Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / clang / a841c19f7860393d6319bf40e9d662284462771d / . / docs / LanguageExtensions.html
blob: eac3c69997becbf8b13ddef108d5c28a036573fd [file] [log] [blame]
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
3<!-- Material used from: HTML 4.01 specs: http://www.w3.org/TR/html401/ -->
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]4<html>
5<head>
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]6 <META http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
Eli Friedman0c706c22011-09-19 23:17:44 +0000[diff] [blame]7 <title>Clang Language Extensions</title>
Sean Hunt64f857b2011-06-23 01:22:53 +0000[diff] [blame]8 <link type="text/css" rel="stylesheet" href="../menu.css">
9 <link type="text/css" rel="stylesheet" href="../content.css">
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]10 <style type="text/css">
11 td {
12 vertical-align: top;
13 }
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000[diff] [blame]14 th { background-color: #ffddaa; }
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]15 </style>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]16</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 +0000[diff] [blame]27<li><a href="#feature_check">Feature Checking Macros</a></li>
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]28<li><a href="#has_include">Include File Checking Macros</a></li>
Chris Lattner81edc9f2009-04-13 02:45:46 +0000[diff] [blame]29<li><a href="#builtinmacros">Builtin Macros</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]30<li><a href="#vectors">Vectors and Extended Vectors</a></li>
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]31<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 +0000[diff] [blame]33<li><a href="#user_specified_system_framework">'User-Specified' System Frameworks</a></li>
Douglas Gregor93a70672012-03-11 04:53:21 +0000[diff] [blame]34<li><a href="#availability">Availability attribute</a></li>
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]35<li><a href="#checking_language_features">Checks for Standard Language Features</a>
Ted Kremenek22c34102009-12-03 02:05:57 +0000[diff] [blame]36 <ul>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]37 <li><a href="#cxx98">C++98</a>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]38 <ul>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]39 <li><a href="#cxx_exceptions">C++ exceptions</a></li>
40 <li><a href="#cxx_rtti">C++ RTTI</a></li>
41 </ul></li>
42 <li><a href="#cxx11">C++11</a>
43 <ul>
44 <li><a href="#cxx_access_control_sfinae">C++11 SFINAE includes access control</a></li>
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]45 <li><a href="#cxx_alias_templates">C++11 alias templates</a></li>
46 <li><a href="#cxx_alignas">C++11 alignment specifiers</a></li>
47 <li><a href="#cxx_attributes">C++11 attributes</a></li>
48 <li><a href="#cxx_constexpr">C++11 generalized constant expressions</a></li>
49 <li><a href="#cxx_decltype">C++11 <tt>decltype()</tt></a></li>
50 <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 +0000[diff] [blame]51 <li><a href="#cxx_defaulted_functions">C++11 defaulted functions</a></li>
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]52 <li><a href="#cxx_delegating_constructor">C++11 delegating constructors</a></li>
53 <li><a href="#cxx_deleted_functions">C++11 deleted functions</a></li>
54 <li><a href="#cxx_explicit_conversions">C++11 explicit conversion functions</a></li>
55 <li><a href="#cxx_generalized_initializers">C++11 generalized initializers</a></li>
56 <li><a href="#cxx_implicit_moves">C++11 implicit move constructors/assignment operators</a></li>
57 <li><a href="#cxx_inheriting_constructors">C++11 inheriting constructors</a></li>
58 <li><a href="#cxx_inline_namespaces">C++11 inline namespaces</a></li>
59 <li><a href="#cxx_lambdas">C++11 lambdas</a></li>
Douglas Gregor7b156dd2012-04-04 00:48:39 +0000[diff] [blame]60 <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 +0000[diff] [blame]61 <li><a href="#cxx_noexcept">C++11 noexcept specification</a></li>
62 <li><a href="#cxx_nonstatic_member_init">C++11 in-class non-static data member initialization</a></li>
63 <li><a href="#cxx_nullptr">C++11 nullptr</a></li>
64 <li><a href="#cxx_override_control">C++11 override control</a></li>
65 <li><a href="#cxx_range_for">C++11 range-based for loop</a></li>
66 <li><a href="#cxx_raw_string_literals">C++11 raw string literals</a></li>
67 <li><a href="#cxx_rvalue_references">C++11 rvalue references</a></li>
68 <li><a href="#cxx_reference_qualified_functions">C++11 reference-qualified functions</a></li>
69 <li><a href="#cxx_static_assert">C++11 <tt>static_assert()</tt></a></li>
70 <li><a href="#cxx_auto_type">C++11 type inference</a></li>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]71 <li><a href="#cxx_strong_enums">C++11 strongly-typed enumerations</a></li>
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]72 <li><a href="#cxx_trailing_return">C++11 trailing return type</a></li>
73 <li><a href="#cxx_unicode_literals">C++11 Unicode string literals</a></li>
74 <li><a href="#cxx_unrestricted_unions">C++11 unrestricted unions</a></li>
75 <li><a href="#cxx_user_literals">C++11 user-defined literals</a></li>
76 <li><a href="#cxx_variadic_templates">C++11 variadic templates</a></li>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]77 </ul></li>
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]78 <li><a href="#c11">C11</a>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]79 <ul>
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]80 <li><a href="#c_alignas">C11 alignment specifiers</a></li>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]81 <li><a href="#c_atomic">C11 atomic operations</a></li>
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]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>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]84 </ul></li>
85</ul></li>
Douglas Gregorafdf1372011-02-03 21:57:35 +0000[diff] [blame]86<li><a href="#checking_type_traits">Checks for Type Traits</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]87<li><a href="#blocks">Blocks</a></li>
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]88<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 +0000[diff] [blame]91 <li><a href="#objc_arc">Automatic reference counting</a></li>
Douglas Gregor5471bc82011-09-08 17:18:35 +0000[diff] [blame]92 <li><a href="#objc_fixed_enum">Enumerations with a fixed underlying type</a></li>
Douglas Gregor8a4e1822012-03-09 23:24:48 +0000[diff] [blame]93 <li><a href="#objc_lambdas">Interoperability with C++11 lambdas</a></li>
Patrick Beardeb382ec2012-04-19 00:25:12 +0000[diff] [blame]94 <li><a href="#objc_object_literals_subscripting">Object Literals and Subscripting</a></li>
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]95 </ul>
96</li>
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]97<li><a href="#overloading-in-c">Function Overloading in C</a></li>
Eli Friedman0c706c22011-09-19 23:17:44 +0000[diff] [blame]98<li><a href="#complex-list-init">Initializer lists for complex numbers in C</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]99<li><a href="#builtins">Builtin Functions</a>
100 <ul>
Hal Finkela841c192012-08-05 22:03:08 +0000[diff] [blame^]101 <li><a href="#__builtin_readcyclecounter">__builtin_readcyclecounter</a></li>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]102 <li><a href="#__builtin_shufflevector">__builtin_shufflevector</a></li>
Chris Lattner21190d52009-09-21 03:09:59 +0000[diff] [blame]103 <li><a href="#__builtin_unreachable">__builtin_unreachable</a></li>
Chris Lattner23aa9c82011-04-09 03:57:26 +0000[diff] [blame]104 <li><a href="#__sync_swap">__sync_swap</a></li>
Douglas Gregorafdf1372011-02-03 21:57:35 +0000[diff] [blame]105 </ul>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]106</li>
Richard Smithe0d3b4c2012-05-03 18:27:39 +0000[diff] [blame]107<li><a href="#non-standard-attributes">Non-standard C++11 Attributes</a>
108<ul>
109 <li><a href="#clang__fallthrough">The <tt>clang::fallthrough</tt> attribute</a></li>
110</ul>
111</li>
Chris Lattner1177f912009-04-09 19:58:15 +0000[diff] [blame]112<li><a href="#targetspecific">Target-Specific Extensions</a>
113 <ul>
114 <li><a href="#x86-specific">X86/X86-64 Language Extensions</a></li>
115 </ul>
116</li>
John McCall87494012011-03-18 03:51:49 +0000[diff] [blame]117<li><a href="#analyzerspecific">Static Analysis-Specific Extensions</a></li>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]118<li><a href="#dynamicanalyzerspecific">Dynamic Analysis-Specific Extensions</a>
Kostya Serebryanyce98c9b2011-11-28 20:51:02 +0000[diff] [blame]119 <ul>
120 <li><a href="#address_sanitizer">AddressSanitizer</a></li>
121 </ul>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]122</li>
123<li><a href="#threadsafety">Thread Safety Annotation Checking</a>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]124 <ul>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]125 <li><a href="#ts_noanal"><tt>no_thread_safety_analysis</tt></a></li>
126 <li><a href="#ts_lockable"><tt>lockable</tt></a></li>
127 <li><a href="#ts_scopedlockable"><tt>scoped_lockable</tt></a></li>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]128 <li><a href="#ts_guardedvar"><tt>guarded_var</tt></a></li>
129 <li><a href="#ts_ptguardedvar"><tt>pt_guarded_var</tt></a></li>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]130 <li><a href="#ts_guardedby"><tt>guarded_by(l)</tt></a></li>
131 <li><a href="#ts_ptguardedby"><tt>pt_guarded_by(l)</tt></a></li>
132 <li><a href="#ts_acquiredbefore"><tt>acquired_before(...)</tt></a></li>
133 <li><a href="#ts_acquiredafter"><tt>acquired_after(...)</tt></a></li>
134 <li><a href="#ts_elf"><tt>exclusive_lock_function(...)</tt></a></li>
135 <li><a href="#ts_slf"><tt>shared_lock_function(...)</tt></a></li>
136 <li><a href="#ts_etf"><tt>exclusive_trylock_function(...)</tt></a></li>
137 <li><a href="#ts_stf"><tt>shared_trylock_function(...)</tt></a></li>
138 <li><a href="#ts_uf"><tt>unlock_function(...)</tt></a></li>
139 <li><a href="#ts_lr"><tt>lock_returned(l)</tt></a></li>
140 <li><a href="#ts_le"><tt>locks_excluded(...)</tt></a></li>
141 <li><a href="#ts_elr"><tt>exclusive_locks_required(...)</tt></a></li>
142 <li><a href="#ts_slr"><tt>shared_locks_required(...)</tt></a></li>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]143 </ul>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]144</li>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]145</ul>
146
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]147<!-- ======================================================================= -->
148<h2 id="intro">Introduction</h2>
149<!-- ======================================================================= -->
150
151<p>This document describes the language extensions provided by Clang. In
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]152addition to the language extensions listed here, Clang aims to support a broad
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]153range of GCC extensions. Please see the <a
154href="http://gcc.gnu.org/onlinedocs/gcc/C-Extensions.html">GCC manual</a> for
155more information on these extensions.</p>
156
157<!-- ======================================================================= -->
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]158<h2 id="feature_check">Feature Checking Macros</h2>
159<!-- ======================================================================= -->
160
161<p>Language extensions can be very useful, but only if you know you can depend
Chris Lattnerc70e1932011-03-21 16:25:11 +0000[diff] [blame]162on them. In order to allow fine-grain features checks, we support three builtin
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]163function-like macros. This allows you to directly test for a feature in your
164code without having to resort to something like autoconf or fragile "compiler
165version checks".</p>
166
167<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]168<h3><a name="__has_builtin">__has_builtin</a></h3>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]169<!-- ======================================================================= -->
170
171<p>This function-like macro takes a single identifier argument that is the name
172of a builtin function. It evaluates to 1 if the builtin is supported or 0 if
173not. It can be used like this:</p>
174
175<blockquote>
176<pre>
177#ifndef __has_builtin // Optional of course.
178 #define __has_builtin(x) 0 // Compatibility with non-clang compilers.
179#endif
180
181...
182#if __has_builtin(__builtin_trap)
183 __builtin_trap();
184#else
185 abort();
186#endif
187...
188</pre>
189</blockquote>
190
191
192<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]193<h3><a name="__has_feature_extension"> __has_feature and __has_extension</a></h3>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]194<!-- ======================================================================= -->
195
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]196<p>These function-like macros take a single identifier argument that is the
197name of a feature. <code>__has_feature</code> evaluates to 1 if the feature
198is both supported by Clang and standardized in the current language standard
199or 0 if not (but see <a href="#has_feature_back_compat">below</a>), while
200<code>__has_extension</code> evaluates to 1 if the feature is supported by
201Clang in the current language (either as a language extension or a standard
202language feature) or 0 if not. They can be used like this:</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]203
204<blockquote>
205<pre>
206#ifndef __has_feature // Optional of course.
