Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / clang / cae40c4d448529fe65de3d87bdbe97d3b54b4d98 / . / docs / UsersManual.html
blob: 2c5139de79c6face776b3eacea4442e259506c9d [file] [log] [blame]
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]3<html>
4<head>
5<title>Clang Compiler User's Manual</title>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]6<link type="text/css" rel="stylesheet" href="../menu.css">
7<link type="text/css" rel="stylesheet" href="../content.css">
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]8<style type="text/css">
9td {
10 vertical-align: top;
11}
12</style>
13</head>
14<body>
15
16<!--#include virtual="../menu.html.incl"-->
17
18<div id="content">
19
20<h1>Clang Compiler User's Manual</h1>
21
22<ul>
23<li><a href="#intro">Introduction</a>
24 <ul>
25 <li><a href="#terminology">Terminology</a></li>
26 <li><a href="#basicusage">Basic Usage</a></li>
27 </ul>
28</li>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]29<li><a href="#commandline">Command Line Options</a>
30 <ul>
31 <li><a href="#cl_diagnostics">Options to Control Error and Warning
32 Messages</a></li>
Chad Rosierb211a4b2012-02-22 19:14:46 +0000[diff] [blame]33 <li><a href="#cl_crash_diagnostics">Options to Control Clang Crash
34 Diagnostics</a></li>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]35 </ul>
36</li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]37<li><a href="#general_features">Language and Target-Independent Features</a>
38 <ul>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]39 <li><a href="#diagnostics">Controlling Errors and Warnings</a>
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]40 <ul>
41 <li><a href="#diagnostics_display">Controlling How Clang Displays Diagnostics</a></li>
42 <li><a href="#diagnostics_mappings">Diagnostic Mappings</a></li>
Chris Lattner67db8cd2010-05-30 23:42:51 +0000[diff] [blame]43 <li><a href="#diagnostics_categories">Diagnostic Categories</a></li>
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]44 <li><a href="#diagnostics_commandline">Controlling Diagnostics via Command Line Flags</a></li>
45 <li><a href="#diagnostics_pragmas">Controlling Diagnostics via Pragmas</a></li>
Ted Kremenekf391fa72011-08-18 01:17:05 +0000[diff] [blame]46 <li><a href="#diagnostics_enable_everything">Enabling All Warnings</a></li>
Ted Kremenek2fb11eb2010-08-24 18:12:35 +0000[diff] [blame]47 <li><a href="#analyzer_diagnositics">Controlling Static Analyzer Diagnostics</a></li>
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]48 </ul>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]49 </li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]50 <li><a href="#precompiledheaders">Precompiled Headers</a></li>
Mike Stump53664ca2009-12-14 23:53:10 +0000[diff] [blame]51 <li><a href="#codegen">Controlling Code Generation</a></li>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]52 </ul>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]53</li>
54<li><a href="#c">C Language Features</a>
55 <ul>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]56 <li><a href="#c_ext">Extensions supported by clang</a></li>
57 <li><a href="#c_modes">Differences between various standard modes</a></li>
58 <li><a href="#c_unimpl_gcc">GCC extensions not implemented yet</a></li>
59 <li><a href="#c_unsupp_gcc">Intentionally unsupported GCC extensions</a></li>
60 <li><a href="#c_ms">Microsoft extensions</a></li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]61 </ul>
62</li>
Richard Smithbdb10182011-11-21 20:54:59 +0000[diff] [blame]63<li><a href="#cxx">C++ Language Features</a>
64 <ul>
65 <li><a href="#cxx_implimits">Controlling implementation limits</a></li>
66 </ul>
67</li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]68<li><a href="#target_features">Target-Specific Features and Limitations</a>
69 <ul>
70 <li><a href="#target_arch">CPU Architectures Features and Limitations</a>
71 <ul>
72 <li><a href="#target_arch_x86">X86</a></li>
Eli Friedman3b658d32009-06-08 05:12:39 +0000[diff] [blame]73 <li><a href="#target_arch_arm">ARM</a></li>
74 <li><a href="#target_arch_other">Other platforms</a></li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]75 </ul>
76 </li>
77 <li><a href="#target_os">Operating System Features and Limitations</a>
78 <ul>
79 <li><a href="#target_os_darwin">Darwin (Mac OS/X)</a></li>
80 <li>Linux, etc.</li>
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]81 <li><a href="#target_os_win32">Windows</a></li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]82 </ul>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]83 </li>
84 </ul>
85</li>
86</ul>
87
88
89<!-- ======================================================================= -->
90<h2 id="intro">Introduction</h2>
91<!-- ======================================================================= -->
92
93<p>The Clang Compiler is an open-source compiler for the C family of programming
94languages, aiming to be the best in class implementation of these languages.
