blob: 40145a4739632b23b17ca1de677bab472ab8f4d8 [file] [log] [blame]
Sean Hunt7e98b472011-06-23 01:21:01 +00001<!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 Lattner5ce933f2009-02-09 08:46:11 +00004<html>
5<head>
Sean Hunt7e98b472011-06-23 01:21:01 +00006 <META http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
Eli Friedman0c706c22011-09-19 23:17:44 +00007 <title>Clang Language Extensions</title>
Sean Hunt64f857b2011-06-23 01:22:53 +00008 <link type="text/css" rel="stylesheet" href="../menu.css">
9 <link type="text/css" rel="stylesheet" href="../content.css">
Sean Hunt7e98b472011-06-23 01:21:01 +000010 <style type="text/css">
11 td {
12 vertical-align: top;
13 }
Benjamin Kramer3419d7c2012-01-15 16:42:14 +000014 th { background-color: #ffddaa; }
Sean Hunt7e98b472011-06-23 01:21:01 +000015 </style>
Chris Lattner5ce933f2009-02-09 08:46:11 +000016</head>
17<body>
18
19<!--#include virtual="../menu.html.incl"-->
20
21<div id="content">
22
23<h1>Clang Language Extensions</h1>
24
25<ul>
26<li><a href="#intro">Introduction</a></li>
Chris Lattner148772a2009-06-13 07:13:28 +000027<li><a href="#feature_check">Feature Checking Macros</a></li>
John Thompson92bd8c72009-11-02 22:28:12 +000028<li><a href="#has_include">Include File Checking Macros</a></li>
Chris Lattner81edc9f2009-04-13 02:45:46 +000029<li><a href="#builtinmacros">Builtin Macros</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +000030<li><a href="#vectors">Vectors and Extended Vectors</a></li>
John McCall48209082010-11-08 19:48:17 +000031<li><a href="#deprecated">Messages on <tt>deprecated</tt> and <tt>unavailable</tt> attributes</a></li>
32<li><a href="#attributes-on-enumerators">Attributes on enumerators</a></li>
Daniel Dunbar85ff9692012-04-05 17:10:06 +000033<li><a href="#user_specified_system_framework">'User-Specified' System Frameworks</a></li>
Douglas Gregor93a70672012-03-11 04:53:21 +000034<li><a href="#availability">Availability attribute</a></li>
Sean Hunt7e98b472011-06-23 01:21:01 +000035<li><a href="#checking_language_features">Checks for Standard Language Features</a>
Ted Kremenek22c34102009-12-03 02:05:57 +000036 <ul>
37 <li><a href="#cxx_exceptions">C++ exceptions</a></li>
38 <li><a href="#cxx_rtti">C++ RTTI</a></li>
Sean Hunt7e98b472011-06-23 01:21:01 +000039 </ul></li>
40<li><a href="#checking_upcoming_features">Checks for Upcoming Standard Language Features</a>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +000041 <ul>
David Blaikie5090e9f2011-10-18 05:49:30 +000042 <li><a href="#cxx0x">C++11</a>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +000043 <ul>
David Blaikie5090e9f2011-10-18 05:49:30 +000044 <li><a href="#cxx_access_control_sfinae">C++11 SFINAE includes
45 access control</a></li>
46 <li><a href="#cxx_alias_templates">C++11 alias templates</a></li>
47 <li><a href="#cxx_alignas">C++11 alignment specifiers</a></li>
48 <li><a href="#cxx_attributes">C++11 attributes</a></li>
49 <li><a href="#cxx_constexpr">C++11 generalized constant expressions</a></li>
50 <li><a href="#cxx_decltype">C++11 <tt>decltype()</tt></a></li>
51 <li><a href="#cxx_default_function_template_args">C++11 default template arguments in function templates</a></li>
Douglas Gregorf695a692011-11-01 01:19:34 +000052 <li><a href="#cxx_defaulted_functions">C++11 defaulted functions</a></li>
David Blaikie5090e9f2011-10-18 05:49:30 +000053 <li><a href="#cxx_delegating_constructor">C++11 delegating constructors</a></li>
54 <li><a href="#cxx_deleted_functions">C++11 deleted functions</a></li>
55 <li><a href="#cxx_explicit_conversions">C++11 explicit conversion functions</a></li>
56 <li><a href="#cxx_generalized_initializers">C++11 generalized initializers</a></li>
57 <li><a href="#cxx_implicit_moves">C++11 implicit move constructors/assignment operators</a></li>
58 <li><a href="#cxx_inheriting_constructors">C++11 inheriting constructors</a></li>
59 <li><a href="#cxx_inline_namespaces">C++11 inline namespaces</a></li>
60 <li><a href="#cxx_lambdas">C++11 lambdas</a></li>
Douglas Gregor7b156dd2012-04-04 00:48:39 +000061 <li><a href="#cxx_local_type_template_args">C++11 local and unnamed types as template arguments</a></li>
David Blaikie5090e9f2011-10-18 05:49:30 +000062 <li><a href="#cxx_noexcept">C++11 noexcept specification</a></li>
63 <li><a href="#cxx_nonstatic_member_init">C++11 in-class non-static data member initialization</a></li>
64 <li><a href="#cxx_nullptr">C++11 nullptr</a></li>
65 <li><a href="#cxx_override_control">C++11 override control</a></li>
66 <li><a href="#cxx_range_for">C++11 range-based for loop</a></li>
67 <li><a href="#cxx_raw_string_literals">C++11 raw string literals</a></li>
68 <li><a href="#cxx_rvalue_references">C++11 rvalue references</a></li>
69 <li><a href="#cxx_reference_qualified_functions">C++11 reference-qualified functions</a></li>
70 <li><a href="#cxx_static_assert">C++11 <tt>static_assert()</tt></a></li>
71 <li><a href="#cxx_auto_type">C++11 type inference</a></li>
72 <li><a href="#cxx_strong_enums">C++11 strongly-typed enumerations</a></li>
73 <li><a href="#cxx_trailing_return">C++11 trailing return type</a></li>
74 <li><a href="#cxx_unicode_literals">C++11 Unicode string literals</a></li>
75 <li><a href="#cxx_unrestricted_unions">C++11 unrestricted unions</a></li>
76 <li><a href="#cxx_user_literals">C++11 user-defined literals</a></li>
77 <li><a href="#cxx_variadic_templates">C++11 variadic templates</a></li>
Douglas Gregorece38942011-08-29 17:28:38 +000078 </ul></li>
Benjamin Kramerffbe9b92011-12-23 17:00:35 +000079 <li><a href="#c11">C11</a>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +000080 <ul>
Benjamin Kramerffbe9b92011-12-23 17:00:35 +000081 <li><a href="#c_alignas">C11 alignment specifiers</a></li>
82 <li><a href="#c_generic_selections">C11 generic selections</a></li>
83 <li><a href="#c_static_assert">C11 <tt>_Static_assert()</tt></a></li>
Sean Hunt7e98b472011-06-23 01:21:01 +000084 </ul></li>
85 </ul> </li>
Douglas Gregorafdf1372011-02-03 21:57:35 +000086<li><a href="#checking_type_traits">Checks for Type Traits</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +000087<li><a href="#blocks">Blocks</a></li>
Douglas Gregor926df6c2011-06-11 01:09:30 +000088<li><a href="#objc_features">Objective-C Features</a>
89 <ul>
90 <li><a href="#objc_instancetype">Related result types</a></li>
John McCallf85e1932011-06-15 23:02:42 +000091 <li><a href="#objc_arc">Automatic reference counting</a></li>
Douglas Gregor5471bc82011-09-08 17:18:35 +000092 <li><a href="#objc_fixed_enum">Enumerations with a fixed underlying type</a></li>
Douglas Gregor8a4e1822012-03-09 23:24:48 +000093 <li><a href="#objc_lambdas">Interoperability with C++11 lambdas</a></li>
Patrick Beard62f12342012-03-20 21:51:03 +000094 <li><a href="#object-literals-subscripting">Object Literals and Subscripting</a></li>
Douglas Gregor926df6c2011-06-11 01:09:30 +000095 </ul>
96</li>
Douglas Gregorcb54d432009-02-13 00:57:04 +000097<li><a href="#overloading-in-c">Function Overloading in C</a></li>
Eli Friedman0c706c22011-09-19 23:17:44 +000098<li><a href="#complex-list-init">Initializer lists for complex numbers in C</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +000099<li><a href="#builtins">Builtin Functions</a>
100 <ul>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000101 <li><a href="#__builtin_shufflevector">__builtin_shufflevector</a></li>
Chris Lattner21190d52009-09-21 03:09:59 +0000102 <li><a href="#__builtin_unreachable">__builtin_unreachable</a></li>
Chris Lattner23aa9c82011-04-09 03:57:26 +0000103 <li><a href="#__sync_swap">__sync_swap</a></li>
Douglas Gregorafdf1372011-02-03 21:57:35 +0000104 </ul>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000105</li>
Chris Lattner1177f912009-04-09 19:58:15 +0000106<li><a href="#targetspecific">Target-Specific Extensions</a>
107 <ul>
108 <li><a href="#x86-specific">X86/X86-64 Language Extensions</a></li>
109 </ul>
110</li>
John McCall87494012011-03-18 03:51:49 +0000111<li><a href="#analyzerspecific">Static Analysis-Specific Extensions</a></li>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000112<li><a href="#dynamicanalyzerspecific">Dynamic Analysis-Specific Extensions</a>
Kostya Serebryanyce98c9b2011-11-28 20:51:02 +0000113 <ul>
114 <li><a href="#address_sanitizer">AddressSanitizer</a></li>
115 </ul>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000116</li>
117<li><a href="#threadsafety">Thread Safety Annotation Checking</a>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000118 <ul>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000119 <li><a href="#ts_noanal"><tt>no_thread_safety_analysis</tt></a></li>
120 <li><a href="#ts_lockable"><tt>lockable</tt></a></li>
121 <li><a href="#ts_scopedlockable"><tt>scoped_lockable</tt></a></li>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000122 <li><a href="#ts_guardedvar"><tt>guarded_var</tt></a></li>
123 <li><a href="#ts_ptguardedvar"><tt>pt_guarded_var</tt></a></li>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000124 <li><a href="#ts_guardedby"><tt>guarded_by(l)</tt></a></li>
125 <li><a href="#ts_ptguardedby"><tt>pt_guarded_by(l)</tt></a></li>
126 <li><a href="#ts_acquiredbefore"><tt>acquired_before(...)</tt></a></li>
127 <li><a href="#ts_acquiredafter"><tt>acquired_after(...)</tt></a></li>
128 <li><a href="#ts_elf"><tt>exclusive_lock_function(...)</tt></a></li>
129 <li><a href="#ts_slf"><tt>shared_lock_function(...)</tt></a></li>
130 <li><a href="#ts_etf"><tt>exclusive_trylock_function(...)</tt></a></li>
131 <li><a href="#ts_stf"><tt>shared_trylock_function(...)</tt></a></li>
132 <li><a href="#ts_uf"><tt>unlock_function(...)</tt></a></li>
133 <li><a href="#ts_lr"><tt>lock_returned(l)</tt></a></li>
134 <li><a href="#ts_le"><tt>locks_excluded(...)</tt></a></li>
135 <li><a href="#ts_elr"><tt>exclusive_locks_required(...)</tt></a></li>
136 <li><a href="#ts_slr"><tt>shared_locks_required(...)</tt></a></li>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000137 </ul>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000138</li>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000139</ul>
140
Chris Lattner5ce933f2009-02-09 08:46:11 +0000141<!-- ======================================================================= -->
142<h2 id="intro">Introduction</h2>
143<!-- ======================================================================= -->
144
145<p>This document describes the language extensions provided by Clang. In
Chris Lattner148772a2009-06-13 07:13:28 +0000146addition to the language extensions listed here, Clang aims to support a broad
Chris Lattner5ce933f2009-02-09 08:46:11 +0000147range of GCC extensions. Please see the <a
148href="http://gcc.gnu.org/onlinedocs/gcc/C-Extensions.html">GCC manual</a> for
149more information on these extensions.</p>
150
151<!-- ======================================================================= -->
Chris Lattner148772a2009-06-13 07:13:28 +0000152<h2 id="feature_check">Feature Checking Macros</h2>
153<!-- ======================================================================= -->
154
155<p>Language extensions can be very useful, but only if you know you can depend
Chris Lattnerc70e1932011-03-21 16:25:11 +0000156on them. In order to allow fine-grain features checks, we support three builtin
Chris Lattner148772a2009-06-13 07:13:28 +0000157function-like macros. This allows you to directly test for a feature in your
158code without having to resort to something like autoconf or fragile "compiler
159version checks".</p>
160
161<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000162<h3><a name="__has_builtin">__has_builtin</a></h3>
Chris Lattner148772a2009-06-13 07:13:28 +0000163<!-- ======================================================================= -->
164
165<p>This function-like macro takes a single identifier argument that is the name
166of a builtin function. It evaluates to 1 if the builtin is supported or 0 if
167not. It can be used like this:</p>
168
169<blockquote>
170<pre>
171#ifndef __has_builtin // Optional of course.