207 #define __has_feature(x) 0 // Compatibility with non-clang compilers.
208#endif
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]209#ifndef __has_extension
210 #define __has_extension __has_feature // Compatibility with pre-3.0 compilers.
211#endif
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]212
213...
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]214#if __has_feature(cxx_rvalue_references)
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]215// This code will only be compiled with the -std=c++11 and -std=gnu++11
216// options, because rvalue references are only standardized in C++11.
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]217#endif
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]218
219#if __has_extension(cxx_rvalue_references)
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]220// This code will be compiled with the -std=c++11, -std=gnu++11, -std=c++98
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]221// and -std=gnu++98 options, because rvalue references are supported as a
222// language extension in C++98.
223#endif
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]224</pre>
225</blockquote>
226
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]227<p id="has_feature_back_compat">For backwards compatibility reasons,
228<code>__has_feature</code> can also be used to test for support for
229non-standardized features, i.e. features not prefixed <code>c_</code>,
230<code>cxx_</code> or <code>objc_</code>.</p>
231
Kostya Serebryanyce98c9b2011-11-28 20:51:02 +0000[diff] [blame]232<p id="has_feature_for_non_language_features">
233Another use of <code>__has_feature</code> is to check for compiler features
234not related to the language standard, such as e.g.
235<a href="AddressSanitizer.html">AddressSanitizer</a>.
236
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]237<p>If the <code>-pedantic-errors</code> option is given,
238<code>__has_extension</code> is equivalent to <code>__has_feature</code>.</p>
239
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]240<p>The feature tag is described along with the language feature below.</p>
241
Richard Smith5297d712012-02-25 10:41:10 +0000[diff] [blame]242<p>The feature name or extension name can also be specified with a preceding and
243following <code>__</code> (double underscore) to avoid interference from a macro
Richard Smith1d9f4c12012-03-01 02:12:07 +0000[diff] [blame]244with the same name. For instance, <code>__cxx_rvalue_references__</code> can be
245used instead of <code>cxx_rvalue_references</code>.</p>
Richard Smith5297d712012-02-25 10:41:10 +0000[diff] [blame]246
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]247<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]248<h3><a name="__has_attribute">__has_attribute</a></h3>
Anders Carlssoncae50952010-10-20 02:31:43 +0000[diff] [blame]249<!-- ======================================================================= -->
250
251<p>This function-like macro takes a single identifier argument that is the name
252of an attribute. It evaluates to 1 if the attribute is supported or 0 if not. It
253can be used like this:</p>
254
255<blockquote>
256<pre>
257#ifndef __has_attribute // Optional of course.
258 #define __has_attribute(x) 0 // Compatibility with non-clang compilers.
259#endif
260
261...
Anders Carlsson961003d2011-01-24 03:54:51 +0000[diff] [blame]262#if __has_attribute(always_inline)
263#define ALWAYS_INLINE __attribute__((always_inline))
Anders Carlssoncae50952010-10-20 02:31:43 +0000[diff] [blame]264#else
Anders Carlsson961003d2011-01-24 03:54:51 +0000[diff] [blame]265#define ALWAYS_INLINE
Anders Carlssoncae50952010-10-20 02:31:43 +0000[diff] [blame]266#endif
267...
268</pre>
269</blockquote>
270
Jean-Daniel Dupas8a5e7fd2012-03-01 14:53:16 +0000[diff] [blame]271<p>The attribute name can also be specified with a preceding and
272following <code>__</code> (double underscore) to avoid interference from a macro
273with the same name. For instance, <code>__always_inline__</code> can be used
274instead of <code>always_inline</code>.</p>
275
Anders Carlssoncae50952010-10-20 02:31:43 +0000[diff] [blame]276<!-- ======================================================================= -->
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]277<h2 id="has_include">Include File Checking Macros</h2>
278<!-- ======================================================================= -->
279
280<p>Not all developments systems have the same include files.
281The <a href="#__has_include">__has_include</a> and
282<a href="#__has_include_next">__has_include_next</a> macros allow you to
283check for the existence of an include file before doing
284a possibly failing #include directive.</p>
285
286<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]287<h3><a name="__has_include">__has_include</a></h3>
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]288<!-- ======================================================================= -->
289
290<p>This function-like macro takes a single file name string argument that
291is the name of an include file. It evaluates to 1 if the file can
292be found using the include paths, or 0 otherwise:</p>
293
294<blockquote>
295<pre>
296// Note the two possible file name string formats.
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]297#if __has_include("myinclude.h") &amp;&amp; __has_include(&lt;stdint.h&gt;)
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]298# include "myinclude.h"
299#endif
300
301// To avoid problem with non-clang compilers not having this macro.
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]302#if defined(__has_include) &amp;&amp; __has_include("myinclude.h")
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]303# include "myinclude.h"
304#endif
305</pre>
306</blockquote>
307
308<p>To test for this feature, use #if defined(__has_include).</p>
309
310<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]311<h3><a name="__has_include_next">__has_include_next</a></h3>
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]312<!-- ======================================================================= -->
313
314<p>This function-like macro takes a single file name string argument that
315is the name of an include file. It is like __has_include except that it
316looks for the second instance of the given file found in the include
317paths. It evaluates to 1 if the second instance of the file can
318be found using the include paths, or 0 otherwise:</p>
319
320<blockquote>
321<pre>
322// Note the two possible file name string formats.
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]323#if __has_include_next("myinclude.h") &amp;&amp; __has_include_next(&lt;stdint.h&gt;)
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]324# include_next "myinclude.h"
325#endif
326
327// To avoid problem with non-clang compilers not having this macro.
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]328#if defined(__has_include_next) &amp;&amp; __has_include_next("myinclude.h")
John Thompson92bd8c72009-11-02 22:28:12 +0000[diff] [blame]329# include_next "myinclude.h"
330#endif
331</pre>
332</blockquote>
333
334<p>Note that __has_include_next, like the GNU extension
335#include_next directive, is intended for use in headers only,
336and will issue a warning if used in the top-level compilation
337file. A warning will also be issued if an absolute path
338is used in the file argument.</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]339
Ted Kremenekd7681502011-10-12 19:46:30 +0000[diff] [blame]340
341<!-- ======================================================================= -->
342<h3><a name="__has_warning">__has_warning</a></h3>
343<!-- ======================================================================= -->
344
345<p>This function-like macro takes a string literal that represents a command
346 line option for a warning and returns true if that is a valid warning
347 option.</p>
348
349<blockquote>
350<pre>
351#if __has_warning("-Wformat")
352...
353#endif
354</pre>
355</blockquote>
356
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]357<!-- ======================================================================= -->
Chris Lattner81edc9f2009-04-13 02:45:46 +0000[diff] [blame]358<h2 id="builtinmacros">Builtin Macros</h2>
359<!-- ======================================================================= -->
360
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000[diff] [blame]361<dl>
362 <dt><code>__BASE_FILE__</code></dt>
363 <dd>Defined to a string that contains the name of the main input
364 file passed to Clang.</dd>
365
366 <dt><code>__COUNTER__</code></dt>
367 <dd>Defined to an integer value that starts at zero and is
368 incremented each time the <code>__COUNTER__</code> macro is
369 expanded.</dd>
370
371 <dt><code>__INCLUDE_LEVEL__</code></dt>
372 <dd>Defined to an integral value that is the include depth of the
373 file currently being translated. For the main file, this value is
374 zero.</dd>
375
376 <dt><code>__TIMESTAMP__</code></dt>
377 <dd>Defined to the date and time of the last modification of the
378 current source file.</dd>
379
380 <dt><code>__clang__</code></dt>
381 <dd>Defined when compiling with Clang</dd>
382
383 <dt><code>__clang_major__</code></dt>
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000[diff] [blame]384 <dd>Defined to the major marketing version number of Clang (e.g., the
385 2 in 2.0.1). Note that marketing version numbers should not be used to
386 check for language features, as different vendors use different numbering
387 schemes. Instead, use the <a href="#feature_check">feature checking
388 macros</a>.</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000[diff] [blame]389
390 <dt><code>__clang_minor__</code></dt>
391 <dd>Defined to the minor version number of Clang (e.g., the 0 in
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000[diff] [blame]392 2.0.1). Note that marketing version numbers should not be used to
393 check for language features, as different vendors use different numbering
394 schemes. Instead, use the <a href="#feature_check">feature checking
395 macros</a>.</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000[diff] [blame]396
397 <dt><code>__clang_patchlevel__</code></dt>
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000[diff] [blame]398 <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 +0000[diff] [blame]399
400 <dt><code>__clang_version__</code></dt>
Chris Lattnerd4b66b92011-12-15 19:06:36 +0000[diff] [blame]401 <dd>Defined to a string that captures the Clang marketing version, including
402 the Subversion tag or revision number, e.g., "1.5 (trunk 102332)".</dd>
Douglas Gregor4290fbd2010-04-30 02:51:06 +0000[diff] [blame]403</dl>
Chris Lattner81edc9f2009-04-13 02:45:46 +0000[diff] [blame]404
405<!-- ======================================================================= -->
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]406<h2 id="vectors">Vectors and Extended Vectors</h2>
407<!-- ======================================================================= -->
408
Anton Yartsevda90c772012-01-15 16:22:24 +0000[diff] [blame]409<p>Supports the GCC, OpenCL, AltiVec and NEON vector extensions.</p>
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000[diff] [blame]410
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000[diff] [blame]411<p>OpenCL vector types are created using <tt>ext_vector_type</tt> attribute. It
412support for <tt>V.xyzw</tt> syntax and other tidbits as seen in OpenCL. An
413example is:</p>
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000[diff] [blame]414
415<blockquote>
416<pre>
417typedef float float4 <b>__attribute__((ext_vector_type(4)))</b>;
418typedef float float2 <b>__attribute__((ext_vector_type(2)))</b>;
419
420float4 foo(float2 a, float2 b) {
421 float4 c;
422 c.xz = a;
423 c.yw = b;
424 return c;
425}
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]426</pre>
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000[diff] [blame]427</blockquote>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]428
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000[diff] [blame]429<p>Query for this feature with
430<tt>__has_extension(attribute_ext_vector_type)</tt>.</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]431
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000[diff] [blame]432<p>Giving <tt>-faltivec</tt> option to clang enables support for AltiVec vector
433syntax and functions. For example:</p>
Anton Yartsevda90c772012-01-15 16:22:24 +0000[diff] [blame]434
435<blockquote>
436<pre>
437vector float foo(vector int a) {
438 vector int b;
439 b = vec_add(a, a) + a;
440 return (vector float)b;
441}
442</pre>
443</blockquote>
444
445<p>NEON vector types are created using <tt>neon_vector_type</tt> and
446<tt>neon_polyvector_type</tt> attributes. For example:</p>
447
448<blockquote>
449<pre>
450typedef <b>__attribute__((neon_vector_type(8)))</b> int8_t int8x8_t;
451typedef <b>__attribute__((neon_polyvector_type(16)))</b> poly8_t poly8x16_t;
452
453int8x8_t foo(int8x8_t a) {
454 int8x8_t v;
455 v = a;
456 return v;
457}
458</pre>
459</blockquote>
460
461<!-- ======================================================================= -->
462<h3><a name="vector_literals">Vector Literals</a></h3>
463<!-- ======================================================================= -->
464
465<p>Vector literals can be used to create vectors from a set of scalars, or
466vectors. Either parentheses or braces form can be used. In the parentheses form
467the number of literal values specified must be one, i.e. referring to a scalar
468value, or must match the size of the vector type being created. If a single
469scalar literal value is specified, the scalar literal value will be replicated
470to all the components of the vector type. In the brackets form any number of
471literals can be specified. For example:</p>
472
473<blockquote>
474<pre>
475typedef int v4si __attribute__((__vector_size__(16)));
476typedef float float4 __attribute__((ext_vector_type(4)));
477typedef float float2 __attribute__((ext_vector_type(2)));
478
479v4si vsi = (v4si){1, 2, 3, 4};
480float4 vf = (float4)(1.0f, 2.0f, 3.0f, 4.0f);
481vector int vi1 = (vector int)(1); // vi1 will be (1, 1, 1, 1).
482vector int vi2 = (vector int){1}; // vi2 will be (1, 0, 0, 0).
483vector int vi3 = (vector int)(1, 2); // error
484vector int vi4 = (vector int){1, 2}; // vi4 will be (1, 2, 0, 0).