95Clang builds on the LLVM optimizer and code generator, allowing it to provide
96high-quality optimization and code generation support for many targets. For
97more general information, please see the <a href="http://clang.llvm.org">Clang
98Web Site</a> or the <a href="http://llvm.org">LLVM Web Site</a>.</p>
99
100<p>This document describes important notes about using Clang as a compiler for
101an end-user, documenting the supported features, command line options, etc. If
102you are interested in using Clang to build a tool that processes code, please
103see <a href="InternalsManual.html">the Clang Internals Manual</a>. If you are
Michael J. Spencer8275fc02012-02-12 09:11:39 +0000[diff] [blame]104interested in the <a href="http://clang-analyzer.llvm.org">Clang
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]105Static Analyzer</a>, please see its web page.</p>
106
107<p>Clang is designed to support the C family of programming languages, which
108includes <a href="#c">C</a>, <a href="#objc">Objective-C</a>, <a
109href="#cxx">C++</a>, and <a href="#objcxx">Objective-C++</a> as well as many
110dialects of those. For language-specific information, please see the
111corresponding language specific section:</p>
112
113<ul>
114<li><a href="#c">C Language</a>: K&amp;R C, ANSI C89, ISO C90, ISO C94
115 (C89+AMD1), ISO C99 (+TC1, TC2, TC3). </li>
116<li><a href="#objc">Objective-C Language</a>: ObjC 1, ObjC 2, ObjC 2.1, plus
117 variants depending on base language.</li>
Richard Smithbdb10182011-11-21 20:54:59 +0000[diff] [blame]118<li><a href="#cxx">C++ Language</a></li>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]119<li><a href="#objcxx">Objective C++ Language</a></li>
120</ul>
121
122<p>In addition to these base languages and their dialects, Clang supports a
123broad variety of language extensions, which are documented in the corresponding
124language section. These extensions are provided to be compatible with the GCC,
125Microsoft, and other popular compilers as well as to improve functionality
126through Clang-specific features. The Clang driver and language features are
127intentionally designed to be as compatible with the GNU GCC compiler as
128reasonably possible, easing migration from GCC to Clang. In most cases, code
129"just works".</p>
130
131<p>In addition to language specific features, Clang has a variety of features
132that depend on what CPU architecture or operating system is being compiled for.
Douglas Gregorcd5a5052009-11-09 15:15:41 +0000[diff] [blame]133Please see the <a href="#target_features">Target-Specific Features and
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]134Limitations</a> section for more details.</p>
135
136<p>The rest of the introduction introduces some basic <a
137href="#terminology">compiler terminology</a> that is used throughout this manual
138and contains a basic <a href="#basicusage">introduction to using Clang</a>
139as a command line compiler.</p>
140
141<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
142<h3 id="terminology">Terminology</h3>
143<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
144
145<p>Front end, parser, backend, preprocessor, undefined behavior, diagnostic,
146 optimizer</p>
147
148<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
149<h3 id="basicusage">Basic Usage</h3>
150<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
151
152<p>Intro to how to use a C compiler for newbies.</p>
153<p>
154compile + link
155
156compile then link
157
158debug info
159
160enabling optimizations
161
162picking a language to use, defaults to C99 by default. Autosenses based on
163extension.
164
165using a makefile
166</p>
167
168
169<!-- ======================================================================= -->
170<h2 id="commandline">Command Line Options</h2>
171<!-- ======================================================================= -->
172
173<p>
174This section is generally an index into other sections. It does not go into
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]175depth on the ones that are covered by other sections. However, the first part
176introduces the language selection and other high level options like -c, -g, etc.