172 #define __has_builtin(x) 0 // Compatibility with non-clang compilers.
173#endif
174
175...
176#if __has_builtin(__builtin_trap)
177 __builtin_trap();
178#else
179 abort();
180#endif
181...
182</pre>
183</blockquote>
184
185
186<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000187<h3><a name="__has_feature_extension"> __has_feature and __has_extension</a></h3>
Chris Lattner148772a2009-06-13 07:13:28 +0000188<!-- ======================================================================= -->
189
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000190<p>These function-like macros take a single identifier argument that is the
191name of a feature. <code>__has_feature</code> evaluates to 1 if the feature
192is both supported by Clang and standardized in the current language standard
193or 0 if not (but see <a href="#has_feature_back_compat">below</a>), while
194<code>__has_extension</code> evaluates to 1 if the feature is supported by
195Clang in the current language (either as a language extension or a standard
196language feature) or 0 if not. They can be used like this:</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000197
198<blockquote>
199<pre>
200#ifndef __has_feature // Optional of course.
201 #define __has_feature(x) 0 // Compatibility with non-clang compilers.
202#endif
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000203#ifndef __has_extension
204 #define __has_extension __has_feature // Compatibility with pre-3.0 compilers.
205#endif
Chris Lattner148772a2009-06-13 07:13:28 +0000206
207...
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000208#if __has_feature(cxx_rvalue_references)
David Blaikie5090e9f2011-10-18 05:49:30 +0000209// This code will only be compiled with the -std=c++11 and -std=gnu++11
210// options, because rvalue references are only standardized in C++11.
Chris Lattner148772a2009-06-13 07:13:28 +0000211#endif
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000212
213#if __has_extension(cxx_rvalue_references)
David Blaikie5090e9f2011-10-18 05:49:30 +0000214// This code will be compiled with the -std=c++11, -std=gnu++11, -std=c++98
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000215// and -std=gnu++98 options, because rvalue references are supported as a
216// language extension in C++98.
217#endif
Chris Lattner148772a2009-06-13 07:13:28 +0000218</pre>
219</blockquote>
220
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000221<p id="has_feature_back_compat">For backwards compatibility reasons,
222<code>__has_feature</code> can also be used to test for support for
223non-standardized features, i.e. features not prefixed <code>c_</code>,
224<code>cxx_</code> or <code>objc_</code>.</p>
225
Kostya Serebryanyce98c9b2011-11-28 20:51:02 +0000226<p id="has_feature_for_non_language_features">
227Another use of <code>__has_feature</code> is to check for compiler features
228not related to the language standard, such as e.g.
229<a href="AddressSanitizer.html">AddressSanitizer</a>.
230
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000231<p>If the <code>-pedantic-errors</code> option is given,
232<code>__has_extension</code> is equivalent to <code>__has_feature</code>.</p>
233
Chris Lattner148772a2009-06-13 07:13:28 +0000234<p>The feature tag is described along with the language feature below.</p>
235
Richard Smith5297d712012-02-25 10:41:10 +0000236<p>The feature name or extension name can also be specified with a preceding and
237following <code>__</code> (double underscore) to avoid interference from a macro
Richard Smith1d9f4c12012-03-01 02:12:07 +0000238with the same name. For instance, <code>__cxx_rvalue_references__</code> can be
239used instead of <code>cxx_rvalue_references</code>.</p>
Richard Smith5297d712012-02-25 10:41:10 +0000240
John Thompson92bd8c72009-11-02 22:28:12 +0000241<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000242<h3><a name="__has_attribute">__has_attribute</a></h3>
Anders Carlssoncae50952010-10-20 02:31:43 +0000243<!-- ======================================================================= -->
244
245<p>This function-like macro takes a single identifier argument that is the name
246of an attribute. It evaluates to 1 if the attribute is supported or 0 if not. It
247can be used like this:</p>
248
249<blockquote>
250<pre>
251#ifndef __has_attribute // Optional of course.
252 #define __has_attribute(x) 0 // Compatibility with non-clang compilers.
253#endif
254
255...
Anders Carlsson961003d2011-01-24 03:54:51 +0000256#if __has_attribute(always_inline)
257#define ALWAYS_INLINE __attribute__((always_inline))
Anders Carlssoncae50952010-10-20 02:31:43 +0000258#else
Anders Carlsson961003d2011-01-24 03:54:51 +0000259#define ALWAYS_INLINE
Anders Carlssoncae50952010-10-20 02:31:43 +0000260#endif
261...
262</pre>
263</blockquote>
264
Jean-Daniel Dupas8a5e7fd2012-03-01 14:53:16 +0000265<p>The attribute name can also be specified with a preceding and
266following <code>__</code> (double underscore) to avoid interference from a macro
267with the same name. For instance, <code>__always_inline__</code> can be used
268instead of <code>always_inline</code>.</p>
269
Anders Carlssoncae50952010-10-20 02:31:43 +0000270<!-- ======================================================================= -->
John Thompson92bd8c72009-11-02 22:28:12 +0000271<h2 id="has_include">Include File Checking Macros</h2>
272<!-- ======================================================================= -->
273
274<p>Not all developments systems have the same include files.
275The <a href="#__has_include">__has_include</a> and
276<a href="#__has_include_next">__has_include_next</a> macros allow you to
277check for the existence of an include file before doing
278a possibly failing #include directive.</p>
279
280<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000281<h3><a name="__has_include">__has_include</a></h3>
John Thompson92bd8c72009-11-02 22:28:12 +0000282<!-- ======================================================================= -->
283
284<p>This function-like macro takes a single file name string argument that
285is the name of an include file. It evaluates to 1 if the file can
286be found using the include paths, or 0 otherwise:</p>
287
288<blockquote>
289<pre>
290// Note the two possible file name string formats.
Sean Hunt7e98b472011-06-23 01:21:01 +0000291#if __has_include("myinclude.h") &amp;&amp; __has_include(&lt;stdint.h&gt;)
John Thompson92bd8c72009-11-02 22:28:12 +0000292# include "myinclude.h"
293#endif
294
295// To avoid problem with non-clang compilers not having this macro.
Sean Hunt7e98b472011-06-23 01:21:01 +0000296#if defined(__has_include) &amp;&amp; __has_include("myinclude.h")
John Thompson92bd8c72009-11-02 22:28:12 +0000297# include "myinclude.h"
298#endif
299</pre>
300</blockquote>
301
302<p>To test for this feature, use #if defined(__has_include).</p>
303
304<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000305<h3><a name="__has_include_next">__has_include_next</a></h3>
John Thompson92bd8c72009-11-02 22:28:12 +0000306<!-- ======================================================================= -->
307
308<p>This function-like macro takes a single file name string argument that
309is the name of an include file. It is like __has_include except that it
310looks for the second instance of the given file found in the include
311paths. It evaluates to 1 if the second instance of the file can
312be found using the include paths, or 0 otherwise:</p>
313
314<blockquote>
315<pre>
316// Note the two possible file name string formats.
Sean Hunt7e98b472011-06-23 01:21:01 +0000317#if __has_include_next("myinclude.h") &amp;&amp; __has_include_next(&lt;stdint.h&gt;)
John Thompson92bd8c72009-11-02 22:28:12 +0000318# include_next "myinclude.h"
319#endif
320
321// To avoid problem with non-clang compilers not having this macro.
Sean Hunt7e98b472011-06-23 01:21:01 +0000322#if defined(__has_include_next) &amp;&amp; __has_include_next("myinclude.h")
John Thompson92bd8c72009-11-02 22:28:12 +0000323# include_next "myinclude.h"
324#endif
325</pre>
326</blockquote>
327
328<p>Note that __has_include_next, like the GNU extension
329#include_next directive, is intended for use in headers only,
330and will issue a warning if used in the top-level compilation
331file. A warning will also be issued if an absolute path
332is used in the file argument.</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000333
Ted Kremenekd7681502011-10-12 19:46:30 +0000334
335<!-- ======================================================================= -->
336<h3><a name="__has_warning">__has_warning</a></h3>
337<!-- ======================================================================= -->
338
339<p>This function-like macro takes a string literal that represents a command
340 line option for a warning and returns true if that is a valid warning
341 option.</p>
342
343<blockquote>
344<pre>
345#if __has_warning("-Wformat")
346...
347#endif
348</pre>
349</blockquote>
350
Chris Lattner148772a2009-06-13 07:13:28 +0000351<!-- ======================================================================= -->
Chris Lattner81edc9f2009-04-13 02:45:46 +0000352<h2 id="builtinmacros">Builtin Macros</h2>
353<!-- ======================================================================= -->
354
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000355<dl>
356 <dt><code>__BASE_FILE__</code></dt>
357 <dd>Defined to a string that contains the name of the main input
358 file passed to Clang.</dd>
359
360 <dt><code>__COUNTER__</code></dt>
361 <dd>Defined to an integer value that starts at zero and is
362 incremented each time the <code>__COUNTER__</code> macro is
363 expanded.</dd>
364
365 <dt><code>__INCLUDE_LEVEL__</code></dt>
366 <dd>Defined to an integral value that is the include depth of the
367 file currently being translated. For the main file, this value is
368 zero.</dd>
369
370 <dt><code>__TIMESTAMP__</code></dt>
371 <dd>Defined to the date and time of the last modification of the
372 current source file.</dd>
373
374 <dt><code>__clang__</code></dt>
375 <dd>Defined when compiling with Clang</dd>
376
377 <dt><code>__clang_major__</code></dt>
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000378 <dd>Defined to the major marketing version number of Clang (e.g., the
379 2 in 2.0.1). Note that marketing version numbers should not be used to
380 check for language features, as different vendors use different numbering
381 schemes. Instead, use the <a href="#feature_check">feature checking
382 macros</a>.</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000383
384 <dt><code>__clang_minor__</code></dt>
385 <dd>Defined to the minor version number of Clang (e.g., the 0 in
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000386 2.0.1). Note that marketing version numbers should not be used to
387 check for language features, as different vendors use different numbering
388 schemes. Instead, use the <a href="#feature_check">feature checking
389 macros</a>.</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000390
391 <dt><code>__clang_patchlevel__</code></dt>
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000392 <dd>Defined to the marketing patch level of Clang (e.g., the 1 in 2.0.1).</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000393
394 <dt><code>__clang_version__</code></dt>
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000395 <dd>Defined to a string that captures the Clang marketing version, including
396 the Subversion tag or revision number, e.g., "1.5 (trunk 102332)".</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000397</dl>
Chris Lattner81edc9f2009-04-13 02:45:46 +0000398
399<!-- ======================================================================= -->
Chris Lattner5ce933f2009-02-09 08:46:11 +0000400<h2 id="vectors">Vectors and Extended Vectors</h2>
401<!-- ======================================================================= -->
402
Anton Yartsevda90c772012-01-15 16:22:24 +0000403<p>Supports the GCC, OpenCL, AltiVec and NEON vector extensions.</p>
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000404
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000405<p>OpenCL vector types are created using <tt>ext_vector_type</tt> attribute. It
406support for <tt>V.xyzw</tt> syntax and other tidbits as seen in OpenCL. An
407example is:</p>
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000408
409<blockquote>
410<pre>
411typedef float float4 <b>__attribute__((ext_vector_type(4)))</b>;
412typedef float float2 <b>__attribute__((ext_vector_type(2)))</b>;
413
414float4 foo(float2 a, float2 b) {
415 float4 c;
416 c.xz = a;
417 c.yw = b;
418 return c;
419}
John McCall48209082010-11-08 19:48:17 +0000420</pre>
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000421</blockquote>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000422
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000423<p>Query for this feature with
424<tt>__has_extension(attribute_ext_vector_type)</tt>.</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000425
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000426<p>Giving <tt>-faltivec</tt> option to clang enables support for AltiVec vector
427syntax and functions. For example:</p>
Anton Yartsevda90c772012-01-15 16:22:24 +0000428
429<blockquote>
430<pre>
431vector float foo(vector int a) {
432 vector int b;
433 b = vec_add(a, a) + a;
434 return (vector float)b;
435}
436</pre>
437</blockquote>
438
439<p>NEON vector types are created using <tt>neon_vector_type</tt> and
440<tt>neon_polyvector_type</tt> attributes. For example:</p>
441
442<blockquote>
443<pre>
444typedef <b>__attribute__((neon_vector_type(8)))</b> int8_t int8x8_t;
445typedef <b>__attribute__((neon_polyvector_type(16)))</b> poly8_t poly8x16_t;
446
447int8x8_t foo(int8x8_t a) {
448 int8x8_t v;
449 v = a;
450 return v;
451}
452</pre>
453</blockquote>
454
455<!-- ======================================================================= -->
456<h3><a name="vector_literals">Vector Literals</a></h3>
457<!-- ======================================================================= -->
458
459<p>Vector literals can be used to create vectors from a set of scalars, or
460vectors. Either parentheses or braces form can be used. In the parentheses form
461the number of literal values specified must be one, i.e. referring to a scalar
462value, or must match the size of the vector type being created. If a single
463scalar literal value is specified, the scalar literal value will be replicated
464to all the components of the vector type. In the brackets form any number of
465literals can be specified. For example:</p>
466
467<blockquote>
468<pre>
469typedef int v4si __attribute__((__vector_size__(16)));
470typedef float float4 __attribute__((ext_vector_type(4)));
471typedef float float2 __attribute__((ext_vector_type(2)));
472
473v4si vsi = (v4si){1, 2, 3, 4};
474float4 vf = (float4)(1.0f, 2.0f, 3.0f, 4.0f);
475vector int vi1 = (vector int)(1); // vi1 will be (1, 1, 1, 1).