485vector int vi5 = (vector int)(1, 2, 3, 4);
486float4 vf = (float4)((float2)(1.0f, 2.0f), (float2)(3.0f, 4.0f));
487</pre>
488</blockquote>
489
490<!-- ======================================================================= -->
491<h3><a name="vector_operations">Vector Operations</a></h3>
492<!-- ======================================================================= -->
493
494<p>The table below shows the support for each operation by vector extension.
495A dash indicates that an operation is not accepted according to a corresponding
496specification.</p>
497
498<table width="500" border="1" cellspacing="0">
499 <tr>
Benjamin Kramer3419d7c2012-01-15 16:42:14 +0000[diff] [blame]500 <th>Operator</th>
501 <th>OpenCL</th>
502 <th>AltiVec</th>
503 <th>GCC</th>
504 <th>NEON</th>
Anton Yartsevda90c772012-01-15 16:22:24 +0000[diff] [blame]505 </tr>
506 <tr>
507 <td>[]</td>
508 <td align="center">yes</td>
509 <td align="center">yes</td>
510 <td align="center">yes</td>
511 <td align="center">-</td>
512 </tr>
513 <tr>
514 <td>unary operators +, -</td>
515 <td align="center">yes</td>
516 <td align="center">yes</td>
517 <td align="center">yes</td>
518 <td align="center">-</td>
519 </tr>
520 <tr>
521 <td>++, --</td>
522 <td align="center">yes</td>
523 <td align="center">yes</td>
524 <td align="center">-</td>
525 <td align="center">-</td>
526 </tr>
527 <tr>
528 <td>+, -, *, /, %</td>
529 <td align="center">yes</td>
530 <td align="center">yes</td>
531 <td align="center">yes</td>
532 <td align="center">-</td>
533 </tr>
534 <tr>
535 <td>bitwise operators &, |, ^, ~</td>
536 <td align="center">yes</td>
537 <td align="center">yes</td>
538 <td align="center">yes</td>
539 <td align="center">-</td>
540 </tr>
541 <tr>
542 <td>&gt&gt, &lt&lt</td>
543 <td align="center">yes</td>
544 <td align="center">yes</td>
545 <td align="center">yes</td>
546 <td align="center">-</td>
547 </tr>
548 <tr>
549 <td>!, &&,||</td>
550 <td align="center">no</td>
551 <td align="center">-</td>
552 <td align="center">-</td>
553 <td align="center">-</td>
554 </tr>
555 <tr>
556 <td>==,!=, >, <, >=, <=</td>
557 <td align="center">yes</td>
558 <td align="center">yes</td>
559 <td align="center">-</td>
560 <td align="center">-</td>
561 </tr>
562 <tr>
563 <td>=</td>
564 <td align="center">yes</td>
565 <td align="center">yes</td>
566 <td align="center">yes</td>
567 <td align="center">yes</td>
568 </tr>
569 <tr>
570 <td>:?</td>
571 <td align="center">yes</td>
572 <td align="center">-</td>
573 <td align="center">-</td>
574 <td align="center">-</td>
575 </tr>
576 <tr>
577 <td>sizeof</td>
578 <td align="center">yes</td>
579 <td align="center">yes</td>
580 <td align="center">yes</td>
581 <td align="center">yes</td>
582 </tr>
583</table>
584
Owen Andersond2bf0cd2010-01-27 01:22:36 +0000[diff] [blame]585<p>See also <a href="#__builtin_shufflevector">__builtin_shufflevector</a>.</p>
586
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]587<!-- ======================================================================= -->
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]588<h2 id="deprecated">Messages on <tt>deprecated</tt> and <tt>unavailable</tt> Attributes</h2>
Fariborz Jahanianc784dc12010-10-06 23:12:32 +0000[diff] [blame]589<!-- ======================================================================= -->
590
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]591<p>An optional string message can be added to the <tt>deprecated</tt>
592and <tt>unavailable</tt> attributes. For example:</p>
Fariborz Jahanianc784dc12010-10-06 23:12:32 +0000[diff] [blame]593
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]594<blockquote>
Chris Lattner4836d6a2010-11-09 19:43:35 +0000[diff] [blame]595<pre>void explode(void) __attribute__((deprecated("extremely unsafe, use 'combust' instead!!!")));</pre>
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]596</blockquote>
597
598<p>If the deprecated or unavailable declaration is used, the message
599will be incorporated into the appropriate diagnostic:</p>
600
601<blockquote>
Benjamin Kramerb4556862012-03-19 19:12:30 +0000[diff] [blame]602<pre>harmless.c:4:3: warning: 'explode' is deprecated: extremely unsafe, use 'combust' instead!!!
603 [-Wdeprecated-declarations]
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]604 explode();
605 ^</pre>
606</blockquote>
607
608<p>Query for this feature
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]609with <tt>__has_extension(attribute_deprecated_with_message)</tt>
610and <tt>__has_extension(attribute_unavailable_with_message)</tt>.</p>
John McCall48209082010-11-08 19:48:17 +0000[diff] [blame]611
612<!-- ======================================================================= -->
613<h2 id="attributes-on-enumerators">Attributes on Enumerators</h2>
614<!-- ======================================================================= -->
615
616<p>Clang allows attributes to be written on individual enumerators.
617This allows enumerators to be deprecated, made unavailable, etc. The
618attribute must appear after the enumerator name and before any
619initializer, like so:</p>
620
621<blockquote>
622<pre>enum OperationMode {
623 OM_Invalid,
624 OM_Normal,
625 OM_Terrified __attribute__((deprecated)),
626 OM_AbortOnError __attribute__((deprecated)) = 4
627};</pre>
628</blockquote>
629
630<p>Attributes on the <tt>enum</tt> declaration do not apply to
631individual enumerators.</p>
632
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]633<p>Query for this feature with <tt>__has_extension(enumerator_attributes)</tt>.</p>
Fariborz Jahanianc784dc12010-10-06 23:12:32 +0000[diff] [blame]634
635<!-- ======================================================================= -->
Daniel Dunbar85ff9692012-04-05 17:10:06 +0000[diff] [blame]636<h2 id="user_specified_system_framework">'User-Specified' System Frameworks</h2>
637<!-- ======================================================================= -->
638
639<p>Clang provides a mechanism by which frameworks can be built in such a way
640that they will always be treated as being 'system frameworks', even if they are
641not present in a system framework directory. This can be useful to system
642framework developers who want to be able to test building other applications
643with development builds of their framework, including the manner in which the
644compiler changes warning behavior for system headers.</p>
645
646<p>Framework developers can opt-in to this mechanism by creating a
647'.system_framework' file at the top-level of their framework. That is, the
648framework should have contents like:</p>
649
650<pre>
651 .../TestFramework.framework
652 .../TestFramework.framework/.system_framework
653 .../TestFramework.framework/Headers
654 .../TestFramework.framework/Headers/TestFramework.h
655 ...
656</pre>
657
658<p>Clang will treat the presence of this file as an indicator that the framework
659should be treated as a system framework, regardless of how it was found in the
660framework search path. For consistency, we recommend that such files never be
661included in installed versions of the framework.</p>
662
663<!-- ======================================================================= -->
Dmitri Gribenkoccc4edf2012-05-27 14:08:44 +0000[diff] [blame]664<h2 id="availability">Availability attribute</h2>
Douglas Gregor93a70672012-03-11 04:53:21 +0000[diff] [blame]665<!-- ======================================================================= -->
666
667<p>Clang introduces the <code>availability</code> attribute, which can
668be placed on declarations to describe the lifecycle of that
669declaration relative to operating system versions. Consider the function declaration for a hypothetical function <code>f</code>:</p>
670
671<pre>
672void f(void) __attribute__((availability(macosx,introduced=10.4,deprecated=10.6,obsoleted=10.7)));
673</pre>
674
675<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>
676
677<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>
678
679<dl>
680 <dt>introduced=<i>version</i></dt>
681 <dd>The first version in which this declaration was introduced.</dd>
682
683 <dt>deprecated=<i>version</i></dt>
684 <dd>The first version in which this declaration was deprecated, meaning that users should migrate away from this API.</dd>
685
686 <dt>obsoleted=<i>version</i></dt>
687 <dd>The first version in which this declaration was obsoleted, meaning that it was removed completely and can no longer be used.</dd>
688
689 <dt>unavailable</dt>
690 <dd>This declaration is never available on this platform.</dd>
691
692 <dt>message=<i>string-literal</i></dt>
693 <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>
694</dl>
695
696<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>
697
698<dl>
699 <dt>ios</dt>
700 <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>
701
702 <dt>macosx</dt>
703 <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>
704</dl>
705
Douglas Gregor594f8412012-03-11 17:21:03 +0000[diff] [blame]706<p>A declaration can be used even when deploying back to a platform
707version prior to when the declaration was introduced. When this
708happens, the declaration is <a
709 href="https://developer.apple.com/library/mac/#documentation/MacOSX/Conceptual/BPFrameworks/Concepts/WeakLinking.html">weakly
710linked</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>
711
Douglas Gregor93a70672012-03-11 04:53:21 +0000[diff] [blame]712<!-- ======================================================================= -->
Ted Kremenek87774fd2009-12-03 02:04:01 +0000[diff] [blame]713<h2 id="checking_language_features">Checks for Standard Language Features</h2>
714<!-- ======================================================================= -->
715
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]716<p>The <tt>__has_feature</tt> macro can be used to query if certain standard
717language features are enabled. The <tt>__has_extension</tt> macro can be used
718to query if language features are available as an extension when compiling for
719a standard which does not provide them. The features which can be tested are
720listed here.</p>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000[diff] [blame]721
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]722<h3 id="cxx98">C++98</h3>
723
724<p>The features listed below are part of the C++98 standard. These features are
725enabled by default when compiling C++ code.</p>
726
727<h4 id="cxx_exceptions">C++ exceptions</h4>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000[diff] [blame]728
Ted Kremenek22c34102009-12-03 02:05:57 +0000[diff] [blame]729<p>Use <tt>__has_feature(cxx_exceptions)</tt> to determine if C++ exceptions have been enabled. For
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]730example, compiling code with <tt>-fno-exceptions</tt> disables C++ exceptions.</p>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000[diff] [blame]731
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]732<h4 id="cxx_rtti">C++ RTTI</h4>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000[diff] [blame]733
Ted Kremenek0eb95602009-12-03 02:06:43 +0000[diff] [blame]734<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 +0000[diff] [blame]735compiling code with <tt>-fno-rtti</tt> disables the use of RTTI.</p>
Ted Kremenek87774fd2009-12-03 02:04:01 +0000[diff] [blame]736
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]737<h3 id="cxx11">C++11</h3>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]738
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]739<p>The features listed below are part of the C++11 standard. As a result, all
740these features are enabled with the <tt>-std=c++11</tt> or <tt>-std=gnu++11</tt>
741option when compiling C++ code.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]742
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]743<h4 id="cxx_access_control_sfinae">C++11 SFINAE includes access control</h4>
Douglas Gregor7822ee32011-05-11 23:45:11 +0000[diff] [blame]744
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]745<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 +0000[diff] [blame]746
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]747<h4 id="cxx_alias_templates">C++11 alias templates</h4>
Richard Smith3e4c6c42011-05-05 21:57:07 +0000[diff] [blame]748
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]749<p>Use <tt>__has_feature(cxx_alias_templates)</tt> or
750<tt>__has_extension(cxx_alias_templates)</tt> to determine if support for
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]751C++11's alias declarations and alias templates is enabled.</p>
Richard Smith3e4c6c42011-05-05 21:57:07 +0000[diff] [blame]752
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]753<h4 id="cxx_alignas">C++11 alignment specifiers</h4>
Peter Collingbournefd5f6862011-10-14 23:44:46 +0000[diff] [blame]754
755<p>Use <tt>__has_feature(cxx_alignas)</tt> or
756<tt>__has_extension(cxx_alignas)</tt> to determine if support for alignment
757specifiers using <tt>alignas</tt> is enabled.</p>
758
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]759<h4 id="cxx_attributes">C++11 attributes</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]760
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]761<p>Use <tt>__has_feature(cxx_attributes)</tt> or
762<tt>__has_extension(cxx_attributes)</tt> to determine if support for attribute
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]763parsing with C++11's square bracket notation is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]764
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]765<h4 id="cxx_constexpr">C++11 generalized constant expressions</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]766
767<p>Use <tt>__has_feature(cxx_constexpr)</tt> to determine if support
768for generalized constant expressions (e.g., <tt>constexpr</tt>) is
Richard Smithb5216aa2012-02-14 22:56:17 +0000[diff] [blame]769enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]770
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]771<h4 id="cxx_decltype">C++11 <tt>decltype()</tt></h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]772
773<p>Use <tt>__has_feature(cxx_decltype)</tt> or
774<tt>__has_extension(cxx_decltype)</tt> to determine if support for the
Douglas Gregor316551f2012-04-10 20:00:33 +0000[diff] [blame]775<tt>decltype()</tt> specifier is enabled. C++11's <tt>decltype</tt>
776does not require type-completeness of a function call expression.
