
| 1026 | |
| 1027 | <pre> |
| 1028 | - (NSString*) initMyStringValue <b>__attribute__((objc_method_family(none)))</b>; |
| 1029 | </pre> |
| 1030 | |
| 1031 | <p>Users who do not wish to change the conventional meaning of a |
| 1032 | method, and who merely want to document its non-standard retain and |
| 1033 | release semantics, should use the |
| 1034 | <a href="#attr_retain_release">retaining behavior attributes</a> |
| 1035 | described below.</p> |
| 1036 | |
| 1037 | <p>Query for this feature with |
| 1038 | <tt>__has_attribute(objc_method_family)</tt>.</p> |
| 1039 | |
| 1040 | <h3 id="attr_retain_release">Objective-C retaining behavior attributes</h3> |
John McCall | 630b7ae | 2011-01-25 04:26:21 +0000 | [diff] [blame] | 1041 | |
| 1042 | <p>In Objective-C, functions and methods are generally assumed to take |
| 1043 | and return objects with +0 retain counts, with some exceptions for |
| 1044 | special methods like <tt>+alloc</tt> and <tt>init</tt>. However, |
| 1045 | there are exceptions, and so Clang provides attributes to allow these |
| 1046 | exceptions to be documented, which helps the analyzer find leaks (and |
John McCall | 8749401 | 2011-03-18 03:51:49 +0000 | [diff] [blame] | 1047 | ignore non-leaks). Some exceptions may be better described using |
| 1048 | the <a href="#attr_method_family"><tt>objc_method_family</tt></a> |
| 1049 | attribute instead.</p> |
John McCall | 630b7ae | 2011-01-25 04:26:21 +0000 | [diff] [blame] | 1050 | |
| 1051 | <p><b>Usage</b>: The <tt>ns_returns_retained</tt>, <tt>ns_returns_not_retained</tt>, |
| 1052 | <tt>ns_returns_autoreleased</tt>, <tt>cf_returns_retained</tt>, |
| 1053 | and <tt>cf_returns_not_retained</tt> attributes can be placed on |
| 1054 | methods and functions that return Objective-C or CoreFoundation |
| 1055 | objects. They are commonly placed at the end of a function prototype |
| 1056 | or method declaration:</p> |
| 1057 | |
| 1058 | <pre> |
| 1059 | id foo() <b>__attribute__((ns_returns_retained))</b>; |
| 1060 | |
| 1061 | - (NSString*) bar: (int) x <b>__attribute__((ns_returns_retained))</b>; |
| 1062 | </pre> |
| 1063 | |
| 1064 | <p>The <tt>*_returns_retained</tt> attributes specify that the |
| 1065 | returned object has a +1 retain count. |
| 1066 | The <tt>*_returns_not_retained</tt> attributes specify that the return |
| 1067 | object has a +0 retain count, even if the normal convention for its |
| 1068 | selector would be +1. <tt>ns_returns_autoreleased</tt> specifies that the |
| 1069 | returned object is +0, but is guaranteed to live at least as long as the |
| 1070 | next flush of an autorelease pool.</p> |
| 1071 | |
| 1072 | <p><b>Usage</b>: The <tt>ns_consumed</tt> and <tt>cf_consumed</tt> |
| 1073 | attributes can be placed on an parameter declaration; they specify |
| 1074 | that the argument is expected to have a +1 retain count, which will be |
| 1075 | balanced in some way by the function or method. |
| 1076 | The <tt>ns_consumes_self</tt> attribute can only be placed on an |
| 1077 | Objective-C method; it specifies that the method expects |
| 1078 | its <tt>self</tt> parameter to have a +1 retain count, which it will |
| 1079 | balance in some way.</p> |
| 1080 | |
| 1081 | <pre> |
| 1082 | void <b>foo(__attribute__((ns_consumed))</b> NSString *string); |
| 1083 | |
| 1084 | - (void) bar <b>__attribute__((ns_consumes_self))</b>; |
| 1085 | - (void) baz: (id) <b>__attribute__((ns_consumed))</b> x; |
| 1086 | </pre> |
Ted Kremenek | ed86931 | 2009-04-10 05:03:33 +0000 | [diff] [blame] | 1087 | |
John McCall | 8749401 | 2011-03-18 03:51:49 +0000 | [diff] [blame] | 1088 | <p>Query for these features with <tt>__has_attribute(ns_consumed)</tt>, |
| 1089 | <tt>__has_attribute(ns_returns_retained)</tt>, etc.</p> |
| 1090 | |
Chris Lattner | 5ce933f | 2009-02-09 08:46:11 +0000 | [diff] [blame] | 1091 | </div> |
| 1092 | </body> |
| 1093 | </html> |