476vector int vi2 = (vector int){1}; // vi2 will be (1, 0, 0, 0).
477vector int vi3 = (vector int)(1, 2); // error
478vector int vi4 = (vector int){1, 2}; // vi4 will be (1, 2, 0, 0).
479vector int vi5 = (vector int)(1, 2, 3, 4);
480float4 vf = (float4)((float2)(1.0f, 2.0f), (float2)(3.0f, 4.0f));
481</pre>
482</blockquote>
483
484<!-- ======================================================================= -->
485<h3><a name="vector_operations">Vector Operations</a></h3>
486<!-- ======================================================================= -->
487
488<p>The table below shows the support for each operation by vector extension.
489A dash indicates that an operation is not accepted according to a corresponding
490specification.</p>
491
492<table width="500" border="1" cellspacing="0">
493 <tr>
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000494 <th>Operator</th>
495 <th>OpenCL</th>
496 <th>AltiVec</th>
497 <th>GCC</th>
498 <th>NEON</th>
Anton Yartsevda90c772012-01-15 16:22:24 +0000499 </tr>
500 <tr>
501 <td>[]</td>
502 <td align="center">yes</td>
503 <td align="center">yes</td>
504 <td align="center">yes</td>
505 <td align="center">-</td>
506 </tr>
507 <tr>
508 <td>unary operators +, -</td>
509 <td align="center">yes</td>
510 <td align="center">yes</td>
511 <td align="center">yes</td>
512 <td align="center">-</td>
513 </tr>
514 <tr>
515 <td>++, --</td>
516 <td align="center">yes</td>
517 <td align="center">yes</td>
518 <td align="center">-</td>
519 <td align="center">-</td>
520 </tr>
521 <tr>
522 <td>+, -, *, /, %</td>
523 <td align="center">yes</td>
524 <td align="center">yes</td>
525 <td align="center">yes</td>
526 <td align="center">-</td>
527 </tr>
528 <tr>
529 <td>bitwise operators &, |, ^, ~</td>
530 <td align="center">yes</td>
531 <td align="center">yes</td>
532 <td align="center">yes</td>
533 <td align="center">-</td>
534 </tr>
535 <tr>
536 <td>&gt&gt, &lt&lt</td>
537 <td align="center">yes</td>
538 <td align="center">yes</td>
539 <td align="center">yes</td>
540 <td align="center">-</td>
541 </tr>
542 <tr>
543 <td>!, &&,||</td>
544 <td align="center">no</td>
545 <td align="center">-</td>
546 <td align="center">-</td>
547 <td align="center">-</td>
548 </tr>
549 <tr>
550 <td>==,!=, >, <, >=, <=</td>
551 <td align="center">yes</td>
552 <td align="center">yes</td>
553 <td align="center">-</td>
554 <td align="center">-</td>
555 </tr>
556 <tr>
557 <td>=</td>
558 <td align="center">yes</td>
559 <td align="center">yes</td>
560 <td align="center">yes</td>
561 <td align="center">yes</td>
562 </tr>
563 <tr>
564 <td>:?</td>
565 <td align="center">yes</td>
566 <td align="center">-</td>
567 <td align="center">-</td>
568 <td align="center">-</td>
569 </tr>
570 <tr>
571 <td>sizeof</td>
572 <td align="center">yes</td>
573 <td align="center">yes</td>
574 <td align="center">yes</td>
575 <td align="center">yes</td>
576 </tr>
577</table>
578
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000579<p>See also <a href="#__builtin_shufflevector">__builtin_shufflevector</a>.</p>
580
Chris Lattner5ce933f2009-02-09 08:46:11 +0000581<!-- ======================================================================= -->
John McCall48209082010-11-08 19:48:17 +0000582<h2 id="deprecated">Messages on <tt>deprecated</tt> and <tt>unavailable</tt> Attributes</h2>
Fariborz Jahanianc784dc12010-10-06 23:12:32 +0000583<!-- ======================================================================= -->
584
John McCall48209082010-11-08 19:48:17 +0000585<p>An optional string message can be added to the <tt>deprecated</tt>
586and <tt>unavailable</tt> attributes. For example:</p>
Fariborz Jahanianc784dc12010-10-06 23:12:32 +0000587
John McCall48209082010-11-08 19:48:17 +0000588<blockquote>
Chris Lattner4836d6a2010-11-09 19:43:35 +0000589<pre>void explode(void) __attribute__((deprecated("extremely unsafe, use 'combust' instead!!!")));</pre>
John McCall48209082010-11-08 19:48:17 +0000590</blockquote>
591
592<p>If the deprecated or unavailable declaration is used, the message
593will be incorporated into the appropriate diagnostic:</p>
594
595<blockquote>
Benjamin Kramerb4556862012-03-19 19:12:30 +0000596<pre>harmless.c:4:3: warning: 'explode' is deprecated: extremely unsafe, use 'combust' instead!!!
597 [-Wdeprecated-declarations]
John McCall48209082010-11-08 19:48:17 +0000598 explode();
599 ^</pre>
600</blockquote>
601
602<p>Query for this feature
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000603with <tt>__has_extension(attribute_deprecated_with_message)</tt>
604and <tt>__has_extension(attribute_unavailable_with_message)</tt>.</p>
John McCall48209082010-11-08 19:48:17 +0000605
606<!-- ======================================================================= -->
607<h2 id="attributes-on-enumerators">Attributes on Enumerators</h2>
608<!-- ======================================================================= -->
609
610<p>Clang allows attributes to be written on individual enumerators.
611This allows enumerators to be deprecated, made unavailable, etc. The
612attribute must appear after the enumerator name and before any
613initializer, like so:</p>
614
615<blockquote>
616<pre>enum OperationMode {
617 OM_Invalid,
618 OM_Normal,
619 OM_Terrified __attribute__((deprecated)),
620 OM_AbortOnError __attribute__((deprecated)) = 4
621};</pre>
622</blockquote>
623
624<p>Attributes on the <tt>enum</tt> declaration do not apply to
625individual enumerators.</p>
626
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000627<p>Query for this feature with <tt>__has_extension(enumerator_attributes)</tt>.</p>
Fariborz Jahanianc784dc12010-10-06 23:12:32 +0000628
629<!-- ======================================================================= -->
Daniel Dunbar85ff9692012-04-05 17:10:06 +0000630<h2 id="user_specified_system_framework">'User-Specified' System Frameworks</h2>
631<!-- ======================================================================= -->
632
633<p>Clang provides a mechanism by which frameworks can be built in such a way
634that they will always be treated as being 'system frameworks', even if they are
635not present in a system framework directory. This can be useful to system
636framework developers who want to be able to test building other applications
637with development builds of their framework, including the manner in which the
638compiler changes warning behavior for system headers.</p>
639
640<p>Framework developers can opt-in to this mechanism by creating a
641'.system_framework' file at the top-level of their framework. That is, the
642framework should have contents like:</p>
643
644<pre>
645 .../TestFramework.framework
646 .../TestFramework.framework/.system_framework
647 .../TestFramework.framework/Headers
648 .../TestFramework.framework/Headers/TestFramework.h
649 ...