777Use <tt>__has_feature(cxx_decltype_incomplete_return_types)</tt>
778or <tt>__has_extension(cxx_decltype_incomplete_return_types)</tt>
779to determine if support for this feature is enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]780
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]781<h4 id="cxx_default_function_template_args">C++11 default template arguments in function templates</h4>
Douglas Gregor07508002011-02-05 20:35:30 +0000[diff] [blame]782
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]783<p>Use <tt>__has_feature(cxx_default_function_template_args)</tt> or
784<tt>__has_extension(cxx_default_function_template_args)</tt> to determine
785if support for default template arguments in function templates is enabled.</p>
Douglas Gregor07508002011-02-05 20:35:30 +0000[diff] [blame]786
Douglas Gregorf695a692011-11-01 01:19:34 +0000[diff] [blame]787<h4 id="cxx_defaulted_functions">C++11 <tt>default</tt>ed functions</h4>
788
789<p>Use <tt>__has_feature(cxx_defaulted_functions)</tt> or
790<tt>__has_extension(cxx_defaulted_functions)</tt> to determine if support for
791defaulted function definitions (with <tt>= default</tt>) is enabled.</p>
792
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]793<h4 id="cxx_delegating_constructors">C++11 delegating constructors</h4>
Sean Huntd9624992011-06-23 06:11:37 +0000[diff] [blame]794
795<p>Use <tt>__has_feature(cxx_delegating_constructors)</tt> to determine if
796support for delegating constructors is enabled.</p>
797
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]798<h4 id="cxx_deleted_functions">C++11 <tt>delete</tt>d functions</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]799
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]800<p>Use <tt>__has_feature(cxx_deleted_functions)</tt> or
801<tt>__has_extension(cxx_deleted_functions)</tt> to determine if support for
Sebastian Redlf6c09772010-08-31 23:28:47 +0000[diff] [blame]802deleted function definitions (with <tt>= delete</tt>) is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]803
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]804<h4 id="cxx_explicit_conversions">C++11 explicit conversion functions</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]805<p>Use <tt>__has_feature(cxx_explicit_conversions)</tt> to determine if support for <tt>explicit</tt> conversion functions is enabled.</p>
806
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]807<h4 id="cxx_generalized_initializers">C++11 generalized initializers</h4>
Sean Hunte1f6dea2011-08-07 00:34:32 +0000[diff] [blame]808
809<p>Use <tt>__has_feature(cxx_generalized_initializers)</tt> to determine if
810support for generalized initializers (using braced lists and
Richard Smith88189552012-02-26 07:09:21 +0000[diff] [blame]811<tt>std::initializer_list</tt>) is enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]812
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]813<h4 id="cxx_implicit_moves">C++11 implicit move constructors/assignment operators</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]814
Sebastian Redl72a81d22011-10-10 18:10:00 +0000[diff] [blame]815<p>Use <tt>__has_feature(cxx_implicit_moves)</tt> to determine if Clang will
816implicitly generate move constructors and move assignment operators where needed.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]817
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]818<h4 id="cxx_inheriting_constructors">C++11 inheriting constructors</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]819
820<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>
821
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]822<h4 id="cxx_inline_namespaces">C++11 inline namespaces</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]823
824<p>Use <tt>__has_feature(cxx_inline_namespaces)</tt> or
825<tt>__has_extension(cxx_inline_namespaces)</tt> to determine if support for
826inline namespaces is enabled.</p>
Sean Hunte1f6dea2011-08-07 00:34:32 +0000[diff] [blame]827
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]828<h4 id="cxx_lambdas">C++11 lambdas</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]829
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]830<p>Use <tt>__has_feature(cxx_lambdas)</tt> or
831<tt>__has_extension(cxx_lambdas)</tt> to determine if support for lambdas
Douglas Gregor46e021e2012-02-23 05:44:09 +0000[diff] [blame]832is enabled. </p>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]833
Douglas Gregor7b156dd2012-04-04 00:48:39 +0000[diff] [blame]834<h4 id="cxx_local_type_template_args">C++11 local and unnamed types as template arguments</h4>
835
836<p>Use <tt>__has_feature(cxx_local_type_template_args)</tt> or
837<tt>__has_extension(cxx_local_type_template_args)</tt> to determine if
838support for local and unnamed types as template arguments is enabled.</p>
839
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]840<h4 id="cxx_noexcept">C++11 noexcept</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]841
842<p>Use <tt>__has_feature(cxx_noexcept)</tt> or
843<tt>__has_extension(cxx_noexcept)</tt> to determine if support for noexcept
844exception specifications is enabled.</p>
845
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]846<h4 id="cxx_nonstatic_member_init">C++11 in-class non-static data member initialization</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]847
848<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 +0000[diff] [blame]849
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]850<h4 id="cxx_nullptr">C++11 <tt>nullptr</tt></h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]851
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]852<p>Use <tt>__has_feature(cxx_nullptr)</tt> or
853<tt>__has_extension(cxx_nullptr)</tt> to determine if support for
Douglas Gregor84ee2ee2011-05-21 23:15:46 +0000[diff] [blame]854<tt>nullptr</tt> is enabled.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]855
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]856<h4 id="cxx_override_control">C++11 <tt>override control</tt></h4>
Anders Carlssonc8b9f792011-03-25 15:04:23 +0000[diff] [blame]857
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]858<p>Use <tt>__has_feature(cxx_override_control)</tt> or
859<tt>__has_extension(cxx_override_control)</tt> to determine if support for
Anders Carlssonc8b9f792011-03-25 15:04:23 +0000[diff] [blame]860the override control keywords is enabled.</p>
861
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]862<h4 id="cxx_reference_qualified_functions">C++11 reference-qualified functions</h4>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]863<p>Use <tt>__has_feature(cxx_reference_qualified_functions)</tt> or
864<tt>__has_extension(cxx_reference_qualified_functions)</tt> to determine
865if support for reference-qualified functions (e.g., member functions with
866<code>&amp;</code> or <code>&amp;&amp;</code> applied to <code>*this</code>)
867is enabled.</p>
Douglas Gregor56209ff2011-01-26 21:25:54 +0000[diff] [blame]868
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]869<h4 id="cxx_range_for">C++11 range-based <tt>for</tt> loop</h4>
Richard Smitha391a462011-04-15 15:14:40 +0000[diff] [blame]870
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]871<p>Use <tt>__has_feature(cxx_range_for)</tt> or
872<tt>__has_extension(cxx_range_for)</tt> to determine if support for the
873range-based for loop is enabled. </p>
Richard Smitha391a462011-04-15 15:14:40 +0000[diff] [blame]874
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]875<h4 id="cxx_raw_string_literals">C++11 raw string literals</h4>
Richard Smith80134582012-03-07 08:57:31 +0000[diff] [blame]876<p>Use <tt>__has_feature(cxx_raw_string_literals)</tt> to determine if support
877for raw string literals (e.g., <tt>R"x(foo\bar)x"</tt>) is enabled.</p>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]878
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]879<h4 id="cxx_rvalue_references">C++11 rvalue references</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]880
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]881<p>Use <tt>__has_feature(cxx_rvalue_references)</tt> or
882<tt>__has_extension(cxx_rvalue_references)</tt> to determine if support for
Douglas Gregor56209ff2011-01-26 21:25:54 +0000[diff] [blame]883rvalue references is enabled. </p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]884
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]885<h4 id="cxx_static_assert">C++11 <tt>static_assert()</tt></h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]886
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]887<p>Use <tt>__has_feature(cxx_static_assert)</tt> or
888<tt>__has_extension(cxx_static_assert)</tt> to determine if support for
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]889compile-time assertions using <tt>static_assert</tt> is enabled.</p>
890
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]891<h4 id="cxx_auto_type">C++11 type inference</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]892
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]893<p>Use <tt>__has_feature(cxx_auto_type)</tt> or
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]894<tt>__has_extension(cxx_auto_type)</tt> to determine C++11 type inference is
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]895supported using the <tt>auto</tt> specifier. If this is disabled, <tt>auto</tt>
896will instead be a storage class specifier, as in C or C++98.</p>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]897
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]898<h4 id="cxx_strong_enums">C++11 strongly typed enumerations</h4>
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]899
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]900<p>Use <tt>__has_feature(cxx_strong_enums)</tt> or
901<tt>__has_extension(cxx_strong_enums)</tt> to determine if support for
902strongly typed, scoped enumerations is enabled.</p>
Sebastian Redlf6c09772010-08-31 23:28:47 +0000[diff] [blame]903
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]904<h4 id="cxx_trailing_return">C++11 trailing return type</h4>
Douglas Gregordab60ad2010-10-01 18:44:50 +0000[diff] [blame]905
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]906<p>Use <tt>__has_feature(cxx_trailing_return)</tt> or
907<tt>__has_extension(cxx_trailing_return)</tt> to determine if support for the
908alternate function declaration syntax with trailing return type is enabled.</p>
Douglas Gregordab60ad2010-10-01 18:44:50 +0000[diff] [blame]909
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]910<h4 id="cxx_unicode_literals">C++11 Unicode string literals</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]911<p>Use <tt>__has_feature(cxx_unicode_literals)</tt> to determine if
912support for Unicode string literals is enabled.</p>
Sebastian Redl4561ecd2011-03-15 21:17:12 +0000[diff] [blame]913
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]914<h4 id="cxx_unrestricted_unions">C++11 unrestricted unions</h4>
Sebastian Redl4561ecd2011-03-15 21:17:12 +0000[diff] [blame]915
Richard Smithec92bc72012-03-03 23:51:05 +0000[diff] [blame]916<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 +0000[diff] [blame]917
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]918<h4 id="cxx_user_literals">C++11 user-defined literals</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]919
Richard Smith9c1dda72012-03-09 08:41:27 +0000[diff] [blame]920<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 +0000[diff] [blame]921
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]922<h4 id="cxx_variadic_templates">C++11 variadic templates</h4>
Douglas Gregorece38942011-08-29 17:28:38 +0000[diff] [blame]923
924<p>Use <tt>__has_feature(cxx_variadic_templates)</tt> or
925<tt>__has_extension(cxx_variadic_templates)</tt> to determine if support
926for variadic templates is enabled.</p>
Douglas Gregor1274ccd2010-10-08 23:50:27 +0000[diff] [blame]927
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]928<h3 id="c11">C11</h3>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]929
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]930<p>The features listed below are part of the C11 standard. As a result, all
931these features are enabled with the <tt>-std=c11</tt> or <tt>-std=gnu11</tt>
932option when compiling C code. Additionally, because these features are all
933backward-compatible, they are available as extensions in all language modes.</p>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]934
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]935<h4 id="c_alignas">C11 alignment specifiers</h4>
Peter Collingbournefd5f6862011-10-14 23:44:46 +0000[diff] [blame]936
937<p>Use <tt>__has_feature(c_alignas)</tt> or <tt>__has_extension(c_alignas)</tt>
938to determine if support for alignment specifiers using <tt>_Alignas</tt>
939is enabled.</p>
940
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]941<h4 id="c_atomic">C11 atomic operations</h4>
942
943<p>Use <tt>__has_feature(c_atomic)</tt> or <tt>__has_extension(c_atomic)</tt>
944to determine if support for atomic types using <tt>_Atomic</tt> is enabled.