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]177</p>
178
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]179
180<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
181<h3 id="cl_diagnostics">Options to Control Error and Warning Messages</h3>
182<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
183
184<p><b>-Werror</b>: Turn warnings into errors.</p>
185<p><b>-Werror=foo</b>: Turn warning "foo" into an error.</p>
186<p><b>-Wno-error=foo</b>: Turn warning "foo" into an warning even if -Werror is
187 specified.</p>
188<p><b>-Wfoo</b>: Enable warning foo</p>
189<p><b>-Wno-foo</b>: Disable warning foo</p>
190<p><b>-w</b>: Disable all warnings.</p>
191<p><b>-pedantic</b>: Warn on language extensions.</p>
192<p><b>-pedantic-errors</b>: Error on language extensions.</p>
193<p><b>-Wsystem-headers</b>: Enable warnings from system headers.</p>
194
Chris Lattner0f0c9632010-04-07 20:49:23 +0000[diff] [blame]195<p><b>-ferror-limit=123</b>: Stop emitting diagnostics after 123 errors have
196 been produced. The default is 20, and the error limit can be disabled with
197 -ferror-limit=0.</p>
198
Douglas Gregor575cf372010-04-20 07:18:24 +0000[diff] [blame]199<p><b>-ftemplate-backtrace-limit=123</b>: Only emit up to 123 template instantiation notes within the template instantiation backtrace for a single warning or error. The default is 10, and the limit can be disabled with -ftemplate-backtrace-limit=0.</p>
200
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]201<!-- ================================================= -->
202<h4 id="cl_diag_formatting">Formatting of Diagnostics</h4>
203<!-- ================================================= -->
204
205<p>Clang aims to produce beautiful diagnostics by default, particularly for new
206users that first come to Clang. However, different people have different
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]207preferences, and sometimes Clang is driven by another program that wants to
208parse simple and consistent output, not a person. For these cases, Clang
209provides a wide range of options to control the exact output format of the
210diagnostics that it generates.</p>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]211
212<dl>
213
214<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
215<dt id="opt_fshow-column"><b>-f[no-]show-column</b>: Print column number in
216diagnostic.</dt>
217<dd>This option, which defaults to on, controls whether or not Clang prints the
218column number of a diagnostic. For example, when this is enabled, Clang will
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]219print something like:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]220
221<pre>
222 test.c:28:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
223 #endif bad
224 ^
225 //
226</pre>
227
228<p>When this is disabled, Clang will print "test.c:28: warning..." with no
229column number.</p>
230</dd>
231
232<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
233<dt id="opt_fshow-source-location"><b>-f[no-]show-source-location</b>: Print
234source file/line/column information in diagnostic.</dt>
235<dd>This option, which defaults to on, controls whether or not Clang prints the
236filename, line number and column number of a diagnostic. For example,
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]237when this is enabled, Clang will print something like:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]238
239<pre>
240 test.c:28:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
241 #endif bad
242 ^
243 //
244</pre>
245
246<p>When this is disabled, Clang will not print the "test.c:28:8: " part.</p>
247</dd>
248
249<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
250<dt id="opt_fcaret-diagnostics"><b>-f[no-]caret-diagnostics</b>: Print source
251line and ranges from source code in diagnostic.</dt>
252<dd>This option, which defaults to on, controls whether or not Clang prints the
253source line, source ranges, and caret when emitting a diagnostic. For example,
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]254when this is enabled, Clang will print something like:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]255
256<pre>
257 test.c:28:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
258 #endif bad
259 ^
260 //
261</pre>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]262</dd>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]263<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
Douglas Gregore8d44dd2010-07-09 16:31:58 +0000[diff] [blame]264<dt id="opt_fcolor_diagnostics"><b>-f[no-]color-diagnostics</b>: </dt>
265<dd>This option, which defaults to on when a color-capable terminal is
266 detected, controls whether or not Clang prints diagnostics in color.