650</pre>
651
652<p>Clang will treat the presence of this file as an indicator that the framework
653should be treated as a system framework, regardless of how it was found in the
654framework search path. For consistency, we recommend that such files never be
655included in installed versions of the framework.</p>
656
657<!-- ======================================================================= -->
Douglas Gregor93a70672012-03-11 04:53:21 +0000658<h2 id="availability">Availability attribute</h2
659<!-- ======================================================================= -->
660
661<p>Clang introduces the <code>availability</code> attribute, which can
662be placed on declarations to describe the lifecycle of that
663declaration relative to operating system versions. Consider the function declaration for a hypothetical function <code>f</code>:</p>
664
665<pre>
666void f(void) __attribute__((availability(macosx,introduced=10.4,deprecated=10.6,obsoleted=10.7)));
667</pre>
668
669<p>The availability attribute states that <code>f</code> was introduced in Mac OS X 10.4, deprecated in Mac OS X 10.6, and obsoleted in Mac OS X 10.7. This information is used by Clang to determine when it is safe to use <code>f</code>: for example, if Clang is instructed to compile code for Mac OS X 10.5, a call to <code>f()</code> succeeds. If Clang is instructed to compile code for Mac OS X 10.6, the call succeeds but Clang emits a warning specifying that the function is deprecated. Finally, if Clang is instructed to compile code for Mac OS X 10.7, the call fails because <code>f()</code> is no longer available.</p>
670
671<p>The availablility attribute is a comma-separated list starting with the platform name and then including clauses specifying important milestones in the declaration's lifetime (in any order) along with additional information. Those clauses can be:</p>
672
673<dl>
674 <dt>introduced=<i>version</i></dt>
675 <dd>The first version in which this declaration was introduced.</dd>
676
677 <dt>deprecated=<i>version</i></dt>
678 <dd>The first version in which this declaration was deprecated, meaning that users should migrate away from this API.</dd>
679
680 <dt>obsoleted=<i>version</i></dt>
681 <dd>The first version in which this declaration was obsoleted, meaning that it was removed completely and can no longer be used.</dd>
682
683 <dt>unavailable</dt>
684 <dd>This declaration is never available on this platform.</dd>
685
686 <dt>message=<i>string-literal</i></dt>
687 <dd>Additional message text that Clang will provide when emitting a warning or error about use of a deprecated or obsoleted declaration. Useful to direct users to replacement APIs.</dd>
688</dl>
689
690<p>Multiple availability attributes can be placed on a declaration, which may correspond to different platforms. Only the availability attribute with the platform corresponding to the target platform will be used; any others will be ignored. If no availability attribute specifies availability for the current target platform, the availability attributes are ignored. Supported platforms are:</p>
691
692<dl>
693 <dt>ios</dt>
694 <dd>Apple's iOS operating system. The minimum deployment target is specified by the <code>-mios-version-min=<i>version</i></code> or <code>-miphoneos-version-min=<i>version</i></code> command-line arguments.</dd>
695
696 <dt>macosx</dt>
697 <dd>Apple's Mac OS X operating system. The minimum deployment target is specified by the <code>-mmacosx-version-min=<i>version</i></code> command-line argument.</dd>
698</dl>
699
Douglas Gregor594f8412012-03-11 17:21:03 +0000700<p>A declaration can be used even when deploying back to a platform
701version prior to when the declaration was introduced. When this
702happens, the declaration is <a
703 href="https://developer.apple.com/library/mac/#documentation/MacOSX/Conceptual/BPFrameworks/Concepts/WeakLinking.html">weakly
704linked</a>, as if the <code>weak_import</code> attribute were added to the declaration. A weakly-linked declaration may or may not be present a run-time, and a program can determine whether the declaration is present by checking whether the address of that declaration is non-NULL.</p>
705
Douglas Gregor93a70672012-03-11 04:53:21 +0000706<!-- ======================================================================= -->
Ted Kremenek87774fd2009-12-03 02:04:01 +0000707<h2 id="checking_language_features">Checks for Standard Language Features</h2>
708<!-- ======================================================================= -->
709
710<p>The <tt>__has_feature</tt> macro can be used to query if certain standard language features are
711enabled. Those features are listed here.</p>
712
Ted Kremenek22c34102009-12-03 02:05:57 +0000713<h3 id="cxx_exceptions">C++ exceptions</h3>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000714
Ted Kremenek22c34102009-12-03 02:05:57 +0000715<p>Use <tt>__has_feature(cxx_exceptions)</tt> to determine if C++ exceptions have been enabled. For
Sean Hunt647ba1b2011-06-23 00:42:53 +0000716example, compiling code with <tt>-fexceptions</tt> enables C++ exceptions.</p>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000717
Ted Kremenek22c34102009-12-03 02:05:57 +0000718<h3 id="cxx_rtti">C++ RTTI</h3>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000719
Ted Kremenek0eb95602009-12-03 02:06:43 +0000720<p>Use <tt>__has_feature(cxx_rtti)</tt> to determine if C++ RTTI has been enabled. For example,
Ted Kremenek22c34102009-12-03 02:05:57 +0000721compiling code with <tt>-fno-rtti</tt> disables the use of RTTI.</p>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000722
723<!-- ======================================================================= -->
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000724<h2 id="checking_upcoming_features">Checks for Upcoming Standard Language Features</h2>
725<!-- ======================================================================= -->
726
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000727<p>The <tt>__has_feature</tt> or <tt>__has_extension</tt> macros can be used
728to query if certain upcoming standard language features are enabled. Those
729features are listed here. Features that are not yet implemented will be
730noted.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000731
David Blaikie5090e9f2011-10-18 05:49:30 +0000732<h3 id="cxx0x">C++11</h3>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000733
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000734<p>The features listed below are slated for inclusion in the upcoming
David Blaikie5090e9f2011-10-18 05:49:30 +0000735C++11 standard. As a result, all these features are enabled
736with the <tt>-std=c++11</tt> option when compiling C++ code.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000737
David Blaikie5090e9f2011-10-18 05:49:30 +0000738<h4 id="cxx_access_control_sfinae">C++11 SFINAE includes access control</h4>
Douglas Gregor7822ee32011-05-11 23:45:11 +0000739
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000740<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 Gregor7822ee32011-05-11 23:45:11 +0000741
David Blaikie5090e9f2011-10-18 05:49:30 +0000742<h4 id="cxx_alias_templates">C++11 alias templates</h4>
Richard Smith3e4c6c42011-05-05 21:57:07 +0000743
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000744<p>Use <tt>__has_feature(cxx_alias_templates)</tt> or
745<tt>__has_extension(cxx_alias_templates)</tt> to determine if support for
David Blaikie5090e9f2011-10-18 05:49:30 +0000746C++11's alias declarations and alias templates is enabled.</p>
Richard Smith3e4c6c42011-05-05 21:57:07 +0000747
David Blaikie5090e9f2011-10-18 05:49:30 +0000748<h4 id="cxx_alignas">C++11 alignment specifiers</h4>
Peter Collingbournefd5f6862011-10-14 23:44:46 +0000749
750<p>Use <tt>__has_feature(cxx_alignas)</tt> or
751<tt>__has_extension(cxx_alignas)</tt> to determine if support for alignment
752specifiers using <tt>alignas</tt> is enabled.</p>
753
David Blaikie5090e9f2011-10-18 05:49:30 +0000754<h4 id="cxx_attributes">C++11 attributes</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000755
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000756<p>Use <tt>__has_feature(cxx_attributes)</tt> or
757<tt>__has_extension(cxx_attributes)</tt> to determine if support for attribute
David Blaikie5090e9f2011-10-18 05:49:30 +0000758parsing with C++11's square bracket notation is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000759
David Blaikie5090e9f2011-10-18 05:49:30 +0000760<h4 id="cxx_constexpr">C++11 generalized constant expressions</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000761
762<p>Use <tt>__has_feature(cxx_constexpr)</tt> to determine if support
763for generalized constant expressions (e.g., <tt>constexpr</tt>) is
Richard Smithb5216aa2012-02-14 22:56:17 +0000764enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000765
David Blaikie5090e9f2011-10-18 05:49:30 +0000766<h4 id="cxx_decltype">C++11 <tt>decltype()</tt></h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000767
768<p>Use <tt>__has_feature(cxx_decltype)</tt> or
769<tt>__has_extension(cxx_decltype)</tt> to determine if support for the
770<tt>decltype()</tt> specifier is enabled.</p>
771
David Blaikie5090e9f2011-10-18 05:49:30 +0000772<h4 id="cxx_default_function_template_args">C++11 default template arguments in function templates</h4>
Douglas Gregor07508002011-02-05 20:35:30 +0000773
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000774<p>Use <tt>__has_feature(cxx_default_function_template_args)</tt> or
775<tt>__has_extension(cxx_default_function_template_args)</tt> to determine
776if support for default template arguments in function templates is enabled.</p>
Douglas Gregor07508002011-02-05 20:35:30 +0000777
Douglas Gregorf695a692011-11-01 01:19:34 +0000778<h4 id="cxx_defaulted_functions">C++11 <tt>default</tt>ed functions</h4>
779
780<p>Use <tt>__has_feature(cxx_defaulted_functions)</tt> or
781<tt>__has_extension(cxx_defaulted_functions)</tt> to determine if support for
782defaulted function definitions (with <tt>= default</tt>) is enabled.</p>
783
David Blaikie5090e9f2011-10-18 05:49:30 +0000784<h4 id="cxx_delegating_constructors">C++11 delegating constructors</h4>
Sean Huntd9624992011-06-23 06:11:37 +0000785
786<p>Use <tt>__has_feature(cxx_delegating_constructors)</tt> to determine if
787support for delegating constructors is enabled.</p>
788
David Blaikie5090e9f2011-10-18 05:49:30 +0000789<h4 id="cxx_deleted_functions">C++11 <tt>delete</tt>d functions</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000790
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000791<p>Use <tt>__has_feature(cxx_deleted_functions)</tt> or
792<tt>__has_extension(cxx_deleted_functions)</tt> to determine if support for
Sebastian Redlf6c09772010-08-31 23:28:47 +0000793deleted function definitions (with <tt>= delete</tt>) is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000794
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000795<h4 id="cxx_explicit_conversions">C++11 explicit conversion functions</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000796<p>Use <tt>__has_feature(cxx_explicit_conversions)</tt> to determine if support for <tt>explicit</tt> conversion functions is enabled.</p>
797
David Blaikie5090e9f2011-10-18 05:49:30 +0000798<h4 id="cxx_generalized_initializers">C++11 generalized initializers</h4>
Sean Hunte1f6dea2011-08-07 00:34:32 +0000799
800<p>Use <tt>__has_feature(cxx_generalized_initializers)</tt> to determine if
801support for generalized initializers (using braced lists and
Richard Smith88189552012-02-26 07:09:21 +0000802<tt>std::initializer_list</tt>) is enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000803
David Blaikie5090e9f2011-10-18 05:49:30 +0000804<h4 id="cxx_implicit_moves">C++11 implicit move constructors/assignment operators</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000805
Sebastian Redl72a81d22011-10-10 18:10:00 +0000806<p>Use <tt>__has_feature(cxx_implicit_moves)</tt> to determine if Clang will
807implicitly generate move constructors and move assignment operators where needed.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000808
David Blaikie5090e9f2011-10-18 05:49:30 +0000809<h4 id="cxx_inheriting_constructors">C++11 inheriting constructors</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000810
811<p>Use <tt>__has_feature(cxx_inheriting_constructors)</tt> to determine if support for inheriting constructors is enabled. Clang does not currently implement this feature.</p>
812
David Blaikie5090e9f2011-10-18 05:49:30 +0000813<h4 id="cxx_inline_namespaces">C++11 inline namespaces</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000814
815<p>Use <tt>__has_feature(cxx_inline_namespaces)</tt> or
816<tt>__has_extension(cxx_inline_namespaces)</tt> to determine if support for
817inline namespaces is enabled.</p>
Sean Hunte1f6dea2011-08-07 00:34:32 +0000818
David Blaikie5090e9f2011-10-18 05:49:30 +0000819<h4 id="cxx_lambdas">C++11 lambdas</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000820
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000821<p>Use <tt>__has_feature(cxx_lambdas)</tt> or
822<tt>__has_extension(cxx_lambdas)</tt> to determine if support for lambdas
Douglas Gregor46e021e2012-02-23 05:44:09 +0000823is enabled. </p>
Douglas Gregorece38942011-08-29 17:28:38 +0000824
Douglas Gregor7b156dd2012-04-04 00:48:39 +0000825<h4 id="cxx_local_type_template_args">C++11 local and unnamed types as template arguments</h4>
826
827<p>Use <tt>__has_feature(cxx_local_type_template_args)</tt> or
828<tt>__has_extension(cxx_local_type_template_args)</tt> to determine if
829support for local and unnamed types as template arguments is enabled.</p>
830
David Blaikie5090e9f2011-10-18 05:49:30 +0000831<h4 id="cxx_noexcept">C++11 noexcept</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000832
833<p>Use <tt>__has_feature(cxx_noexcept)</tt> or
834<tt>__has_extension(cxx_noexcept)</tt> to determine if support for noexcept
835exception specifications is enabled.</p>
836
David Blaikie5090e9f2011-10-18 05:49:30 +0000837<h4 id="cxx_nonstatic_member_init">C++11 in-class non-static data member initialization</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000838