945Clang also provides <a href="#__c11_atomic">a set of builtins</a> which can be
Richard Smithc495e602012-04-19 17:46:52 +0000[diff] [blame]946used to implement the <tt>&lt;stdatomic.h&gt;</tt> operations on
947<tt>_Atomic</tt> types.</p>
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]948
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]949<h4 id="c_generic_selections">C11 generic selections</h4>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]950
951<p>Use <tt>__has_feature(c_generic_selections)</tt> or
952<tt>__has_extension(c_generic_selections)</tt> to determine if support for
953generic selections is enabled.</p>
954
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]955<p>As an extension, the C11 generic selection expression is available in all
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]956languages supported by Clang. The syntax is the same as that given in the
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]957C11 standard.</p>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]958
959<p>In C, type compatibility is decided according to the rules given in the
960appropriate standard, but in C++, which lacks the type compatibility rules
961used in C, types are considered compatible only if they are equivalent.</p>
962
Benjamin Kramerffbe9b92011-12-23 17:00:35 +0000[diff] [blame]963<h4 id="c_static_assert">C11 <tt>_Static_assert()</tt></h4>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]964
965<p>Use <tt>__has_feature(c_static_assert)</tt> or
966<tt>__has_extension(c_static_assert)</tt> to determine if support for
967compile-time assertions using <tt>_Static_assert</tt> is enabled.</p>
968
Sean Hunt4ef4c6b2010-01-13 08:31:49 +0000[diff] [blame]969<!-- ======================================================================= -->
Douglas Gregorafdf1372011-02-03 21:57:35 +0000[diff] [blame]970<h2 id="checking_type_traits">Checks for Type Traits</h2>
971<!-- ======================================================================= -->
972
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]973<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 +0000[diff] [blame]974<blockquote>
975<pre>
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]976#if __has_extension(is_convertible_to)
Douglas Gregorafdf1372011-02-03 21:57:35 +0000[diff] [blame]977template&lt;typename From, typename To&gt;
978struct is_convertible_to {
979 static const bool value = __is_convertible_to(From, To);
980};
981#else
982// Emulate type trait
983#endif
984</pre>
985</blockquote>
986
987<p>The following type traits are supported by Clang:</p>
988<ul>
989 <li><code>__has_nothrow_assign</code> (GNU, Microsoft)</li>
990 <li><code>__has_nothrow_copy</code> (GNU, Microsoft)</li>
991 <li><code>__has_nothrow_constructor</code> (GNU, Microsoft)</li>
992 <li><code>__has_trivial_assign</code> (GNU, Microsoft)</li>
993 <li><code>__has_trivial_copy</code> (GNU, Microsoft)</li>
994 <li><code>__has_trivial_constructor</code> (GNU, Microsoft)</li>
995 <li><code>__has_trivial_destructor</code> (GNU, Microsoft)</li>
996 <li><code>__has_virtual_destructor</code> (GNU, Microsoft)</li>
997 <li><code>__is_abstract</code> (GNU, Microsoft)</li>
998 <li><code>__is_base_of</code> (GNU, Microsoft)</li>
999 <li><code>__is_class</code> (GNU, Microsoft)</li>
1000 <li><code>__is_convertible_to</code> (Microsoft)</li>
1001 <li><code>__is_empty</code> (GNU, Microsoft)</li>
1002 <li><code>__is_enum</code> (GNU, Microsoft)</li>
1003 <li><code>__is_pod</code> (GNU, Microsoft)</li>
1004 <li><code>__is_polymorphic</code> (GNU, Microsoft)</li>
1005 <li><code>__is_union</code> (GNU, Microsoft)</li>
1006 <li><code>__is_literal(type)</code>: Determines whether the given type is a literal type</li>
Douglas Gregor5e9392b2011-12-03 18:14:24 +0000[diff] [blame]1007 <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 +0000[diff] [blame]1008 <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 +0000[diff] [blame]1009 <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>
1010 <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 +0000[diff] [blame]1011</ul>
1012
1013<!-- ======================================================================= -->
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1014<h2 id="blocks">Blocks</h2>
1015<!-- ======================================================================= -->
1016
Chris Lattnera7dbdf52009-03-09 07:03:22 +0000[diff] [blame]1017<p>The syntax and high level language feature description is in <a
1018href="BlockLanguageSpec.txt">BlockLanguageSpec.txt</a>. Implementation and ABI
1019details for the clang implementation are in <a
Chris Lattner5d7650b2010-03-16 21:43:03 +0000[diff] [blame]1020href="Block-ABI-Apple.txt">Block-ABI-Apple.txt</a>.</p>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1021
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]1022
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]1023<p>Query for this feature with __has_extension(blocks).</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]1024
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1025<!-- ======================================================================= -->
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]1026<h2 id="objc_features">Objective-C Features</h2>
1027<!-- ======================================================================= -->
1028
1029<h3 id="objc_instancetype">Related result types</h3>
1030
1031<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>
1032
1033<blockquote>
1034<pre>
1035@interface NSObject
1036+ (id)alloc;
1037- (id)init;
1038@end
1039
1040@interface NSArray : NSObject
1041@end
1042</pre>
1043</blockquote>
1044
1045<p>and this common initialization pattern</p>
1046
1047<blockquote>
1048<pre>
1049NSArray *array = [[NSArray alloc] init];
1050</pre>
1051</blockquote>
1052
1053<p>the type of the expression <code>[NSArray alloc]</code> is
1054<code>NSArray*</code> because <code>alloc</code> implicitly has a
1055related result type. Similarly, the type of the expression
1056<code>[[NSArray alloc] init]</code> is <code>NSArray*</code>, since
1057<code>init</code> has a related result type and its receiver is known
1058to 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>
1059
Douglas Gregore97179c2011-09-08 01:46:34 +0000[diff] [blame]1060<p>A method with a related result type can be declared by using the
1061type <tt>instancetype</tt> as its result type. <tt>instancetype</tt>
1062is a contextual keyword that is only permitted in the result type of
1063an Objective-C method, e.g.</p>
1064
1065<pre>
1066@interface A
1067+ (<b>instancetype</b>)constructAnA;
1068@end
1069</pre>
1070
1071<p>The related result type can also be inferred for some methods.
1072To determine whether a method has an inferred related result type, the first
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]1073word in the camel-case selector (e.g., "init" in "initWithObjects") is
Douglas Gregor8a0ace62011-11-03 18:33:01 +0000[diff] [blame]1074considered, and the method will have a related result type if its return
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1075type is compatible with the type of its class and if</p>
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]1076
1077<ul>
1078
1079 <li>the first word is "alloc" or "new", and the method is a class
1080 method, or</li>
1081
1082 <li>the first word is "autorelease", "init", "retain", or "self",
1083 and the method is an instance method.</li>
1084
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1085</ul>
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]1086
1087<p>If a method with a related result type is overridden by a subclass
1088method, the subclass method must also return a type that is compatible
1089with the subclass type. For example:</p>
1090
1091<blockquote>
1092<pre>
1093@interface NSString : NSObject
1094- (NSUnrelated *)init; // incorrect usage: NSUnrelated is not NSString or a superclass of NSString
1095@end
1096</pre>
1097</blockquote>
1098
1099<p>Related result types only affect the type of a message send or
1100property access via the given method. In all other respects, a method
Douglas Gregore97179c2011-09-08 01:46:34 +0000[diff] [blame]1101with a related result type is treated the same way as method that
1102returns <tt>id</tt>.</p>
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]1103
Douglas Gregoraebb6532011-09-08 17:19:31 +0000[diff] [blame]1104<p>Use <tt>__has_feature(objc_instancetype)</tt> to determine whether
1105the <tt>instancetype</tt> contextual keyword is available.</p>
1106
Douglas Gregor926df6c2011-06-11 01:09:30 +0000[diff] [blame]1107<!-- ======================================================================= -->
John McCallf85e1932011-06-15 23:02:42 +0000[diff] [blame]1108<h2 id="objc_arc">Automatic reference counting </h2>
1109<!-- ======================================================================= -->
1110
1111<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>
1112
1113<!-- ======================================================================= -->
Douglas Gregor5471bc82011-09-08 17:18:35 +0000[diff] [blame]1114<h2 id="objc_fixed_enum">Enumerations with a fixed underlying type</h2>
1115<!-- ======================================================================= -->
1116
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]1117<p>Clang provides support for C++11 enumerations with a fixed
Douglas Gregor5471bc82011-09-08 17:18:35 +0000[diff] [blame]1118underlying type within Objective-C. For example, one can write an
1119enumeration type as:</p>
1120
1121<pre>
1122typedef enum : unsigned char { Red, Green, Blue } Color;
1123</pre>
1124
1125<p>This specifies that the underlying type, which is used to store the
1126enumeration value, is <tt>unsigned char</tt>.</p>
1127
1128<p>Use <tt>__has_feature(objc_fixed_enum)</tt> to determine whether
1129support for fixed underlying types is available in Objective-C.</p>
1130
1131<!-- ======================================================================= -->
Douglas Gregor8a4e1822012-03-09 23:24:48 +0000[diff] [blame]1132<h2 id="objc_lambdas">Interoperability with C++11 lambdas</h2>
1133<!-- ======================================================================= -->
1134
1135<p>Clang provides interoperability between C++11 lambdas and
1136blocks-based APIs, by permitting a lambda to be implicitly converted
1137to a block pointer with the corresponding signature. For example,
1138consider an API such as <code>NSArray</code>'s array-sorting
1139method:</p>
1140
1141<pre> - (NSArray *)sortedArrayUsingComparator:(NSComparator)cmptr; </pre>
1142
1143<p><code>NSComparator</code> is simply a typedef for the block pointer
1144<code>NSComparisonResult (^)(id, id)</code>, and parameters of this
1145type are generally provided with block literals as arguments. However,
1146one can also use a C++11 lambda so long as it provides the same
1147signature (in this case, accepting two parameters of type
1148<code>id</code> and returning an <code>NSComparisonResult</code>):</p>
1149
1150<pre>
1151 NSArray *array = @[@"string 1", @"string 21", @"string 12", @"String 11",
1152 @"String 02"];
1153 const NSStringCompareOptions comparisonOptions
1154 = NSCaseInsensitiveSearch | NSNumericSearch |
1155 NSWidthInsensitiveSearch | NSForcedOrderingSearch;
1156 NSLocale *currentLocale = [NSLocale currentLocale];
1157 NSArray *sorted
1158 = [array sortedArrayUsingComparator:<b>[=](id s1, id s2) -&gt; NSComparisonResult {
1159 NSRange string1Range = NSMakeRange(0, [s1 length]);
1160 return [s1 compare:s2 options:comparisonOptions
1161 range:string1Range locale:currentLocale];
1162 }</b>];
1163 NSLog(@"sorted: %@", sorted);
1164</pre>
1165
1166<p>This code relies on an implicit conversion from the type of the
1167lambda expression (an unnamed, local class type called the <i>closure
1168type</i>) to the corresponding block pointer type. The conversion
1169itself is expressed by a conversion operator in that closure type
1170that produces a block pointer with the same signature as the lambda
1171itself, e.g.,</p>
1172
1173<pre>
1174 operator NSComparisonResult (^)(id, id)() const;
1175</pre>
1176
1177<p>This conversion function returns a new block that simply forwards
1178the two parameters to the lambda object (which it captures by copy),
1179then returns the result. The returned block is first copied (with
1180<tt>Block_copy</tt>) and then autoreleased. As an optimization, if a
1181lambda expression is immediately converted to a block pointer (as in
1182the first example, above), then the block is not copied and
1183autoreleased: rather, it is given the same lifetime as a block literal
1184written at that point in the program, which avoids the overhead of
1185copying a block to the heap in the common case.</p>
1186
Douglas Gregorbccda482012-03-10 22:20:11 +0000[diff] [blame]1187<p>The conversion from a lambda to a block pointer is only available
1188in Objective-C++, and not in C++ with blocks, due to its use of
1189Objective-C memory management (autorelease).</p>
1190
Douglas Gregor8a4e1822012-03-09 23:24:48 +0000[diff] [blame]1191<!-- ======================================================================= -->
Patrick Beardeb382ec2012-04-19 00:25:12 +0000[diff] [blame]1192<h2 id="objc_object_literals_subscripting">Object Literals and Subscripting</h2>
Patrick Beard62f12342012-03-20 21:51:03 +0000[diff] [blame]1193<!-- ======================================================================= -->
1194
Patrick Beardeb382ec2012-04-19 00:25:12 +0000[diff] [blame]1195<p>Clang provides support for <a href="ObjectiveCLiterals.html">Object Literals
1196and Subscripting</a> in Objective-C, which simplifies common Objective-C
1197programming patterns, makes programs more concise, and improves the safety of
1198container creation. There are several feature macros associated with object
1199literals and subscripting: <code>__has_feature(objc_array_literals)</code>
1200tests the availability of array literals;
1201<code>__has_feature(objc_dictionary_literals)</code> tests the availability of
1202dictionary literals; <code>__has_feature(objc_subscripting)</code> tests the
1203availability of object subscripting.</p>
Patrick Beard62f12342012-03-20 21:51:03 +0000[diff] [blame]1204
1205<!-- ======================================================================= -->
Fariborz Jahanian790880b2012-06-18 17:13:17 +0000[diff] [blame]1206<h2 id="objc_default_synthesize_properties">Objective-C Autosynthesis of Properties</h2>
1207<!-- ======================================================================= -->
1208
1209<p> Clang provides support for autosynthesis of declared properties. Using this
1210feature, clang provides default synthesis of those properties not declared @dynamic
1211and not having user provided backing getter and setter methods.