267 When this option is enabled, Clang will use colors to highlight
268 specific parts of the diagnostic, e.g.,
Chris Lattner4e1c53d2011-01-24 03:47:34 +0000[diff] [blame]269 <pre>
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]270 <b><span style="color:black">test.c:28:8: <span style="color:magenta">warning</span>: extra tokens at end of #endif directive [-Wextra-tokens]</span></b>
Chris Lattner4e1c53d2011-01-24 03:47:34 +0000[diff] [blame]271 #endif bad
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]272 <span style="color:green">^</span>
273 <span style="color:green">//</span>
Douglas Gregore8d44dd2010-07-09 16:31:58 +0000[diff] [blame]274</pre>
Chris Lattner4e1c53d2011-01-24 03:47:34 +0000[diff] [blame]275
276<p>When this is disabled, Clang will just print:</p>
277
278<pre>
279 test.c:2:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
280 #endif bad
281 ^
282 //
283</pre>
284</dd>
Douglas Gregore8d44dd2010-07-09 16:31:58 +0000[diff] [blame]285<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
Douglas Gregorc9471b02011-05-21 17:07:29 +0000[diff] [blame]286<dt id="opt_fdiagnostics-format"><b>-fdiagnostics-format=clang/msvc/vi</b>:
287Changes diagnostic output format to better match IDEs and command line tools.</dt>
288<dd>This option controls the output format of the filename, line number, and column printed in diagnostic messages. The options, and their affect on formatting a simple conversion diagnostic, follow:
289
290 <dl>
291 <dt><b>clang</b> (default)</dt>
292 <dd>
293 <pre>t.c:3:11: warning: conversion specifies type 'char *' but the argument has type 'int'</pre>
294 </dd>
295
296 <dt><b>msvc</b></dt>
297 <dd>
298 <pre>t.c(3,11) : warning: conversion specifies type 'char *' but the argument has type 'int'</pre>
299 </dd>
300
301 <dt><b>vi</b></dt>
302 <dd>
303 <pre>t.c +3:11: warning: conversion specifies type 'char *' but the argument has type 'int'</pre>
304 </dd>
305 </dl>
306</dd>
307
308<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
Argyrios Kyrtzidis477aab62011-05-25 05:05:01 +0000[diff] [blame]309<dt id="opt_fdiagnostics-show-name"><b>-f[no-]diagnostics-show-name</b>:
310Enable the display of the diagnostic name.</dt>
311<dd>This option, which defaults to off, controls whether or not
Benjamin Kramer665a8dc2012-01-15 15:26:07 +0000[diff] [blame]312Clang prints the associated name.<p></p></dd>
Argyrios Kyrtzidis477aab62011-05-25 05:05:01 +0000[diff] [blame]313<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]314<dt id="opt_fdiagnostics-show-option"><b>-f[no-]diagnostics-show-option</b>:
315Enable <tt>[-Woption]</tt> information in diagnostic line.</dt>
316<dd>This option, which defaults to on,
317controls whether or not Clang prints the associated <A
318href="#cl_diag_warning_groups">warning group</a> option name when outputting
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]319a warning diagnostic. For example, in this output:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]320
321<pre>
322 test.c:28:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
323 #endif bad
324 ^
325 //
326</pre>
327
328<p>Passing <b>-fno-diagnostics-show-option</b> will prevent Clang from printing
329the [<a href="#opt_Wextra-tokens">-Wextra-tokens</a>] information in the
330diagnostic. This information tells you the flag needed to enable or disable the
331diagnostic, either from the command line or through <a
332href="#pragma_GCC_diagnostic">#pragma GCC diagnostic</a>.</dd>
333
Chris Lattner28a43a42010-05-05 01:35:28 +0000[diff] [blame]334<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
335<dt id="opt_fdiagnostics-show-category"><b>-fdiagnostics-show-category=none/id/name</b>:
336Enable printing category information in diagnostic line.</dt>
337<dd>This option, which defaults to "none",
338controls whether or not Clang prints the category associated with a diagnostic
339when emitting it. Each diagnostic may or many not have an associated category,
340if it has one, it is listed in the diagnostic categorization field of the
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]341diagnostic line (in the []'s).