839<p>Use <tt>__has_feature(cxx_nonstatic_member_init)</tt> to determine whether in-class initialization of non-static data members is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000840
David Blaikie5090e9f2011-10-18 05:49:30 +0000841<h4 id="cxx_nullptr">C++11 <tt>nullptr</tt></h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000842
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000843<p>Use <tt>__has_feature(cxx_nullptr)</tt> or
844<tt>__has_extension(cxx_nullptr)</tt> to determine if support for
Douglas Gregor84ee2ee2011-05-21 23:15:46 +0000845<tt>nullptr</tt> is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000846
David Blaikie5090e9f2011-10-18 05:49:30 +0000847<h4 id="cxx_override_control">C++11 <tt>override control</tt></h4>
Anders Carlssonc8b9f792011-03-25 15:04:23 +0000848
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000849<p>Use <tt>__has_feature(cxx_override_control)</tt> or
850<tt>__has_extension(cxx_override_control)</tt> to determine if support for
Anders Carlssonc8b9f792011-03-25 15:04:23 +0000851the override control keywords is enabled.</p>
852
David Blaikie5090e9f2011-10-18 05:49:30 +0000853<h4 id="cxx_reference_qualified_functions">C++11 reference-qualified functions</h4>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000854<p>Use <tt>__has_feature(cxx_reference_qualified_functions)</tt> or
855<tt>__has_extension(cxx_reference_qualified_functions)</tt> to determine
856if support for reference-qualified functions (e.g., member functions with
857<code>&amp;</code> or <code>&amp;&amp;</code> applied to <code>*this</code>)
858is enabled.</p>
Douglas Gregor56209ff2011-01-26 21:25:54 +0000859
David Blaikie5090e9f2011-10-18 05:49:30 +0000860<h4 id="cxx_range_for">C++11 range-based <tt>for</tt> loop</h4>
Richard Smitha391a462011-04-15 15:14:40 +0000861
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000862<p>Use <tt>__has_feature(cxx_range_for)</tt> or
863<tt>__has_extension(cxx_range_for)</tt> to determine if support for the
864range-based for loop is enabled. </p>
Richard Smitha391a462011-04-15 15:14:40 +0000865
David Blaikie5090e9f2011-10-18 05:49:30 +0000866<h4 id="cxx_raw_string_literals">C++11 raw string literals</h4>
Richard Smith80134582012-03-07 08:57:31 +0000867<p>Use <tt>__has_feature(cxx_raw_string_literals)</tt> to determine if support
868for raw string literals (e.g., <tt>R"x(foo\bar)x"</tt>) is enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000869
David Blaikie5090e9f2011-10-18 05:49:30 +0000870<h4 id="cxx_rvalue_references">C++11 rvalue references</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000871
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000872<p>Use <tt>__has_feature(cxx_rvalue_references)</tt> or
873<tt>__has_extension(cxx_rvalue_references)</tt> to determine if support for
Douglas Gregor56209ff2011-01-26 21:25:54 +0000874rvalue references is enabled. </p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000875
David Blaikie5090e9f2011-10-18 05:49:30 +0000876<h4 id="cxx_static_assert">C++11 <tt>static_assert()</tt></h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000877
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000878<p>Use <tt>__has_feature(cxx_static_assert)</tt> or
879<tt>__has_extension(cxx_static_assert)</tt> to determine if support for
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000880compile-time assertions using <tt>static_assert</tt> is enabled.</p>
881
David Blaikie5090e9f2011-10-18 05:49:30 +0000882<h4 id="cxx_auto_type">C++11 type inference</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000883
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000884<p>Use <tt>__has_feature(cxx_auto_type)</tt> or
David Blaikie5090e9f2011-10-18 05:49:30 +0000885<tt>__has_extension(cxx_auto_type)</tt> to determine C++11 type inference is
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000886supported using the <tt>auto</tt> specifier. If this is disabled, <tt>auto</tt>
887will instead be a storage class specifier, as in C or C++98.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000888
David Blaikie5090e9f2011-10-18 05:49:30 +0000889<h4 id="cxx_strong_enums">C++11 strongly typed enumerations</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000890
Douglas Gregorece38942011-08-29 17:28:38 +0000891<p>Use <tt>__has_feature(cxx_strong_enums)</tt> or
892<tt>__has_extension(cxx_strong_enums)</tt> to determine if support for
893strongly typed, scoped enumerations is enabled.</p>
Sebastian Redlf6c09772010-08-31 23:28:47 +0000894
David Blaikie5090e9f2011-10-18 05:49:30 +0000895<h4 id="cxx_trailing_return">C++11 trailing return type</h4>
Douglas Gregordab60ad2010-10-01 18:44:50 +0000896
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000897<p>Use <tt>__has_feature(cxx_trailing_return)</tt> or
898<tt>__has_extension(cxx_trailing_return)</tt> to determine if support for the
899alternate function declaration syntax with trailing return type is enabled.</p>
Douglas Gregordab60ad2010-10-01 18:44:50 +0000900
David Blaikie5090e9f2011-10-18 05:49:30 +0000901<h4 id="cxx_unicode_literals">C++11 Unicode string literals</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000902<p>Use <tt>__has_feature(cxx_unicode_literals)</tt> to determine if
903support for Unicode string literals is enabled.</p>
Sebastian Redl4561ecd2011-03-15 21:17:12 +0000904
David Blaikie5090e9f2011-10-18 05:49:30 +0000905<h4 id="cxx_unrestricted_unions">C++11 unrestricted unions</h4>
Sebastian Redl4561ecd2011-03-15 21:17:12 +0000906
Richard Smithec92bc72012-03-03 23:51:05 +0000907<p>Use <tt>__has_feature(cxx_unrestricted_unions)</tt> to determine if support for unrestricted unions is enabled.</p>
Douglas Gregor1274ccd2010-10-08 23:50:27 +0000908
David Blaikie5090e9f2011-10-18 05:49:30 +0000909<h4 id="cxx_user_literals">C++11 user-defined literals</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000910
Richard Smith9c1dda72012-03-09 08:41:27 +0000911<p>Use <tt>__has_feature(cxx_user_literals)</tt> to determine if support for user-defined literals is enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000912
David Blaikie5090e9f2011-10-18 05:49:30 +0000913<h4 id="cxx_variadic_templates">C++11 variadic templates</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000914
915<p>Use <tt>__has_feature(cxx_variadic_templates)</tt> or
916<tt>__has_extension(cxx_variadic_templates)</tt> to determine if support
917for variadic templates is enabled.</p>
Douglas Gregor1274ccd2010-10-08 23:50:27 +0000918
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000919<h3 id="c11">C11</h3>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000920
921<p>The features listed below are slated for inclusion in the upcoming
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000922C11 standard. As a result, all these features are enabled
923with the <tt>-std=c11</tt> option when compiling C code.</p>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000924
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000925<h4 id="c_alignas">C11 alignment specifiers</h4>
Peter Collingbournefd5f6862011-10-14 23:44:46 +0000926
927<p>Use <tt>__has_feature(c_alignas)</tt> or <tt>__has_extension(c_alignas)</tt>
928to determine if support for alignment specifiers using <tt>_Alignas</tt>
929is enabled.</p>
930
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000931<h4 id="c_generic_selections">C11 generic selections</h4>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000932
933<p>Use <tt>__has_feature(c_generic_selections)</tt> or
934<tt>__has_extension(c_generic_selections)</tt> to determine if support for
935generic selections is enabled.</p>
936
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000937<p>As an extension, the C11 generic selection expression is available in all
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000938languages supported by Clang. The syntax is the same as that given in the
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000939C11 standard.</p>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000940
941<p>In C, type compatibility is decided according to the rules given in the
942appropriate standard, but in C++, which lacks the type compatibility rules
943used in C, types are considered compatible only if they are equivalent.</p>
944
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000945<h4 id="c_static_assert">C11 <tt>_Static_assert()</tt></h4>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000946
947<p>Use <tt>__has_feature(c_static_assert)</tt> or
948<tt>__has_extension(c_static_assert)</tt> to determine if support for
949compile-time assertions using <tt>_Static_assert</tt> is enabled.</p>
950
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000951<!-- ======================================================================= -->
Douglas Gregorafdf1372011-02-03 21:57:35 +0000952<h2 id="checking_type_traits">Checks for Type Traits</h2>
953<!-- ======================================================================= -->
954
Sean Hunt7e98b472011-06-23 01:21:01 +0000955<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 Gregorafdf1372011-02-03 21:57:35 +0000956<blockquote>
957<pre>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000958#if __has_extension(is_convertible_to)
Douglas Gregorafdf1372011-02-03 21:57:35 +0000959template&lt;typename From, typename To&gt;
960struct is_convertible_to {
961 static const bool value = __is_convertible_to(From, To);
962};
963#else
964// Emulate type trait
965#endif
966</pre>
967</blockquote>
968
969<p>The following type traits are supported by Clang:</p>
970<ul>
971 <li><code>__has_nothrow_assign</code> (GNU, Microsoft)</li>
972 <li><code>__has_nothrow_copy</code> (GNU, Microsoft)</li>
973 <li><code>__has_nothrow_constructor</code> (GNU, Microsoft)</li>
974 <li><code>__has_trivial_assign</code> (GNU, Microsoft)</li>
975 <li><code>__has_trivial_copy</code> (GNU, Microsoft)</li>
976 <li><code>__has_trivial_constructor</code> (GNU, Microsoft)</li>
977 <li><code>__has_trivial_destructor</code> (GNU, Microsoft)</li>
978 <li><code>__has_virtual_destructor</code> (GNU, Microsoft)</li>
979 <li><code>__is_abstract</code> (GNU, Microsoft)</li>
980 <li><code>__is_base_of</code> (GNU, Microsoft)</li>
981 <li><code>__is_class</code> (GNU, Microsoft)</li>
982 <li><code>__is_convertible_to</code> (Microsoft)</li>
983 <li><code>__is_empty</code> (GNU, Microsoft)</li>
984 <li><code>__is_enum</code> (GNU, Microsoft)</li>
985 <li><code>__is_pod</code> (GNU, Microsoft)</li>
986 <li><code>__is_polymorphic</code> (GNU, Microsoft)</li>
987 <li><code>__is_union</code> (GNU, Microsoft)</li>
988 <li><code>__is_literal(type)</code>: Determines whether the given type is a literal type</li>
Douglas Gregor5e9392b2011-12-03 18:14:24 +0000989 <li><code>__is_final</code>: Determines whether the given type is declared with a <code>final</code> class-virt-specifier.</li>
David Blaikie5090e9f2011-10-18 05:49:30 +0000990 <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++11 standard library.</li>
Douglas Gregor4ca8ac22012-02-24 07:38:34 +0000991 <li><code>__is_trivially_assignable(totype, fromtype)</code>: Determines whether a value of type <tt>totype</tt> can be assigned to from a value of type <tt>fromtype</tt> such that no non-trivial functions are called as part of that assignment. This trait is required to implement the C++11 standard library.</li>
992 <li><code>__is_trivially_constructible(type, argtypes...)</code>: Determines whether a value of type <tt>type</tt> can be direct-initialized with arguments of types <tt>argtypes...</tt> such that no non-trivial functions are called as part of that initialization. This trait is required to implement the C++11 standard library.</li>
Douglas Gregorafdf1372011-02-03 21:57:35 +0000993</ul>
994
995<!-- ======================================================================= -->
Chris Lattner5ce933f2009-02-09 08:46:11 +0000996<h2 id="blocks">Blocks</h2>
997<!-- ======================================================================= -->
998
Chris Lattnera7dbdf52009-03-09 07:03:22 +0000999<p>The syntax and high level language feature description is in <a
1000href="BlockLanguageSpec.txt">BlockLanguageSpec.txt</a>. Implementation and ABI
1001details for the clang implementation are in <a
Chris Lattner5d7650b2010-03-16 21:43:03 +00001002href="Block-ABI-Apple.txt">Block-ABI-Apple.txt</a>.</p>
Chris Lattner5ce933f2009-02-09 08:46:11 +00001003
Chris Lattner148772a2009-06-13 07:13:28 +00001004
Peter Collingbournec1b5fa42011-05-13 20:54:45 +00001005<p>Query for this feature with __has_extension(blocks).</p>
Chris Lattner148772a2009-06-13 07:13:28 +00001006
Chris Lattner5ce933f2009-02-09 08:46:11 +00001007<!-- ======================================================================= -->
Douglas Gregor926df6c2011-06-11 01:09:30 +00001008<h2 id="objc_features">Objective-C Features</h2>
1009<!-- ======================================================================= -->
1010
1011<h3 id="objc_instancetype">Related result types</h3>
1012
1013<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>
1014
1015<blockquote>
1016<pre>
1017@interface NSObject
1018+ (id)alloc;
1019- (id)init;
1020@end
1021
1022@interface NSArray : NSObject
1023@end
1024</pre>
1025</blockquote>
1026
1027<p>and this common initialization pattern</p>
1028
1029<blockquote>
1030<pre>
1031NSArray *array = [[NSArray alloc] init];
1032</pre>
1033</blockquote>
1034
1035<p>the type of the expression <code>[NSArray alloc]</code> is
1036<code>NSArray*</code> because <code>alloc</code> implicitly has a
1037related result type. Similarly, the type of the expression
1038<code>[[NSArray alloc] init]</code> is <code>NSArray*</code>, since
1039<code>init</code> has a related result type and its receiver is known
1040to 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>
1041
Douglas Gregore97179c2011-09-08 01:46:34 +00001042<p>A method with a related result type can be declared by using the
1043type <tt>instancetype</tt> as its result type. <tt>instancetype</tt>
1044is a contextual keyword that is only permitted in the result type of
1045an Objective-C method, e.g.</p>
1046
1047<pre>
1048@interface A
1049+ (<b>instancetype</b>)constructAnA;
1050@end
1051</pre>
1052
1053<p>The related result type can also be inferred for some methods.