1212<code>__has_feature(objc_default_synthesize_properties)</code> checks for availability
1213of this feature in version of clang being used.</p>
1214
1215<!-- ======================================================================= -->
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]1216<h2 id="overloading-in-c">Function Overloading in C</h2>
1217<!-- ======================================================================= -->
1218
Chris Lattnerf161d412009-02-13 21:51:45 +0000[diff] [blame]1219<p>Clang provides support for C++ function overloading in C. Function
1220overloading in C is introduced using the <tt>overloadable</tt> attribute. For
1221example, one might provide several overloaded versions of a <tt>tgsin</tt>
1222function that invokes the appropriate standard function computing the sine of a
1223value with <tt>float</tt>, <tt>double</tt>, or <tt>long double</tt>
1224precision:</p>
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]1225
1226<blockquote>
1227<pre>
1228#include &lt;math.h&gt;
1229float <b>__attribute__((overloadable))</b> tgsin(float x) { return sinf(x); }
1230double <b>__attribute__((overloadable))</b> tgsin(double x) { return sin(x); }
1231long double <b>__attribute__((overloadable))</b> tgsin(long double x) { return sinl(x); }
1232</pre>
1233</blockquote>
1234
1235<p>Given these declarations, one can call <tt>tgsin</tt> with a
1236<tt>float</tt> value to receive a <tt>float</tt> result, with a
1237<tt>double</tt> to receive a <tt>double</tt> result, etc. Function
1238overloading in C follows the rules of C++ function overloading to pick
1239the best overload given the call arguments, with a few C-specific
1240semantics:</p>
1241<ul>
1242 <li>Conversion from <tt>float</tt> or <tt>double</tt> to <tt>long
1243 double</tt> is ranked as a floating-point promotion (per C99) rather
1244 than as a floating-point conversion (as in C++).</li>
1245
1246 <li>A conversion from a pointer of type <tt>T*</tt> to a pointer of type
1247 <tt>U*</tt> is considered a pointer conversion (with conversion
1248 rank) if <tt>T</tt> and <tt>U</tt> are compatible types.</li>
1249
1250 <li>A conversion from type <tt>T</tt> to a value of type <tt>U</tt>
1251 is permitted if <tt>T</tt> and <tt>U</tt> are compatible types. This
1252 conversion is given "conversion" rank.</li>
1253</ul>
1254
1255<p>The declaration of <tt>overloadable</tt> functions is restricted to
1256function declarations and definitions. Most importantly, if any
1257function with a given name is given the <tt>overloadable</tt>
1258attribute, then all function declarations and definitions with that
1259name (and in that scope) must have the <tt>overloadable</tt>
Chris Lattnerf161d412009-02-13 21:51:45 +0000[diff] [blame]1260attribute. This rule even applies to redeclarations of functions whose original
1261declaration had the <tt>overloadable</tt> attribute, e.g.,</p>
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]1262
1263<blockquote>
1264<pre>
1265int f(int) __attribute__((overloadable));
1266float f(float); <i>// error: declaration of "f" must have the "overloadable" attribute</i>
1267
1268int g(int) __attribute__((overloadable));
1269int g(int) { } <i>// error: redeclaration of "g" must also have the "overloadable" attribute</i>
1270</pre>
1271</blockquote>
1272
Douglas Gregor965acbb2009-02-18 07:07:28 +0000[diff] [blame]1273<p>Functions marked <tt>overloadable</tt> must have
1274prototypes. Therefore, the following code is ill-formed:</p>
1275
1276<blockquote>
1277<pre>
1278int h() __attribute__((overloadable)); <i>// error: h does not have a prototype</i>
1279</pre>
1280</blockquote>
1281
1282<p>However, <tt>overloadable</tt> functions are allowed to use a
1283ellipsis 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>
1284
1285<blockquote>
1286<pre>
Chris Lattner02246802009-02-18 22:27:46 +0000[diff] [blame]1287void honeypot(...) __attribute__((overloadable, unavailable)); <i>// calling me is an error</i>
Douglas Gregor965acbb2009-02-18 07:07:28 +0000[diff] [blame]1288</pre>
1289</blockquote>
1290
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]1291<p>Functions declared with the <tt>overloadable</tt> attribute have
1292their names mangled according to the same rules as C++ function
1293names. For example, the three <tt>tgsin</tt> functions in our
1294motivating example get the mangled names <tt>_Z5tgsinf</tt>,
Chris Lattner71b48d62010-11-28 18:19:13 +0000[diff] [blame]1295<tt>_Z5tgsind</tt>, and <tt>_Z5tgsine</tt>, respectively. There are two
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]1296caveats to this use of name mangling:</p>
1297
1298<ul>
1299
1300 <li>Future versions of Clang may change the name mangling of
1301 functions overloaded in C, so you should not depend on an specific
1302 mangling. To be completely safe, we strongly urge the use of
1303 <tt>static inline</tt> with <tt>overloadable</tt> functions.</li>
1304
1305 <li>The <tt>overloadable</tt> attribute has almost no meaning when
1306 used in C++, because names will already be mangled and functions are
1307 already overloadable. However, when an <tt>overloadable</tt>
1308 function occurs within an <tt>extern "C"</tt> linkage specification,
1309 it's name <i>will</i> be mangled in the same way as it would in
1310 C.</li>
1311</ul>
1312
Peter Collingbournec1b5fa42011-05-13 20:54:45 +0000[diff] [blame]1313<p>Query for this feature with __has_extension(attribute_overloadable).</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]1314
Eli Friedman0c706c22011-09-19 23:17:44 +0000[diff] [blame]1315<!-- ======================================================================= -->
1316<h2 id="complex-list-init">Initializer lists for complex numbers in C</h2>
1317<!-- ======================================================================= -->
1318
1319<p>clang supports an extension which allows the following in C:</p>
1320
1321<blockquote>
1322<pre>
1323#include &lt;math.h&gt;
1324#include &lt;complex.h&gt;
1325complex float x = { 1.0f, INFINITY }; // Init to (1, Inf)
1326</pre>
1327</blockquote>
1328
1329<p>This construct is useful because there is no way to separately
1330initialize the real and imaginary parts of a complex variable in
1331standard C, given that clang does not support <code>_Imaginary</code>.
1332(clang also supports the <code>__real__</code> and <code>__imag__</code>
1333extensions from gcc, which help in some cases, but are not usable in
1334static initializers.)
1335
1336<p>Note that this extension does not allow eliding the braces; the
1337meaning of the following two lines is different:</p>
1338
1339<blockquote>
1340<pre>
1341complex float x[] = { { 1.0f, 1.0f } }; // [0] = (1, 1)
1342complex float x[] = { 1.0f, 1.0f }; // [0] = (1, 0), [1] = (1, 0)
1343</pre>
1344</blockquote>
1345
1346<p>This extension also works in C++ mode, as far as that goes, but does not
1347 apply to the C++ <code>std::complex</code>. (In C++11, list
1348 initialization allows the same syntax to be used with
1349 <code>std::complex</code> with the same meaning.)
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]1350
Douglas Gregorcb54d432009-02-13 00:57:04 +0000[diff] [blame]1351<!-- ======================================================================= -->
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1352<h2 id="builtins">Builtin Functions</h2>
1353<!-- ======================================================================= -->
1354
1355<p>Clang supports a number of builtin library functions with the same syntax as
1356GCC, including things like <tt>__builtin_nan</tt>,
1357<tt>__builtin_constant_p</tt>, <tt>__builtin_choose_expr</tt>,
1358<tt>__builtin_types_compatible_p</tt>, <tt>__sync_fetch_and_add</tt>, etc. In
1359addition to the GCC builtins, Clang supports a number of builtins that GCC does
1360not, which are listed here.</p>
1361
1362<p>Please note that Clang does not and will not support all of the GCC builtins
1363for vector operations. Instead of using builtins, you should use the functions
1364defined in target-specific header files like <tt>&lt;xmmintrin.h&gt;</tt>, which
1365define portable wrappers for these. Many of the Clang versions of these
1366functions are implemented directly in terms of <a href="#vectors">extended
1367vector support</a> instead of builtins, in order to reduce the number of
1368builtins that we need to implement.</p>
1369
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1370<!-- ======================================================================= -->
Hal Finkela841c192012-08-05 22:03:08 +0000[diff] [blame^]1371<h3><a name="__builtin_readcyclecounter">__builtin_readcyclecounter</a></h3>
1372<!-- ======================================================================= -->
1373
1374<p><tt>__builtin_readcyclecounter</tt> is used to access the cycle counter
1375register (or a similar low-latency, high-accuracy clock) on those targets that
1376support it.
1377</p>
1378
1379<p><b>Syntax:</b></p>
1380
1381<pre>
1382__builtin_readcyclecounter()
1383</pre>
1384
1385<p><b>Example of Use:</b></p>
1386
1387<pre>
1388unsigned long long t0 = __builtin_readcyclecounter();
1389do_something();
1390unsigned long long t1 = __builtin_readcyclecounter();
1391unsigned long long cycles_to_do_something = t1 - t0; // assuming no overflow
1392</pre>
1393
1394<p><b>Description:</b></p>
1395
1396<p>The __builtin_readcyclecounter() builtin returns the cycle counter value,
1397which may be either global or process/thread-specific depending on the target.
1398As the backing counters often overflow quickly (on the order of
1399seconds) this should only be used for timing small intervals. When not
1400supported by the target, the return value is always zero. This builtin
1401takes no arguments and produces an unsigned long long result.
1402</p>
1403
1404<p>Query for this feature with __has_builtin(__builtin_readcyclecounter).</p>
1405
1406<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1407<h3><a name="__builtin_shufflevector">__builtin_shufflevector</a></h3>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1408<!-- ======================================================================= -->
1409
Chris Lattneraad826b2009-09-16 18:56:12 +0000[diff] [blame]1410<p><tt>__builtin_shufflevector</tt> is used to express generic vector
Chris Lattner6f72da52009-02-13 20:00:20 +0000[diff] [blame]1411permutation/shuffle/swizzle operations. This builtin is also very important for
1412the implementation of various target-specific header files like
1413<tt>&lt;xmmintrin.h&gt;</tt>.
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1414</p>
1415
1416<p><b>Syntax:</b></p>
1417
1418<pre>
Chris Lattner6f72da52009-02-13 20:00:20 +0000[diff] [blame]1419__builtin_shufflevector(vec1, vec2, index1, index2, ...)
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1420</pre>
1421
1422<p><b>Examples:</b></p>
1423
1424<pre>
Chris Lattner6f72da52009-02-13 20:00:20 +0000[diff] [blame]1425 // Identity operation - return 4-element vector V1.
1426 __builtin_shufflevector(V1, V1, 0, 1, 2, 3)
1427
1428 // "Splat" element 0 of V1 into a 4-element result.
1429 __builtin_shufflevector(V1, V1, 0, 0, 0, 0)
1430
1431 // Reverse 4-element vector V1.
1432 __builtin_shufflevector(V1, V1, 3, 2, 1, 0)
1433
1434 // Concatenate every other element of 4-element vectors V1 and V2.
1435 __builtin_shufflevector(V1, V2, 0, 2, 4, 6)
1436
1437 // Concatenate every other element of 8-element vectors V1 and V2.
1438 __builtin_shufflevector(V1, V2, 0, 2, 4, 6, 8, 10, 12, 14)
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1439</pre>
1440
1441<p><b>Description:</b></p>
1442
Chris Lattner6f72da52009-02-13 20:00:20 +0000[diff] [blame]1443<p>The first two arguments to __builtin_shufflevector are vectors that have the
1444same element type. The remaining arguments are a list of integers that specify
1445the elements indices of the first two vectors that should be extracted and
1446returned in a new vector. These element indices are numbered sequentially
1447starting with the first vector, continuing into the second vector. Thus, if
1448vec1 is a 4-element vector, index 5 would refer to the second element of vec2.
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1449</p>
1450
Chris Lattner6f72da52009-02-13 20:00:20 +0000[diff] [blame]1451<p>The result of __builtin_shufflevector is a vector
1452with the same element type as vec1/vec2 but that has an element count equal to
1453the number of indices specified.