Chris Lattner28a43a42010-05-05 01:35:28 +0000[diff] [blame]342
343<p>For example, a format string warning will produce these three renditions
344based on the setting of this option:</p>
345
346<pre>
347 t.c:3:11: warning: conversion specifies type 'char *' but the argument has type 'int' [-Wformat]
348 t.c:3:11: warning: conversion specifies type 'char *' but the argument has type 'int' [-Wformat<b>,1</b>]
349 t.c:3:11: warning: conversion specifies type 'char *' but the argument has type 'int' [-Wformat<b>,Format String</b>]
350</pre>
351
352<p>This category can be used by clients that want to group diagnostics by
353category, so it should be a high level category. We want dozens of these, not
354hundreds or thousands of them.</p>
355</dd>
356
357
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]358
359<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
360<dt id="opt_fdiagnostics-fixit-info"><b>-f[no-]diagnostics-fixit-info</b>:
361Enable "FixIt" information in the diagnostics output.</dt>
362<dd>This option, which defaults to on, controls whether or not Clang prints the
363information on how to fix a specific diagnostic underneath it when it knows.
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]364For example, in this output:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]365
366<pre>
367 test.c:28:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
368 #endif bad
369 ^
370 //
371</pre>
372
373<p>Passing <b>-fno-diagnostics-fixit-info</b> will prevent Clang from printing
374the "//" line at the end of the message. This information is useful for users
375who may not understand what is wrong, but can be confusing for machine
376parsing.</p>
377</dd>
378
379<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
Chris Lattner2a9cc232009-04-21 05:35:32 +0000[diff] [blame]380<dt id="opt_fdiagnostics-print-source-range-info">
381<b>-f[no-]diagnostics-print-source-range-info</b>:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]382Print machine parsable information about source ranges.</dt>
383<dd>This option, which defaults to off, controls whether or not Clang prints
384information about source ranges in a machine parsable format after the
385file/line/column number information. The information is a simple sequence of
386brace enclosed ranges, where each range lists the start and end line/column
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]387locations. For example, in this output:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]388
389<pre>
390exprs.c:47:15:{47:8-47:14}{47:17-47:24}: error: invalid operands to binary expression ('int *' and '_Complex float')
391 P = (P-42) + Gamma*4;
392 ~~~~~~ ^ ~~~~~~~
393</pre>
394
Chris Lattner2a9cc232009-04-21 05:35:32 +0000[diff] [blame]395<p>The {}'s are generated by -fdiagnostics-print-source-range-info.</p>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]396</dd>
397
Douglas Gregor4786c152010-08-19 20:24:43 +0000[diff] [blame]398<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
399<dt id="opt_fdiagnostics-parseable-fixits">
400<b>-fdiagnostics-parseable-fixits</b>:
401Print Fix-Its in a machine parseable form.</dt>
402<dd><p>This option makes Clang print available Fix-Its in a machine parseable format at the end of diagnostics. The following example illustrates the format:</p>
403
404<pre>
Douglas Gregorbf5e09d2010-08-20 03:17:33 +0000[diff] [blame]405 fix-it:"t.cpp":{7:25-7:29}:"Gamma"
Douglas Gregor4786c152010-08-19 20:24:43 +0000[diff] [blame]406</pre>
407
NAKAMURA Takumi44626362011-04-05 00:57:02 +0000[diff] [blame]408<p>The range printed is a half-open range, so in this example the characters at
409column 25 up to but not including column 29 on line 7 in t.cpp should be
410replaced with the string &quot;Gamma&quot;. Either the range or the replacement
411string may be empty (representing strict insertions and strict erasures,
412respectively). Both the file name and the insertion string escape backslash (as
413&quot;\\&quot;), tabs (as &quot;\t&quot;), newlines (as &quot;\n&quot;), double
414quotes(as &quot;\&quot;&quot;) and non-printable characters (as octal
415&quot;\xxx&quot;).</p>
Douglas Gregor4786c152010-08-19 20:24:43 +0000[diff] [blame]416</dd>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]417
418</dl>
419
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]420
421
422<!-- ===================================================== -->