1054To determine whether a method has an inferred related result type, the first
Douglas Gregor926df6c2011-06-11 01:09:30 +00001055word in the camel-case selector (e.g., "init" in "initWithObjects") is
Douglas Gregor8a0ace62011-11-03 18:33:01 +00001056considered, and the method will have a related result type if its return
Sean Hunt7e98b472011-06-23 01:21:01 +00001057type is compatible with the type of its class and if</p>
Douglas Gregor926df6c2011-06-11 01:09:30 +00001058
1059<ul>
1060
1061 <li>the first word is "alloc" or "new", and the method is a class
1062 method, or</li>
1063
1064 <li>the first word is "autorelease", "init", "retain", or "self",
1065 and the method is an instance method.</li>
1066
Sean Hunt7e98b472011-06-23 01:21:01 +00001067</ul>
Douglas Gregor926df6c2011-06-11 01:09:30 +00001068
1069<p>If a method with a related result type is overridden by a subclass
1070method, the subclass method must also return a type that is compatible
1071with the subclass type. For example:</p>
1072
1073<blockquote>
1074<pre>
1075@interface NSString : NSObject
1076- (NSUnrelated *)init; // incorrect usage: NSUnrelated is not NSString or a superclass of NSString
1077@end
1078</pre>
1079</blockquote>
1080
1081<p>Related result types only affect the type of a message send or
1082property access via the given method. In all other respects, a method
Douglas Gregore97179c2011-09-08 01:46:34 +00001083with a related result type is treated the same way as method that
1084returns <tt>id</tt>.</p>
Douglas Gregor926df6c2011-06-11 01:09:30 +00001085
Douglas Gregoraebb6532011-09-08 17:19:31 +00001086<p>Use <tt>__has_feature(objc_instancetype)</tt> to determine whether
1087the <tt>instancetype</tt> contextual keyword is available.</p>
1088
Douglas Gregor926df6c2011-06-11 01:09:30 +00001089<!-- ======================================================================= -->
John McCallf85e1932011-06-15 23:02:42 +00001090<h2 id="objc_arc">Automatic reference counting </h2>
1091<!-- ======================================================================= -->
1092
1093<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>
1094
1095<!-- ======================================================================= -->
Douglas Gregor5471bc82011-09-08 17:18:35 +00001096<h2 id="objc_fixed_enum">Enumerations with a fixed underlying type</h2>
1097<!-- ======================================================================= -->
1098
David Blaikie5090e9f2011-10-18 05:49:30 +00001099<p>Clang provides support for C++11 enumerations with a fixed
Douglas Gregor5471bc82011-09-08 17:18:35 +00001100underlying type within Objective-C. For example, one can write an
1101enumeration type as:</p>
1102
1103<pre>
1104typedef enum : unsigned char { Red, Green, Blue } Color;
1105</pre>
1106
1107<p>This specifies that the underlying type, which is used to store the
1108enumeration value, is <tt>unsigned char</tt>.</p>
1109
1110<p>Use <tt>__has_feature(objc_fixed_enum)</tt> to determine whether
1111support for fixed underlying types is available in Objective-C.</p>
1112
1113<!-- ======================================================================= -->
Douglas Gregor8a4e1822012-03-09 23:24:48 +00001114<h2 id="objc_lambdas">Interoperability with C++11 lambdas</h2>
1115<!-- ======================================================================= -->
1116
1117<p>Clang provides interoperability between C++11 lambdas and
1118blocks-based APIs, by permitting a lambda to be implicitly converted
1119to a block pointer with the corresponding signature. For example,
1120consider an API such as <code>NSArray</code>'s array-sorting
1121method:</p>
1122
1123<pre> - (NSArray *)sortedArrayUsingComparator:(NSComparator)cmptr; </pre>
1124
1125<p><code>NSComparator</code> is simply a typedef for the block pointer
1126<code>NSComparisonResult (^)(id, id)</code>, and parameters of this
1127type are generally provided with block literals as arguments. However,
1128one can also use a C++11 lambda so long as it provides the same
1129signature (in this case, accepting two parameters of type
1130<code>id</code> and returning an <code>NSComparisonResult</code>):</p>
1131
1132<pre>
1133 NSArray *array = @[@"string 1", @"string 21", @"string 12", @"String 11",
1134 @"String 02"];
1135 const NSStringCompareOptions comparisonOptions
1136 = NSCaseInsensitiveSearch | NSNumericSearch |
1137 NSWidthInsensitiveSearch | NSForcedOrderingSearch;
1138 NSLocale *currentLocale = [NSLocale currentLocale];
1139 NSArray *sorted
1140 = [array sortedArrayUsingComparator:<b>[=](id s1, id s2) -&gt; NSComparisonResult {
1141 NSRange string1Range = NSMakeRange(0, [s1 length]);
1142 return [s1 compare:s2 options:comparisonOptions
1143 range:string1Range locale:currentLocale];
1144 }</b>];
1145 NSLog(@"sorted: %@", sorted);
1146</pre>
1147
1148<p>This code relies on an implicit conversion from the type of the
1149lambda expression (an unnamed, local class type called the <i>closure
1150type</i>) to the corresponding block pointer type. The conversion
1151itself is expressed by a conversion operator in that closure type
1152that produces a block pointer with the same signature as the lambda
1153itself, e.g.,</p>
1154
1155<pre>
1156 operator NSComparisonResult (^)(id, id)() const;
1157</pre>
1158
1159<p>This conversion function returns a new block that simply forwards
1160the two parameters to the lambda object (which it captures by copy),
1161then returns the result. The returned block is first copied (with
1162<tt>Block_copy</tt>) and then autoreleased. As an optimization, if a
1163lambda expression is immediately converted to a block pointer (as in
1164the first example, above), then the block is not copied and
1165autoreleased: rather, it is given the same lifetime as a block literal
1166written at that point in the program, which avoids the overhead of
1167copying a block to the heap in the common case.</p>
1168
Douglas Gregorbccda482012-03-10 22:20:11 +00001169<p>The conversion from a lambda to a block pointer is only available
1170in Objective-C++, and not in C++ with blocks, due to its use of
1171Objective-C memory management (autorelease).</p>
1172
Douglas Gregor8a4e1822012-03-09 23:24:48 +00001173<!-- ======================================================================= -->
Patrick Beard62f12342012-03-20 21:51:03 +00001174<h2 id="object-literals-subscripting">Object Literals and Subscripting</h2>
1175<!-- ======================================================================= -->
1176
Patrick Beard1526f542012-03-21 16:22:53 +00001177<p>Clang provides support for <a href="ObjectiveCLiterals.html">Object Literals and Subscripting</a> in Objective-C, which simplifies common Objective-C programming patterns, makes programs more concise, and improves the safety of container creation. There are several feature macros associated with object literals and subscripting: <code>__has_feature(objc_array_literals)</code> tests the availability of array literals; <code>__has_feature(objc_dictionary_literals)</code> tests the availability of dictionary literals; <code>__has_feature(objc_subscripting)</code> tests the availability of object subscripting.</p>
Patrick Beard62f12342012-03-20 21:51:03 +00001178
1179<!-- ======================================================================= -->
Douglas Gregorcb54d432009-02-13 00:57:04 +00001180<h2 id="overloading-in-c">Function Overloading in C</h2>
1181<!-- ======================================================================= -->
1182
Chris Lattnerf161d412009-02-13 21:51:45 +00001183<p>Clang provides support for C++ function overloading in C. Function
1184overloading in C is introduced using the <tt>overloadable</tt> attribute. For
1185example, one might provide several overloaded versions of a <tt>tgsin</tt>
1186function that invokes the appropriate standard function computing the sine of a
1187value with <tt>float</tt>, <tt>double</tt>, or <tt>long double</tt>
1188precision:</p>
Douglas Gregorcb54d432009-02-13 00:57:04 +00001189
1190<blockquote>
1191<pre>
1192#include &lt;math.h&gt;
1193float <b>__attribute__((overloadable))</b> tgsin(float x) { return sinf(x); }
1194double <b>__attribute__((overloadable))</b> tgsin(double x) { return sin(x); }
1195long double <b>__attribute__((overloadable))</b> tgsin(long double x) { return sinl(x); }
1196</pre>
1197</blockquote>
1198
1199<p>Given these declarations, one can call <tt>tgsin</tt> with a
1200<tt>float</tt> value to receive a <tt>float</tt> result, with a
1201<tt>double</tt> to receive a <tt>double</tt> result, etc. Function
1202overloading in C follows the rules of C++ function overloading to pick
1203the best overload given the call arguments, with a few C-specific
1204semantics:</p>
1205<ul>
1206 <li>Conversion from <tt>float</tt> or <tt>double</tt> to <tt>long
1207 double</tt> is ranked as a floating-point promotion (per C99) rather
1208 than as a floating-point conversion (as in C++).</li>
1209
1210 <li>A conversion from a pointer of type <tt>T*</tt> to a pointer of type
1211 <tt>U*</tt> is considered a pointer conversion (with conversion
1212 rank) if <tt>T</tt> and <tt>U</tt> are compatible types.</li>
1213
1214 <li>A conversion from type <tt>T</tt> to a value of type <tt>U</tt>
1215 is permitted if <tt>T</tt> and <tt>U</tt> are compatible types. This
1216 conversion is given "conversion" rank.</li>
1217</ul>
1218
1219<p>The declaration of <tt>overloadable</tt> functions is restricted to
1220function declarations and definitions. Most importantly, if any
1221function with a given name is given the <tt>overloadable</tt>
1222attribute, then all function declarations and definitions with that
1223name (and in that scope) must have the <tt>overloadable</tt>
Chris Lattnerf161d412009-02-13 21:51:45 +00001224attribute. This rule even applies to redeclarations of functions whose original
1225declaration had the <tt>overloadable</tt> attribute, e.g.,</p>
Douglas Gregorcb54d432009-02-13 00:57:04 +00001226
1227<blockquote>
1228<pre>
1229int f(int) __attribute__((overloadable));
1230float f(float); <i>// error: declaration of "f" must have the "overloadable" attribute</i>
1231
1232int g(int) __attribute__((overloadable));
1233int g(int) { } <i>// error: redeclaration of "g" must also have the "overloadable" attribute</i>
1234</pre>
1235</blockquote>
1236
Douglas Gregor965acbb2009-02-18 07:07:28 +00001237<p>Functions marked <tt>overloadable</tt> must have
1238prototypes. Therefore, the following code is ill-formed:</p>
1239
1240<blockquote>
1241<pre>
1242int h() __attribute__((overloadable)); <i>// error: h does not have a prototype</i>
1243</pre>
1244</blockquote>
1245
1246<p>However, <tt>overloadable</tt> functions are allowed to use a
1247ellipsis 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>
1248
1249<blockquote>
1250<pre>
Chris Lattner02246802009-02-18 22:27:46 +00001251void honeypot(...) __attribute__((overloadable, unavailable)); <i>// calling me is an error</i>
Douglas Gregor965acbb2009-02-18 07:07:28 +00001252</pre>
1253</blockquote>
1254
Douglas Gregorcb54d432009-02-13 00:57:04 +00001255<p>Functions declared with the <tt>overloadable</tt> attribute have
1256their names mangled according to the same rules as C++ function
1257names. For example, the three <tt>tgsin</tt> functions in our
1258motivating example get the mangled names <tt>_Z5tgsinf</tt>,
Chris Lattner71b48d62010-11-28 18:19:13 +00001259<tt>_Z5tgsind</tt>, and <tt>_Z5tgsine</tt>, respectively. There are two
Douglas Gregorcb54d432009-02-13 00:57:04 +00001260caveats to this use of name mangling:</p>
1261
1262<ul>
1263
1264 <li>Future versions of Clang may change the name mangling of
1265 functions overloaded in C, so you should not depend on an specific
1266 mangling. To be completely safe, we strongly urge the use of
1267 <tt>static inline</tt> with <tt>overloadable</tt> functions.</li>
1268
1269 <li>The <tt>overloadable</tt> attribute has almost no meaning when
1270 used in C++, because names will already be mangled and functions are
1271 already overloadable. However, when an <tt>overloadable</tt>
1272 function occurs within an <tt>extern "C"</tt> linkage specification,
1273 it's name <i>will</i> be mangled in the same way as it would in
1274 C.</li>
1275</ul>
1276
Peter Collingbournec1b5fa42011-05-13 20:54:45 +00001277<p>Query for this feature with __has_extension(attribute_overloadable).</p>
Chris Lattner148772a2009-06-13 07:13:28 +00001278
Eli Friedman0c706c22011-09-19 23:17:44 +00001279<!-- ======================================================================= -->
1280<h2 id="complex-list-init">Initializer lists for complex numbers in C</h2>
1281<!-- ======================================================================= -->
1282
1283<p>clang supports an extension which allows the following in C:</p>
1284
1285<blockquote>
1286<pre>
1287#include &lt;math.h&gt;
1288#include &lt;complex.h&gt;
1289complex float x = { 1.0f, INFINITY }; // Init to (1, Inf)
1290</pre>
1291</blockquote>
1292
1293<p>This construct is useful because there is no way to separately
1294initialize the real and imaginary parts of a complex variable in
1295standard C, given that clang does not support <code>_Imaginary</code>.