1454</p>
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1455
Chris Lattner21190d52009-09-21 03:09:59 +0000[diff] [blame]1456<p>Query for this feature with __has_builtin(__builtin_shufflevector).</p>
1457
1458<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1459<h3><a name="__builtin_unreachable">__builtin_unreachable</a></h3>
Chris Lattner21190d52009-09-21 03:09:59 +0000[diff] [blame]1460<!-- ======================================================================= -->
1461
1462<p><tt>__builtin_unreachable</tt> is used to indicate that a specific point in
1463the program cannot be reached, even if the compiler might otherwise think it
1464can. This is useful to improve optimization and eliminates certain warnings.
1465For example, without the <tt>__builtin_unreachable</tt> in the example below,
1466the compiler assumes that the inline asm can fall through and prints a "function
1467declared 'noreturn' should not return" warning.
1468</p>
1469
1470<p><b>Syntax:</b></p>
1471
1472<pre>
1473__builtin_unreachable()
1474</pre>
1475
1476<p><b>Example of Use:</b></p>
1477
1478<pre>
1479void myabort(void) __attribute__((noreturn));
1480void myabort(void) {
1481 asm("int3");
1482 __builtin_unreachable();
1483}
1484</pre>
1485
1486<p><b>Description:</b></p>
1487
1488<p>The __builtin_unreachable() builtin has completely undefined behavior. Since
1489it has undefined behavior, it is a statement that it is never reached and the
1490optimizer can take advantage of this to produce better code. This builtin takes
1491no arguments and produces a void result.
1492</p>
1493
1494<p>Query for this feature with __has_builtin(__builtin_unreachable).</p>
1495
Chris Lattner23aa9c82011-04-09 03:57:26 +0000[diff] [blame]1496<!-- ======================================================================= -->
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1497<h3><a name="__sync_swap">__sync_swap</a></h3>
Chris Lattner23aa9c82011-04-09 03:57:26 +0000[diff] [blame]1498<!-- ======================================================================= -->
1499
1500<p><tt>__sync_swap</tt> is used to atomically swap integers or pointers in
1501memory.
1502</p>
1503
1504<p><b>Syntax:</b></p>
1505
1506<pre>
1507<i>type</i> __sync_swap(<i>type</i> *ptr, <i>type</i> value, ...)
1508</pre>
1509
1510<p><b>Example of Use:</b></p>
1511
1512<pre>
Sean Hunt7e98b472011-06-23 01:21:01 +0000[diff] [blame]1513int old_value = __sync_swap(&amp;value, new_value);
Chris Lattner23aa9c82011-04-09 03:57:26 +0000[diff] [blame]1514</pre>
1515
1516<p><b>Description:</b></p>
1517
1518<p>The __sync_swap() builtin extends the existing __sync_*() family of atomic
1519intrinsics to allow code to atomically swap the current value with the new
1520value. More importantly, it helps developers write more efficient and correct
1521code by avoiding expensive loops around __sync_bool_compare_and_swap() or
1522relying on the platform specific implementation details of
1523__sync_lock_test_and_set(). The __sync_swap() builtin is a full barrier.
1524</p>
1525
Richard Smithfafbf062012-04-11 17:55:32 +0000[diff] [blame]1526<!-- ======================================================================= -->
1527<h3><a name="__c11_atomic">__c11_atomic builtins</a></h3>
1528<!-- ======================================================================= -->
1529
1530<p>Clang provides a set of builtins which are intended to be used to implement
1531C11's <tt>&lt;stdatomic.h&gt;</tt> header. These builtins provide the semantics
1532of the <tt>_explicit</tt> form of the corresponding C11 operation, and are named
1533with a <tt>__c11_</tt> prefix. The supported operations are:</p>
1534
1535<ul>
1536 <li><tt>__c11_atomic_init</tt></li>
1537 <li><tt>__c11_atomic_thread_fence</tt></li>
1538 <li><tt>__c11_atomic_signal_fence</tt></li>
1539 <li><tt>__c11_atomic_is_lock_free</tt></li>
1540 <li><tt>__c11_atomic_store</tt></li>
1541 <li><tt>__c11_atomic_load</tt></li>
1542 <li><tt>__c11_atomic_exchange</tt></li>
1543 <li><tt>__c11_atomic_compare_exchange_strong</tt></li>
1544 <li><tt>__c11_atomic_compare_exchange_weak</tt></li>
1545 <li><tt>__c11_atomic_fetch_add</tt></li>
1546 <li><tt>__c11_atomic_fetch_sub</tt></li>
1547 <li><tt>__c11_atomic_fetch_and</tt></li>
1548 <li><tt>__c11_atomic_fetch_or</tt></li>
1549 <li><tt>__c11_atomic_fetch_xor</tt></li>
1550</ul>
1551
Richard Smithe0d3b4c2012-05-03 18:27:39 +0000[diff] [blame]1552<!-- ======================================================================= -->
1553<h2 id="non-standard-attributes">Non-standard C++11 Attributes</h2>
1554<!-- ======================================================================= -->
1555
Richard Smith207653c2012-05-03 20:05:46 +0000[diff] [blame]1556<p>Clang supports one non-standard C++11 attribute. It resides in the
1557<tt>clang</tt> attribute namespace.</p>
Richard Smithe0d3b4c2012-05-03 18:27:39 +0000[diff] [blame]1558
1559<!-- ======================================================================= -->
1560<h3 id="clang__fallthrough">The <tt>clang::fallthrough</tt> attribute</h3>
1561<!-- ======================================================================= -->
1562
Richard Smith207653c2012-05-03 20:05:46 +0000[diff] [blame]1563<p>The <tt>clang::fallthrough</tt> attribute is used along with the
1564<tt>-Wimplicit-fallthrough</tt> argument to annotate intentional fall-through
1565between switch labels. It can only be applied to a null statement placed at a
Richard Smithe0d3b4c2012-05-03 18:27:39 +0000[diff] [blame]1566point of execution between any statement and the next switch label. It is common
1567to mark these places with a specific comment, but this attribute is meant to
1568replace comments with a more strict annotation, which can be checked by the
1569compiler. This attribute doesn't change semantics of the code and can be used
Richard Smith207653c2012-05-03 20:05:46 +0000[diff] [blame]1570wherever an intended fall-through occurs. It is designed to mimic
1571control-flow statements like <tt>break;</tt>, so it can be placed in most places
1572where <tt>break;</tt> can, but only if there are no statements on the execution
1573path between it and the next switch label.</p>
Richard Smithe0d3b4c2012-05-03 18:27:39 +0000[diff] [blame]1574<p>Here is an example:</p>
1575<pre>
1576// compile with -Wimplicit-fallthrough
1577switch (n) {
1578case 33:
1579 f();
1580case 44: // warning: unannotated fall-through
1581 g();
1582 <b>[[clang::fallthrough]];</b>
1583case 55: // no warning
1584 if (x) {
1585 h();
1586 break;
1587 }
1588 else {
1589 i();
1590 <b>[[clang::fallthrough]];</b>
1591 }
1592case 66: // no warning
1593 p();
Richard Smithfff4a442012-05-03 20:10:41 +0000[diff] [blame]1594 <b>[[clang::fallthrough]];</b> // warning: fallthrough annotation does not directly precede case label
Richard Smithe0d3b4c2012-05-03 18:27:39 +0000[diff] [blame]1595 q();
1596case 77: // warning: unannotated fall-through
1597 r();
1598}
1599</pre>
Chris Lattner21190d52009-09-21 03:09:59 +0000[diff] [blame]1600
Chris Lattner1177f912009-04-09 19:58:15 +0000[diff] [blame]1601<!-- ======================================================================= -->
1602<h2 id="targetspecific">Target-Specific Extensions</h2>
1603<!-- ======================================================================= -->
1604
1605<p>Clang supports some language features conditionally on some targets.</p>
1606
1607<!-- ======================================================================= -->
1608<h3 id="x86-specific">X86/X86-64 Language Extensions</h3>
1609<!-- ======================================================================= -->
1610
1611<p>The X86 backend has these language extensions:</p>
1612
1613<!-- ======================================================================= -->
1614<h4 id="x86-gs-segment">Memory references off the GS segment</h4>
1615<!-- ======================================================================= -->
1616
1617<p>Annotating a pointer with address space #256 causes it to be code generated
Chris Lattnera021e7c2009-05-05 18:54:47 +0000[diff] [blame]1618relative to the X86 GS segment register, and address space #257 causes it to be
1619relative to the X86 FS segment. Note that this is a very very low-level
1620feature that should only be used if you know what you're doing (for example in
1621an OS kernel).</p>
Chris Lattner1177f912009-04-09 19:58:15 +0000[diff] [blame]1622
1623<p>Here is an example:</p>
1624
1625<pre>
1626#define GS_RELATIVE __attribute__((address_space(256)))
1627int foo(int GS_RELATIVE *P) {
1628 return *P;
1629}
1630</pre>
1631
1632<p>Which compiles to (on X86-32):</p>
1633
1634<pre>
1635_foo:
1636 movl 4(%esp), %eax
1637 movl %gs:(%eax), %eax
1638 ret
1639</pre>
1640
Ted Kremeneked869312009-04-10 05:03:33 +0000[diff] [blame]1641<!-- ======================================================================= -->
1642<h2 id="analyzerspecific">Static Analysis-Specific Extensions</h2>
1643<!-- ======================================================================= -->
1644
1645<p>Clang supports additional attributes that are useful for documenting program
1646invariants and rules for static analysis tools. The extensions documented here
1647are used by the <a
1648href="http://clang.llvm.org/StaticAnalysis.html">path-sensitive static analyzer
1649engine</a> that is part of Clang's Analysis library.</p>
1650
John McCall87494012011-03-18 03:51:49 +0000[diff] [blame]1651<h3 id="attr_analyzer_noreturn">The <tt>analyzer_noreturn</tt> attribute</h3>
Ted Kremeneked869312009-04-10 05:03:33 +0000[diff] [blame]1652
1653<p>Clang's static analysis engine understands the standard <tt>noreturn</tt>
Ted Kremenek4df21142009-04-10 05:04:22 +0000[diff] [blame]1654attribute. This attribute, which is typically affixed to a function prototype,
1655indicates that a call to a given function never returns. Function prototypes for
1656common functions like <tt>exit</tt> are typically annotated with this attribute,
1657as well as a variety of common assertion handlers. Users can educate the static
1658analyzer about their own custom assertion handles (thus cutting down on false
1659positives due to false paths) by marking their own &quot;panic&quot; functions
1660with this attribute.</p>
Ted Kremeneked869312009-04-10 05:03:33 +0000[diff] [blame]1661
1662<p>While useful, <tt>noreturn</tt> is not applicable in all cases. Sometimes
Nick Lewycky625b5862009-06-14 04:08:08 +0000[diff] [blame]1663there are special functions that for all intents and purposes should be
1664considered panic functions (i.e., they are only called when an internal program
1665error occurs) but may actually return so that the program can fail gracefully.
1666The <tt>analyzer_noreturn</tt> attribute allows one to annotate such functions
1667as being interpreted as &quot;no return&quot; functions by the analyzer (thus
Chris Lattner28935892009-04-10 05:54:56 +0000[diff] [blame]1668pruning bogus paths) but will not affect compilation (as in the case of
Ted Kremeneked869312009-04-10 05:03:33 +0000[diff] [blame]1669<tt>noreturn</tt>).</p>
1670
1671<p><b>Usage</b>: The <tt>analyzer_noreturn</tt> attribute can be placed in the
Chris Lattner28935892009-04-10 05:54:56 +0000[diff] [blame]1672same places where the <tt>noreturn</tt> attribute can be placed. It is commonly
Ted Kremeneked869312009-04-10 05:03:33 +0000[diff] [blame]1673placed at the end of function prototypes:</p>
1674
1675<pre>
1676 void foo() <b>__attribute__((analyzer_noreturn))</b>;
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]1677</pre>
1678
John McCall87494012011-03-18 03:51:49 +0000[diff] [blame]1679<p>Query for this feature with
1680<tt>__has_attribute(analyzer_noreturn)</tt>.</p>
Chris Lattner148772a2009-06-13 07:13:28 +0000[diff] [blame]1681
John McCall87494012011-03-18 03:51:49 +0000[diff] [blame]1682<h3 id="attr_method_family">The <tt>objc_method_family</tt> attribute</h3>
1683
1684<p>Many methods in Objective-C have conventional meanings determined
1685by their selectors. For the purposes of static analysis, it is
1686sometimes useful to be able to mark a method as having a particular
1687conventional meaning despite not having the right selector, or as not
1688having the conventional meaning that its selector would suggest.