423<h4 id="cl_diag_warning_groups">Individual Warning Groups</h4>
424<!-- ===================================================== -->
425
426<p>TODO: Generate this from tblgen. Define one anchor per warning group.</p>
427
428
429<dl>
430
431
432<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
433<dt id="opt_Wextra-tokens"><b>-Wextra-tokens</b>: Warn about excess tokens at
434 the end of a preprocessor directive.</dt>
435<dd>This option, which defaults to on, enables warnings about extra tokens at
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]436the end of preprocessor directives. For example:
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]437
438<pre>
439 test.c:28:8: warning: extra tokens at end of #endif directive [-Wextra-tokens]
440 #endif bad
441 ^
442</pre>
443
444<p>These extra tokens are not strictly conforming, and are usually best handled
445by commenting them out.</p>
446
447<p>This option is also enabled by <a href="">-Wfoo</a>, <a href="">-Wbar</a>,
448 and <a href="">-Wbaz</a>.</p>
449</dd>
450
Jeffrey Yasskin21d07e42010-06-05 01:39:57 +0000[diff] [blame]451<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
452<dt id="opt_Wambiguous-member-template"><b>-Wambiguous-member-template</b>:
453Warn about unqualified uses of a member template whose name resolves
454to another template at the location of the use.</dt>
455<dd>This option, which defaults to on, enables a warning in the
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]456following code:
Jeffrey Yasskin21d07e42010-06-05 01:39:57 +0000[diff] [blame]457
458<pre>
459template&lt;typename T> struct set{};
460template&lt;typename T> struct trait { typedef const T& type; };
461struct Value {
462 template&lt;typename T> void set(typename trait&lt;T>::type value) {}
463};
464void foo() {
465 Value v;
466 v.set&lt;double>(3.2);
467}
468</pre>
469
470<p>C++ [basic.lookup.classref] requires this to be an error, but,
471because it's hard to work around, Clang downgrades it to a warning as
472an extension.</p>
473</dd>
474
Jeffrey Yasskin57d12fd2010-06-07 15:58:05 +0000[diff] [blame]475<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
476<dt id="opt_Wbind-to-temporary-copy"><b>-Wbind-to-temporary-copy</b>: Warn about
477an unusable copy constructor when binding a reference to a temporary.</dt>
478<dd>This option, which defaults to on, enables warnings about binding a
479reference to a temporary when the temporary doesn't have a usable copy
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]480constructor. For example:
Jeffrey Yasskin57d12fd2010-06-07 15:58:05 +0000[diff] [blame]481
482<pre>
483 struct NonCopyable {
484 NonCopyable();
485 private:
486 NonCopyable(const NonCopyable&);
487 };
488 void foo(const NonCopyable&);
489 void bar() {
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]490 foo(NonCopyable()); // Disallowed in C++98; allowed in C++11.
Jeffrey Yasskin57d12fd2010-06-07 15:58:05 +0000[diff] [blame]491 }
492</pre>
493<pre>
494 struct NonCopyable2 {
495 NonCopyable2();
496 NonCopyable2(NonCopyable2&);
497 };
498 void foo(const NonCopyable2&);
499 void bar() {
David Blaikie5090e9f2011-10-18 05:49:30 +0000[diff] [blame]500 foo(NonCopyable2()); // Disallowed in C++98; allowed in C++11.
Jeffrey Yasskin57d12fd2010-06-07 15:58:05 +0000[diff] [blame]501 }
502</pre>
503
504<p>Note that if <tt>NonCopyable2::NonCopyable2()</tt> has a default
505argument whose instantiation produces a compile error, that error will
506still be a hard error in C++98 mode even if this warning is turned
507off.</p>
508
509</dd>
510
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]511</dl>
512
Chad Rosiercd376a12012-02-22 20:26:10 +0000[diff] [blame]513<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
514<h3 id="cl_crash_diagnostics">Options to Control Clang Crash Diagnostics</h3>
515<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
516
517<p>As unbelievable as it may sound, Clang does crash from time to time.
518Generally, this only occurs to those living on the
519<a href="http://llvm.org/releases/download.html#svn">bleeding edge</a>. Clang
520goes to great lengths to assist you in filing a bug report. Specifically, Clang
521generates preprocessed source file(s) and associated run script(s) upon a
522crash. These files should be attached to a bug report to ease reproducibility
523of the failure. Below are the command line options to control the crash
524diagnostics.