1296(clang also supports the <code>__real__</code> and <code>__imag__</code>
1297extensions from gcc, which help in some cases, but are not usable in
1298static initializers.)
1299
1300<p>Note that this extension does not allow eliding the braces; the
1301meaning of the following two lines is different:</p>
1302
1303<blockquote>
1304<pre>
1305complex float x[] = { { 1.0f, 1.0f } }; // [0] = (1, 1)
1306complex float x[] = { 1.0f, 1.0f }; // [0] = (1, 0), [1] = (1, 0)
1307</pre>
1308</blockquote>
1309
1310<p>This extension also works in C++ mode, as far as that goes, but does not
1311 apply to the C++ <code>std::complex</code>. (In C++11, list
1312 initialization allows the same syntax to be used with
1313 <code>std::complex</code> with the same meaning.)
Chris Lattner148772a2009-06-13 07:13:28 +00001314
Douglas Gregorcb54d432009-02-13 00:57:04 +00001315<!-- ======================================================================= -->
Chris Lattner5ce933f2009-02-09 08:46:11 +00001316<h2 id="builtins">Builtin Functions</h2>
1317<!-- ======================================================================= -->
1318
1319<p>Clang supports a number of builtin library functions with the same syntax as
1320GCC, including things like <tt>__builtin_nan</tt>,
1321<tt>__builtin_constant_p</tt>, <tt>__builtin_choose_expr</tt>,
1322<tt>__builtin_types_compatible_p</tt>, <tt>__sync_fetch_and_add</tt>, etc. In
1323addition to the GCC builtins, Clang supports a number of builtins that GCC does
1324not, which are listed here.</p>
1325
1326<p>Please note that Clang does not and will not support all of the GCC builtins
1327for vector operations. Instead of using builtins, you should use the functions
1328defined in target-specific header files like <tt>&lt;xmmintrin.h&gt;</tt>, which
1329define portable wrappers for these. Many of the Clang versions of these
1330functions are implemented directly in terms of <a href="#vectors">extended
1331vector support</a> instead of builtins, in order to reduce the number of
1332builtins that we need to implement.</p>
1333
Chris Lattner5ce933f2009-02-09 08:46:11 +00001334<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +00001335<h3><a name="__builtin_shufflevector">__builtin_shufflevector</a></h3>
Chris Lattner5ce933f2009-02-09 08:46:11 +00001336<!-- ======================================================================= -->
1337
Chris Lattneraad826b2009-09-16 18:56:12 +00001338<p><tt>__builtin_shufflevector</tt> is used to express generic vector
Chris Lattner6f72da52009-02-13 20:00:20 +00001339permutation/shuffle/swizzle operations. This builtin is also very important for
1340the implementation of various target-specific header files like
1341<tt>&lt;xmmintrin.h&gt;</tt>.
Chris Lattner5ce933f2009-02-09 08:46:11 +00001342</p>
1343
1344<p><b>Syntax:</b></p>
1345
1346<pre>
Chris Lattner6f72da52009-02-13 20:00:20 +00001347__builtin_shufflevector(vec1, vec2, index1, index2, ...)
Chris Lattner5ce933f2009-02-09 08:46:11 +00001348</pre>
1349
1350<p><b>Examples:</b></p>
1351
1352<pre>
Chris Lattner6f72da52009-02-13 20:00:20 +00001353 // Identity operation - return 4-element vector V1.
1354 __builtin_shufflevector(V1, V1, 0, 1, 2, 3)
1355
1356 // "Splat" element 0 of V1 into a 4-element result.
1357 __builtin_shufflevector(V1, V1, 0, 0, 0, 0)
1358
1359 // Reverse 4-element vector V1.
1360 __builtin_shufflevector(V1, V1, 3, 2, 1, 0)
1361
1362 // Concatenate every other element of 4-element vectors V1 and V2.
1363 __builtin_shufflevector(V1, V2, 0, 2, 4, 6)
1364
1365 // Concatenate every other element of 8-element vectors V1 and V2.
1366 __builtin_shufflevector(V1, V2, 0, 2, 4, 6, 8, 10, 12, 14)
Chris Lattner5ce933f2009-02-09 08:46:11 +00001367</pre>
1368
1369<p><b>Description:</b></p>
1370
Chris Lattner6f72da52009-02-13 20:00:20 +00001371<p>The first two arguments to __builtin_shufflevector are vectors that have the
1372same element type. The remaining arguments are a list of integers that specify
1373the elements indices of the first two vectors that should be extracted and
1374returned in a new vector. These element indices are numbered sequentially
1375starting with the first vector, continuing into the second vector. Thus, if
1376vec1 is a 4-element vector, index 5 would refer to the second element of vec2.
Chris Lattner5ce933f2009-02-09 08:46:11 +00001377</p>
1378
Chris Lattner6f72da52009-02-13 20:00:20 +00001379<p>The result of __builtin_shufflevector is a vector
1380with the same element type as vec1/vec2 but that has an element count equal to
1381the number of indices specified.
1382</p>
Chris Lattner5ce933f2009-02-09 08:46:11 +00001383
Chris Lattner21190d52009-09-21 03:09:59 +00001384<p>Query for this feature with __has_builtin(__builtin_shufflevector).</p>
1385
1386<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +00001387<h3><a name="__builtin_unreachable">__builtin_unreachable</a></h3>
Chris Lattner21190d52009-09-21 03:09:59 +00001388<!-- ======================================================================= -->
1389
1390<p><tt>__builtin_unreachable</tt> is used to indicate that a specific point in
1391the program cannot be reached, even if the compiler might otherwise think it
1392can. This is useful to improve optimization and eliminates certain warnings.
1393For example, without the <tt>__builtin_unreachable</tt> in the example below,
1394the compiler assumes that the inline asm can fall through and prints a "function
1395declared 'noreturn' should not return" warning.
1396</p>
1397
1398<p><b>Syntax:</b></p>
1399
1400<pre>
1401__builtin_unreachable()
1402</pre>
1403
1404<p><b>Example of Use:</b></p>
1405
1406<pre>
1407void myabort(void) __attribute__((noreturn));
1408void myabort(void) {
1409 asm("int3");
1410 __builtin_unreachable();
1411}
1412</pre>
1413
1414<p><b>Description:</b></p>
1415
1416<p>The __builtin_unreachable() builtin has completely undefined behavior. Since
1417it has undefined behavior, it is a statement that it is never reached and the
1418optimizer can take advantage of this to produce better code. This builtin takes
1419no arguments and produces a void result.
1420</p>
1421
1422<p>Query for this feature with __has_builtin(__builtin_unreachable).</p>
1423
Chris Lattner23aa9c82011-04-09 03:57:26 +00001424<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +00001425<h3><a name="__sync_swap">__sync_swap</a></h3>
Chris Lattner23aa9c82011-04-09 03:57:26 +00001426<!-- ======================================================================= -->
1427
1428<p><tt>__sync_swap</tt> is used to atomically swap integers or pointers in
1429memory.
1430</p>
1431
1432<p><b>Syntax:</b></p>
1433
1434<pre>
1435<i>type</i> __sync_swap(<i>type</i> *ptr, <i>type</i> value, ...)
1436</pre>
1437
1438<p><b>Example of Use:</b></p>
1439
1440<pre>
Sean Hunt7e98b472011-06-23 01:21:01 +00001441int old_value = __sync_swap(&amp;value, new_value);
Chris Lattner23aa9c82011-04-09 03:57:26 +00001442</pre>
1443
1444<p><b>Description:</b></p>
1445
1446<p>The __sync_swap() builtin extends the existing __sync_*() family of atomic
1447intrinsics to allow code to atomically swap the current value with the new
1448value. More importantly, it helps developers write more efficient and correct
1449code by avoiding expensive loops around __sync_bool_compare_and_swap() or
1450relying on the platform specific implementation details of
1451__sync_lock_test_and_set(). The __sync_swap() builtin is a full barrier.
1452</p>
1453
Chris Lattner21190d52009-09-21 03:09:59 +00001454
Chris Lattner1177f912009-04-09 19:58:15 +00001455<!-- ======================================================================= -->
1456<h2 id="targetspecific">Target-Specific Extensions</h2>
1457<!-- ======================================================================= -->
1458
1459<p>Clang supports some language features conditionally on some targets.</p>
1460
1461<!-- ======================================================================= -->
1462<h3 id="x86-specific">X86/X86-64 Language Extensions</h3>
1463<!-- ======================================================================= -->
1464
1465<p>The X86 backend has these language extensions:</p>
1466
1467<!-- ======================================================================= -->
1468<h4 id="x86-gs-segment">Memory references off the GS segment</h4>
1469<!-- ======================================================================= -->
1470
1471<p>Annotating a pointer with address space #256 causes it to be code generated
Chris Lattnera021e7c2009-05-05 18:54:47 +00001472relative to the X86 GS segment register, and address space #257 causes it to be
1473relative to the X86 FS segment. Note that this is a very very low-level
1474feature that should only be used if you know what you're doing (for example in
1475an OS kernel).</p>
Chris Lattner1177f912009-04-09 19:58:15 +00001476
1477<p>Here is an example:</p>
1478
1479<pre>
1480#define GS_RELATIVE __attribute__((address_space(256)))
1481int foo(int GS_RELATIVE *P) {
1482 return *P;
1483}
1484</pre>
1485
1486<p>Which compiles to (on X86-32):</p>
1487
1488<pre>
1489_foo:
1490 movl 4(%esp), %eax
1491 movl %gs:(%eax), %eax
1492 ret
1493</pre>
1494
Ted Kremeneked869312009-04-10 05:03:33 +00001495<!-- ======================================================================= -->
1496<h2 id="analyzerspecific">Static Analysis-Specific Extensions</h2>
1497<!-- ======================================================================= -->
1498
1499<p>Clang supports additional attributes that are useful for documenting program
1500invariants and rules for static analysis tools. The extensions documented here
1501are used by the <a
1502href="http://clang.llvm.org/StaticAnalysis.html">path-sensitive static analyzer
1503engine</a> that is part of Clang's Analysis library.</p>
1504
John McCall87494012011-03-18 03:51:49 +00001505<h3 id="attr_analyzer_noreturn">The <tt>analyzer_noreturn</tt> attribute</h3>
Ted Kremeneked869312009-04-10 05:03:33 +00001506
1507<p>Clang's static analysis engine understands the standard <tt>noreturn</tt>
Ted Kremenek4df21142009-04-10 05:04:22 +00001508attribute. This attribute, which is typically affixed to a function prototype,
1509indicates that a call to a given function never returns. Function prototypes for
1510common functions like <tt>exit</tt> are typically annotated with this attribute,
1511as well as a variety of common assertion handlers. Users can educate the static
1512analyzer about their own custom assertion handles (thus cutting down on false
1513positives due to false paths) by marking their own &quot;panic&quot; functions
1514with this attribute.</p>
Ted Kremeneked869312009-04-10 05:03:33 +00001515
1516<p>While useful, <tt>noreturn</tt> is not applicable in all cases. Sometimes
Nick Lewycky625b5862009-06-14 04:08:08 +00001517there are special functions that for all intents and purposes should be
1518considered panic functions (i.e., they are only called when an internal program
1519error occurs) but may actually return so that the program can fail gracefully.
1520The <tt>analyzer_noreturn</tt> attribute allows one to annotate such functions
1521as being interpreted as &quot;no return&quot; functions by the analyzer (thus
Chris Lattner28935892009-04-10 05:54:56 +00001522pruning bogus paths) but will not affect compilation (as in the case of
Ted Kremeneked869312009-04-10 05:03:33 +00001523<tt>noreturn</tt>).</p>
1524
1525<p><b>Usage</b>: The <tt>analyzer_noreturn</tt> attribute can be placed in the
Chris Lattner28935892009-04-10 05:54:56 +00001526same places where the <tt>noreturn</tt> attribute can be placed. It is commonly
Ted Kremeneked869312009-04-10 05:03:33 +00001527placed at the end of function prototypes:</p>
1528
1529<pre>
1530 void foo() <b>__attribute__((analyzer_noreturn))</b>;
Chris Lattner148772a2009-06-13 07:13:28 +00001531</pre>
1532
John McCall87494012011-03-18 03:51:49 +00001533<p>Query for this feature with
1534<tt>__has_attribute(analyzer_noreturn)</tt>.</p>
Chris Lattner148772a2009-06-13 07:13:28 +00001535
John McCall87494012011-03-18 03:51:49 +00001536<h3 id="attr_method_family">The <tt>objc_method_family</tt> attribute</h3>
1537
1538<p>Many methods in Objective-C have conventional meanings determined
1539by their selectors. For the purposes of static analysis, it is
1540sometimes useful to be able to mark a method as having a particular
1541conventional meaning despite not having the right selector, or as not
1542having the conventional meaning that its selector would suggest.