1689For these use cases, we provide an attribute to specifically describe
1690the <q>method family</q> that a method belongs to.</p>
1691
1692<p><b>Usage</b>: <tt>__attribute__((objc_method_family(X)))</tt>,
1693where <tt>X</tt> is one of <tt>none</tt>, <tt>alloc</tt>, <tt>copy</tt>,
1694<tt>init</tt>, <tt>mutableCopy</tt>, or <tt>new</tt>. This attribute
1695can only be placed at the end of a method declaration:</p>
1696
1697<pre>
1698 - (NSString*) initMyStringValue <b>__attribute__((objc_method_family(none)))</b>;
1699</pre>
1700
1701<p>Users who do not wish to change the conventional meaning of a
1702method, and who merely want to document its non-standard retain and
1703release semantics, should use the
1704<a href="#attr_retain_release">retaining behavior attributes</a>
1705described below.</p>
1706
1707<p>Query for this feature with
1708<tt>__has_attribute(objc_method_family)</tt>.</p>
1709
1710<h3 id="attr_retain_release">Objective-C retaining behavior attributes</h3>
John McCall630b7ae2011-01-25 04:26:21 +0000[diff] [blame]1711
1712<p>In Objective-C, functions and methods are generally assumed to take
1713and return objects with +0 retain counts, with some exceptions for
1714special methods like <tt>+alloc</tt> and <tt>init</tt>. However,
1715there are exceptions, and so Clang provides attributes to allow these
1716exceptions to be documented, which helps the analyzer find leaks (and
John McCall87494012011-03-18 03:51:49 +0000[diff] [blame]1717ignore non-leaks). Some exceptions may be better described using
1718the <a href="#attr_method_family"><tt>objc_method_family</tt></a>
1719attribute instead.</p>
John McCall630b7ae2011-01-25 04:26:21 +0000[diff] [blame]1720
1721<p><b>Usage</b>: The <tt>ns_returns_retained</tt>, <tt>ns_returns_not_retained</tt>,
1722<tt>ns_returns_autoreleased</tt>, <tt>cf_returns_retained</tt>,
1723and <tt>cf_returns_not_retained</tt> attributes can be placed on
1724methods and functions that return Objective-C or CoreFoundation
1725objects. They are commonly placed at the end of a function prototype
1726or method declaration:</p>
1727
1728<pre>
1729 id foo() <b>__attribute__((ns_returns_retained))</b>;
1730
1731 - (NSString*) bar: (int) x <b>__attribute__((ns_returns_retained))</b>;
1732</pre>
1733
1734<p>The <tt>*_returns_retained</tt> attributes specify that the
1735returned object has a +1 retain count.
1736The <tt>*_returns_not_retained</tt> attributes specify that the return
1737object has a +0 retain count, even if the normal convention for its
1738selector would be +1. <tt>ns_returns_autoreleased</tt> specifies that the
1739returned object is +0, but is guaranteed to live at least as long as the
1740next flush of an autorelease pool.</p>
1741
1742<p><b>Usage</b>: The <tt>ns_consumed</tt> and <tt>cf_consumed</tt>
1743attributes can be placed on an parameter declaration; they specify
1744that the argument is expected to have a +1 retain count, which will be
1745balanced in some way by the function or method.
1746The <tt>ns_consumes_self</tt> attribute can only be placed on an
1747Objective-C method; it specifies that the method expects
1748its <tt>self</tt> parameter to have a +1 retain count, which it will
1749balance in some way.</p>
1750
1751<pre>
1752 void <b>foo(__attribute__((ns_consumed))</b> NSString *string);
1753
1754 - (void) bar <b>__attribute__((ns_consumes_self))</b>;
1755 - (void) baz: (id) <b>__attribute__((ns_consumed))</b> x;
1756</pre>
Ted Kremeneked869312009-04-10 05:03:33 +0000[diff] [blame]1757
John McCall87494012011-03-18 03:51:49 +0000[diff] [blame]1758<p>Query for these features with <tt>__has_attribute(ns_consumed)</tt>,
1759<tt>__has_attribute(ns_returns_retained)</tt>, etc.</p>
1760
Kostya Serebryanyce98c9b2011-11-28 20:51:02 +0000[diff] [blame]1761<!-- ======================================================================= -->
1762<h2 id="dynamicanalyzerspecific">Dynamic Analysis-Specific Extensions</h2>
1763<!-- ======================================================================= -->
1764<h3 id="address_sanitizer">AddressSanitizer</h3>
1765<p> Use <code>__has_feature(address_sanitizer)</code>
1766to check if the code is being built with <a
1767 href="AddressSanitizer.html">AddressSanitizer</a>.
1768</p>
Kostya Serebryany71efba02012-01-24 19:25:38 +0000[diff] [blame]1769<p>Use <tt>__attribute__((no_address_safety_analysis))</tt> on a function
1770declaration to specify that address safety instrumentation (e.g.
1771AddressSanitizer) should not be applied to that function.
1772</p>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]1773
1774<!-- ======================================================================= -->
Caitlin Sadowski73cbbc82011-07-28 18:38:36 +0000[diff] [blame]1775<h2 id="threadsafety">Thread-Safety Annotation Checking</h2>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]1776<!-- ======================================================================= -->
1777
1778<p>Clang supports additional attributes for checking basic locking policies in
1779multithreaded programs.
1780Clang currently parses the following list of attributes, although
1781<b>the implementation for these annotations is currently in development.</b>
1782For more details, see the
1783<a href="http://gcc.gnu.org/wiki/ThreadSafetyAnnotation">GCC implementation</a>.
1784</p>
1785
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1786<h4 id="ts_noanal">no_thread_safety_analysis</h4>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]1787
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1788<p>Use <tt>__attribute__((no_thread_safety_analysis))</tt> on a function
1789declaration to specify that the thread safety analysis should not be run on that
1790function. This attribute provides an escape hatch (e.g. for situations when it
1791is difficult to annotate the locking policy). </p>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]1792
1793<h4 id="ts_lockable">lockable</h4>
1794
1795<p>Use <tt>__attribute__((lockable))</tt> on a class definition to specify
1796that it has a lockable type (e.g. a Mutex class). This annotation is primarily
1797used to check consistency.</p>
1798
1799<h4 id="ts_scopedlockable">scoped_lockable</h4>
1800
1801<p>Use <tt>__attribute__((scoped_lockable))</tt> on a class definition to
1802specify that it has a "scoped" lockable type. Objects of this type will acquire
1803the lock upon construction and release it upon going out of scope.
1804 This annotation is primarily used to check
1805consistency.</p>
1806
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1807<h4 id="ts_guardedvar">guarded_var</h4>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]1808
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1809<p>Use <tt>__attribute__((guarded_var))</tt> on a variable declaration to
1810specify that the variable must be accessed while holding some lock.</p>
1811
1812<h4 id="ts_ptguardedvar">pt_guarded_var</h4>
1813
1814<p>Use <tt>__attribute__((pt_guarded_var))</tt> on a pointer declaration to
1815specify that the pointer must be dereferenced while holding some lock.</p>
1816
1817<h4 id="ts_guardedby">guarded_by(l)</h4>
1818
1819<p>Use <tt>__attribute__((guarded_by(l)))</tt> on a variable declaration to
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1820specify that the variable must be accessed while holding lock <tt>l</tt>.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1821
1822<h4 id="ts_ptguardedby">pt_guarded_by(l)</h4>
1823
1824<p>Use <tt>__attribute__((pt_guarded_by(l)))</tt> on a pointer declaration to
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1825specify that the pointer must be dereferenced while holding lock <tt>l</tt>.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1826
1827<h4 id="ts_acquiredbefore">acquired_before(...)</h4>
1828
1829<p>Use <tt>__attribute__((acquired_before(...)))</tt> on a declaration
1830of a lockable variable to specify that the lock must be acquired before all
1831attribute arguments. Arguments must be lockable type, and there must be at
1832least one argument.</p>
1833
1834<h4 id="ts_acquiredafter">acquired_after(...)</h4>
1835
1836<p>Use <tt>__attribute__((acquired_after(...)))</tt> on a declaration
1837of a lockable variable to specify that the lock must be acquired after all
1838attribute arguments. Arguments must be lockable type, and there must be at
1839least one argument.</p>
1840
1841<h4 id="ts_elf">exclusive_lock_function(...)</h4>
1842
1843<p>Use <tt>__attribute__((exclusive_lock_function(...)))</tt> on a function
1844declaration to specify that the function acquires all listed locks
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1845exclusively. This attribute takes zero or more arguments: either of lockable
1846type or integers indexing into function parameters of lockable type. If no
1847arguments are given, the acquired lock is implicitly <tt>this</tt> of the
1848enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1849
1850<h4 id="ts_slf">shared_lock_function(...)</h4>
1851
1852<p>Use <tt>__attribute__((shared_lock_function(...)))</tt> on a function
1853declaration to specify that the function acquires all listed locks, although
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1854 the locks may be shared (e.g. read locks). This attribute takes zero or more
1855arguments: either of lockable type or integers indexing into function
1856parameters of lockable type. If no arguments are given, the acquired lock is
1857implicitly <tt>this</tt> of the enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1858
1859<h4 id="ts_etf">exclusive_trylock_function(...)</h4>
1860
1861<p>Use <tt>__attribute__((exclusive_lock_function(...)))</tt> on a function
1862declaration to specify that the function will try (without blocking) to acquire
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1863all listed locks exclusively. This attribute takes one or more arguments. The
1864first argument is an integer or boolean value specifying the return value of a
1865successful lock acquisition. The remaining arugments are either of lockable type
1866or integers indexing into function parameters of lockable type. If only one
1867argument is given, the acquired lock is implicitly <tt>this</tt> of the
1868enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1869
1870<h4 id="ts_stf">shared_trylock_function(...)</h4>
1871
1872<p>Use <tt>__attribute__((shared_lock_function(...)))</tt> on a function
1873declaration to specify that the function will try (without blocking) to acquire
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1874all listed locks, although the locks may be shared (e.g. read locks). This
1875attribute takes one or more arguments. The first argument is an integer or
1876boolean value specifying the return value of a successful lock acquisition. The
1877remaining arugments are either of lockable type or integers indexing into
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1878function parameters of lockable type. If only one argument is given, the
1879acquired lock is implicitly <tt>this</tt> of the enclosing object.</p>
1880
1881<h4 id="ts_uf">unlock_function(...)</h4>
1882
1883<p>Use <tt>__attribute__((unlock_function(...)))</tt> on a function
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1884declaration to specify that the function release all listed locks. This
1885attribute takes zero or more arguments: either of lockable type or integers
1886indexing into function parameters of lockable type. If no arguments are given,
1887the acquired lock is implicitly <tt>this</tt> of the enclosing object.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1888
1889<h4 id="ts_lr">lock_returned(l)</h4>
1890
1891<p>Use <tt>__attribute__((lock_returned(l)))</tt> on a function
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1892declaration to specify that the function returns lock <tt>l</tt> (<tt>l</tt>
1893must be of lockable type). This annotation is used to aid in resolving lock
1894expressions.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1895
1896<h4 id="ts_le">locks_excluded(...)</h4>
1897
1898<p>Use <tt>__attribute__((locks_excluded(...)))</tt> on a function declaration
Caitlin Sadowskib51e0312011-08-09 17:59:31 +0000[diff] [blame]1899to specify that the function must not be called with the listed locks. Arguments
1900must be lockable type, and there must be at least one argument.</p>
Caitlin Sadowskidb33e142011-07-28 20:12:35 +0000[diff] [blame]1901
1902<h4 id="ts_elr">exclusive_locks_required(...)</h4>
1903
1904<p>Use <tt>__attribute__((exclusive_locks_required(...)))</tt> on a function
1905declaration to specify that the function must be called while holding the listed
1906exclusive locks. Arguments must be lockable type, and there must be at
1907least one argument.</p>
1908
1909<h4 id="ts_slr">shared_locks_required(...)</h4>
1910
1911<p>Use <tt>__attribute__((shared_locks_required(...)))</tt> on a function
1912declaration to specify that the function must be called while holding the listed
1913shared locks. Arguments must be lockable type, and there must be at
1914least one argument.</p>
Caitlin Sadowskifdde9e72011-07-28 17:21:07 +0000[diff] [blame]1915
Chris Lattner5ce933f2009-02-09 08:46:11 +0000[diff] [blame]1916</div>
1917</body>
1918</html>