525</p>
526
527<p><b>-fno-crash-diagnostics</b>: Disable auto-generation of preprocessed
528source files during a clang crash.</p>
529
530<p>The -fno-crash-diagnostics flag can be helpful for speeding the process of
531generating a delta reduced test case.</p>
532
533
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]534<!-- ======================================================================= -->
535<h2 id="general_features">Language and Target-Independent Features</h2>
536<!-- ======================================================================= -->
537
538
539<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
540<h3 id="diagnostics">Controlling Errors and Warnings</h3>
541<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
542
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]543<p>Clang provides a number of ways to control which code constructs cause it to
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]544emit errors and warning messages, and how they are displayed to the console.</p>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]545
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]546<h4 id="diagnostics_display">Controlling How Clang Displays Diagnostics</h4>
Chris Lattner65a795b2009-04-20 06:00:23 +0000[diff] [blame]547
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]548<p>When Clang emits a diagnostic, it includes rich information in the output,
549and gives you fine-grain control over which information is printed. Clang has
550the ability to print this information, and these are the options that control
551it:</p>
552
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]553<ol>
554<li>A file/line/column indicator that shows exactly where the diagnostic occurs
555 in your code [<a href="#opt_fshow-column">-fshow-column</a>, <a
556 href="#opt_fshow-source-location">-fshow-source-location</a>].</li>
557<li>A categorization of the diagnostic as a note, warning, error, or fatal
558 error.</li>
559<li>A text string that describes what the problem is.</li>
560<li>An option that indicates how to control the diagnostic (for diagnostics that
561 support it) [<a
562 href="#opt_fdiagnostics-show-option">-fdiagnostics-show-option</a>].</li>
Chris Lattner3f145382010-05-24 21:35:18 +0000[diff] [blame]563<li>A <a href="#diagnostics_categories">high-level category</a> for the
564 diagnostic for clients that want to group diagnostics by class (for
565 diagnostics that support it) [<a
Chris Lattner28a43a42010-05-05 01:35:28 +0000[diff] [blame]566 href="#opt_fdiagnostics-show-category">-fdiagnostics-show-category</a>].</li>
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]567<li>The line of source code that the issue occurs on, along with a caret and
568 ranges that indicate the important locations [<a
569 href="opt_fcaret-diagnostics">-fcaret-diagnostics</a>].</li>
570<li>"FixIt" information, which is a concise explanation of how to fix the
571 problem (when Clang is certain it knows) [<a
572 href="opt_fdiagnostics-fixit-info">-fdiagnostics-fixit-info</a>].</li>
573<li>A machine-parsable representation of the ranges involved (off by
574 default) [<a
Chris Lattner2a9cc232009-04-21 05:35:32 +0000[diff] [blame]575 href="opt_fdiagnostics-print-source-range-info">-fdiagnostics-print-source-range-info</a>].</li>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]576</ol>
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]577
578<p>For more information please see <a href="#cl_diag_formatting">Formatting of
579Diagnostics</a>.</p>
580
Chris Lattner3f145382010-05-24 21:35:18 +0000[diff] [blame]581
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]582<h4 id="diagnostics_mappings">Diagnostic Mappings</h4>
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]583
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]584<p>All diagnostics are mapped into one of these 5 classes:</p>
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]585
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]586<ul>
587<li>Ignored</li>
588<li>Note</li>
589<li>Warning</li>
590<li>Error</li>
591<li>Fatal</li>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]592</ul>
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]593
Chris Lattner3f145382010-05-24 21:35:18 +0000[diff] [blame]594<h4 id="diagnostics_categories">Diagnostic Categories</h4>
595
596<p>Though not shown by default, diagnostics may each be associated with a
597 high-level category. This category is intended to make it possible to triage
598 builds that produce a large number of errors or warnings in a grouped way.
599</p>
600
601<p>Categories are not shown by default, but they can be turned on with the
602<a href="#opt_fdiagnostics-show-category">-fdiagnostics-show-category</a> option.
603When set to "<tt>name</tt>", the category is printed textually in the diagnostic
604output. When it is set to "<tt>id</tt>", a category number is printed. The
605mapping of category names to category id's can be obtained by running '<tt>clang
606 --print-diagnostic-categories</tt>'.