1543For these use cases, we provide an attribute to specifically describe
1544the <q>method family</q> that a method belongs to.</p>
1545
1546<p><b>Usage</b>: <tt>__attribute__((objc_method_family(X)))</tt>,
1547where <tt>X</tt> is one of <tt>none</tt>, <tt>alloc</tt>, <tt>copy</tt>,
1548<tt>init</tt>, <tt>mutableCopy</tt>, or <tt>new</tt>. This attribute
1549can only be placed at the end of a method declaration:</p>
1550
1551<pre>
1552 - (NSString*) initMyStringValue <b>__attribute__((objc_method_family(none)))</b>;
1553</pre>
1554
1555<p>Users who do not wish to change the conventional meaning of a
1556method, and who merely want to document its non-standard retain and
1557release semantics, should use the
1558<a href="#attr_retain_release">retaining behavior attributes</a>
1559described below.</p>
1560
1561<p>Query for this feature with
1562<tt>__has_attribute(objc_method_family)</tt>.</p>
1563
1564<h3 id="attr_retain_release">Objective-C retaining behavior attributes</h3>
John McCall630b7ae2011-01-25 04:26:21 +00001565
1566<p>In Objective-C, functions and methods are generally assumed to take
1567and return objects with +0 retain counts, with some exceptions for
1568special methods like <tt>+alloc</tt> and <tt>init</tt>. However,
1569there are exceptions, and so Clang provides attributes to allow these
1570exceptions to be documented, which helps the analyzer find leaks (and
John McCall87494012011-03-18 03:51:49 +00001571ignore non-leaks). Some exceptions may be better described using
1572the <a href="#attr_method_family"><tt>objc_method_family</tt></a>
1573attribute instead.</p>
John McCall630b7ae2011-01-25 04:26:21 +00001574
1575<p><b>Usage</b>: The <tt>ns_returns_retained</tt>, <tt>ns_returns_not_retained</tt>,
1576<tt>ns_returns_autoreleased</tt>, <tt>cf_returns_retained</tt>,
1577and <tt>cf_returns_not_retained</tt> attributes can be placed on
1578methods and functions that return Objective-C or CoreFoundation
1579objects. They are commonly placed at the end of a function prototype
1580or method declaration:</p>
1581
1582<pre>
1583 id foo() <b>__attribute__((ns_returns_retained))</b>;
1584
1585 - (NSString*) bar: (int) x <b>__attribute__((ns_returns_retained))</b>;
1586</pre>
1587
1588<p>The <tt>*_returns_retained</tt> attributes specify that the
1589returned object has a +1 retain count.
1590The <tt>*_returns_not_retained</tt> attributes specify that the return
1591object has a +0 retain count, even if the normal convention for its
1592selector would be +1. <tt>ns_returns_autoreleased</tt> specifies that the
1593returned object is +0, but is guaranteed to live at least as long as the
1594next flush of an autorelease pool.</p>
1595
1596<p><b>Usage</b>: The <tt>ns_consumed</tt> and <tt>cf_consumed</tt>
1597attributes can be placed on an parameter declaration; they specify
1598that the argument is expected to have a +1 retain count, which will be
1599balanced in some way by the function or method.
1600The <tt>ns_consumes_self</tt> attribute can only be placed on an
1601Objective-C method; it specifies that the method expects
1602its <tt>self</tt> parameter to have a +1 retain count, which it will
1603balance in some way.</p>
1604
1605<pre>
1606 void <b>foo(__attribute__((ns_consumed))</b> NSString *string);
1607
1608 - (void) bar <b>__attribute__((ns_consumes_self))</b>;
1609 - (void) baz: (id) <b>__attribute__((ns_consumed))</b> x;
1610</pre>
Ted Kremeneked869312009-04-10 05:03:33 +00001611
John McCall87494012011-03-18 03:51:49 +00001612<p>Query for these features with <tt>__has_attribute(ns_consumed)</tt>,
1613<tt>__has_attribute(ns_returns_retained)</tt>, etc.</p>
1614
Kostya Serebryanyce98c9b2011-11-28 20:51:02 +00001615<!-- ======================================================================= -->
1616<h2 id="dynamicanalyzerspecific">Dynamic Analysis-Specific Extensions</h2>
1617<!-- ======================================================================= -->
1618<h3 id="address_sanitizer">AddressSanitizer</h3>
1619<p> Use <code>__has_feature(address_sanitizer)</code>
1620to check if the code is being built with <a
1621 href="AddressSanitizer.html">AddressSanitizer</a>.
1622</p>
Kostya Serebryany71efba02012-01-24 19:25:38 +00001623<p>Use <tt>__attribute__((no_address_safety_analysis))</tt> on a function
1624declaration to specify that address safety instrumentation (e.g.
1625AddressSanitizer) should not be applied to that function.
1626</p>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +00001627
1628<!-- ======================================================================= -->
Caitlin Sadowski73cbbc82011-07-28 18:38:36 +00001629<h2 id="threadsafety">Thread-Safety Annotation Checking</h2>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +00001630<!-- ======================================================================= -->
1631
1632<p>Clang supports additional attributes for checking basic locking policies in
1633multithreaded programs.
1634Clang currently parses the following list of attributes, although
1635<b>the implementation for these annotations is currently in development.</b>
1636For more details, see the
1637<a href="http://gcc.gnu.org/wiki/ThreadSafetyAnnotation">GCC implementation</a>.
1638</p>
1639
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001640<h4 id="ts_noanal">no_thread_safety_analysis</h4>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +00001641
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001642<p>Use <tt>__attribute__((no_thread_safety_analysis))</tt> on a function
1643declaration to specify that the thread safety analysis should not be run on that
1644function. This attribute provides an escape hatch (e.g. for situations when it
1645is difficult to annotate the locking policy). </p>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +00001646
1647<h4 id="ts_lockable">lockable</h4>
1648
1649<p>Use <tt>__attribute__((lockable))</tt> on a class definition to specify
1650that it has a lockable type (e.g. a Mutex class). This annotation is primarily
1651used to check consistency.</p>
1652
1653<h4 id="ts_scopedlockable">scoped_lockable</h4>
1654
1655<p>Use <tt>__attribute__((scoped_lockable))</tt> on a class definition to
1656specify that it has a "scoped" lockable type. Objects of this type will acquire
1657the lock upon construction and release it upon going out of scope.
1658 This annotation is primarily used to check
1659consistency.</p>
1660
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001661<h4 id="ts_guardedvar">guarded_var</h4>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +00001662
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001663<p>Use <tt>__attribute__((guarded_var))</tt> on a variable declaration to
1664specify that the variable must be accessed while holding some lock.</p>
1665
1666<h4 id="ts_ptguardedvar">pt_guarded_var</h4>
1667
1668<p>Use <tt>__attribute__((pt_guarded_var))</tt> on a pointer declaration to
1669specify that the pointer must be dereferenced while holding some lock.</p>
1670
1671<h4 id="ts_guardedby">guarded_by(l)</h4>
1672
1673<p>Use <tt>__attribute__((guarded_by(l)))</tt> on a variable declaration to
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001674specify that the variable must be accessed while holding lock <tt>l</tt>.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001675
1676<h4 id="ts_ptguardedby">pt_guarded_by(l)</h4>
1677
1678<p>Use <tt>__attribute__((pt_guarded_by(l)))</tt> on a pointer declaration to
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001679specify that the pointer must be dereferenced while holding lock <tt>l</tt>.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001680
1681<h4 id="ts_acquiredbefore">acquired_before(...)</h4>
1682
1683<p>Use <tt>__attribute__((acquired_before(...)))</tt> on a declaration
1684of a lockable variable to specify that the lock must be acquired before all
1685attribute arguments. Arguments must be lockable type, and there must be at
1686least one argument.</p>
1687
1688<h4 id="ts_acquiredafter">acquired_after(...)</h4>
1689
1690<p>Use <tt>__attribute__((acquired_after(...)))</tt> on a declaration
1691of a lockable variable to specify that the lock must be acquired after all
1692attribute arguments. Arguments must be lockable type, and there must be at
1693least one argument.</p>
1694
1695<h4 id="ts_elf">exclusive_lock_function(...)</h4>
1696
1697<p>Use <tt>__attribute__((exclusive_lock_function(...)))</tt> on a function
1698declaration to specify that the function acquires all listed locks
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001699exclusively. This attribute takes zero or more arguments: either of lockable
1700type or integers indexing into function parameters of lockable type. If no
1701arguments are given, the acquired lock is implicitly <tt>this</tt> of the
1702enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001703
1704<h4 id="ts_slf">shared_lock_function(...)</h4>
1705
1706<p>Use <tt>__attribute__((shared_lock_function(...)))</tt> on a function
1707declaration to specify that the function acquires all listed locks, although
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001708 the locks may be shared (e.g. read locks). This attribute takes zero or more
1709arguments: either of lockable type or integers indexing into function
1710parameters of lockable type. If no arguments are given, the acquired lock is
1711implicitly <tt>this</tt> of the enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001712
1713<h4 id="ts_etf">exclusive_trylock_function(...)</h4>
1714
1715<p>Use <tt>__attribute__((exclusive_lock_function(...)))</tt> on a function
1716declaration to specify that the function will try (without blocking) to acquire
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001717all listed locks exclusively. This attribute takes one or more arguments. The
1718first argument is an integer or boolean value specifying the return value of a
1719successful lock acquisition. The remaining arugments are either of lockable type
1720or integers indexing into function parameters of lockable type. If only one
1721argument is given, the acquired lock is implicitly <tt>this</tt> of the
1722enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001723
1724<h4 id="ts_stf">shared_trylock_function(...)</h4>
1725
1726<p>Use <tt>__attribute__((shared_lock_function(...)))</tt> on a function
1727declaration to specify that the function will try (without blocking) to acquire
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001728all listed locks, although the locks may be shared (e.g. read locks). This
1729attribute takes one or more arguments. The first argument is an integer or
1730boolean value specifying the return value of a successful lock acquisition. The
1731remaining arugments are either of lockable type or integers indexing into
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001732function parameters of lockable type. If only one argument is given, the
1733acquired lock is implicitly <tt>this</tt> of the enclosing object.</p>
1734
1735<h4 id="ts_uf">unlock_function(...)</h4>
1736
1737<p>Use <tt>__attribute__((unlock_function(...)))</tt> on a function
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001738declaration to specify that the function release all listed locks. This
1739attribute takes zero or more arguments: either of lockable type or integers
1740indexing into function parameters of lockable type. If no arguments are given,
1741the acquired lock is implicitly <tt>this</tt> of the enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001742
1743<h4 id="ts_lr">lock_returned(l)</h4>
1744
1745<p>Use <tt>__attribute__((lock_returned(l)))</tt> on a function
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001746declaration to specify that the function returns lock <tt>l</tt> (<tt>l</tt>
1747must be of lockable type). This annotation is used to aid in resolving lock
1748expressions.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001749
1750<h4 id="ts_le">locks_excluded(...)</h4>
1751
1752<p>Use <tt>__attribute__((locks_excluded(...)))</tt> on a function declaration
Caitlin Sadowskib51e0312011-08-09 17:59:31 +00001753to specify that the function must not be called with the listed locks. Arguments
1754must be lockable type, and there must be at least one argument.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +00001755
1756<h4 id="ts_elr">exclusive_locks_required(...)</h4>
1757
1758<p>Use <tt>__attribute__((exclusive_locks_required(...)))</tt> on a function
1759declaration to specify that the function must be called while holding the listed
1760exclusive locks. Arguments must be lockable type, and there must be at
1761least one argument.</p>
1762
1763<h4 id="ts_slr">shared_locks_required(...)</h4>
1764
1765<p>Use <tt>__attribute__((shared_locks_required(...)))</tt> on a function
1766declaration to specify that the function must be called while holding the listed
1767shared locks. Arguments must be lockable type, and there must be at
1768least one argument.</p>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +00001769
Chris Lattner5ce933f2009-02-09 08:46:11 +00001770</div>
1771</body>
1772</html>