607</p>
608
609<h4 id="diagnostics_commandline">Controlling Diagnostics via Command Line
610 Flags</h4>
Chris Lattner8217f4e2009-04-20 06:26:18 +0000[diff] [blame]611
612<p>-W flags, -pedantic, etc</p>
613
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]614<h4 id="diagnostics_pragmas">Controlling Diagnostics via Pragmas</h4>
615
616<p>Clang can also control what diagnostics are enabled through the use of
617pragmas in the source code. This is useful for turning off specific warnings
618in a section of source code. Clang supports GCC's pragma for compatibility
619with existing source code, as well as several extensions. </p>
620
621<p>The pragma may control any warning that can be used from the command line.
622Warnings may be set to ignored, warning, error, or fatal. The following
623example code will tell Clang or GCC to ignore the -Wall warnings:</p>
624
625<pre>
626#pragma GCC diagnostic ignored "-Wall"
627</pre>
628
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]629<p>In addition to all of the functionality provided by GCC's pragma, Clang
Chris Lattner3401cf82009-07-12 21:22:10 +0000[diff] [blame]630also allows you to push and pop the current warning state. This is particularly
631useful when writing a header file that will be compiled by other people, because
632you don't know what warning flags they build with.</p>
633
634<p>In the below example
635-Wmultichar is ignored for only a single line of code, after which the
636diagnostics return to whatever state had previously existed.</p>
637
638<pre>
639#pragma clang diagnostic push
640#pragma clang diagnostic ignored "-Wmultichar"
641
642char b = 'df'; // no warning.
643
644#pragma clang diagnostic pop
645</pre>
646
647<p>The push and pop pragmas will save and restore the full diagnostic state of
648the compiler, regardless of how it was set. That means that it is possible to
649use push and pop around GCC compatible diagnostics and Clang will push and pop
650them appropriately, while GCC will ignore the pushes and pops as unknown
651pragmas. It should be noted that while Clang supports the GCC pragma, Clang and
652GCC do not support the exact same set of warnings, so even when using GCC
653compatible #pragmas there is no guarantee that they will have identical behaviour
654on both compilers. </p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]655
Ted Kremenekf391fa72011-08-18 01:17:05 +0000[diff] [blame]656<h4 id="diagnostics_enable_everything">Enabling All Warnings</h4>
657
658<p>In addition to the traditional <tt>-W</tt> flags, one can enable <b>all</b>
659 warnings by passing <tt>-Weverything</tt>.
660 This works as expected with <tt>-Werror</tt>,
661 and also includes the warnings from <tt>-pedantic</tt>.</p>
662
663<p>Note that when combined with <tt>-w</tt> (which disables all warnings), that
664 flag wins.</p>
665
Ted Kremenek2fb11eb2010-08-24 18:12:35 +0000[diff] [blame]666<h4 id="analyzer_diagnositics">Controlling Static Analyzer Diagnostics</h4>
667
668<p>While not strictly part of the compiler, the diagnostics from Clang's <a
669href="http://clang-analyzer.llvm.org">static analyzer</a> can also be influenced
670by the user via changes to the source code. This can be done in two ways:
671
672<ul>
673
674<li id="analyzer_annotations"><b>Annotations</b>: The static analyzer recognizes various GCC-style
675attributes (e.g., <tt>__attribute__((nonnull)))</tt>) that can either suppress
676static analyzer warnings or teach the analyzer about code invariants which
677enable it to find more bugs. While many of these attributes are standard GCC
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]678attributes, additional ones have been added to Clang to specifically support the
Ted Kremenek2fb11eb2010-08-24 18:12:35 +0000[diff] [blame]679static analyzer. Detailed information on these annotations can be found in the
680<a href="http://clang-analyzer.llvm.org/annotations.html">analyzer's
681documentation</a>.</li>
682
683<li><b><tt>__clang_analyzer__</tt></b>: When the static analyzer is using Clang
684to parse source files, it implicitly defines the preprocessor macro
685<tt>__clang_analyzer__</tt>. While discouraged, code can use this macro to
686selectively exclude code the analyzer examines. Here is an example:
687
688<pre>
689#ifndef __clang_analyzer__
690// Code not to be analyzed
691#endif
692</pre>
693
694In general, this usage is discouraged. Instead, we prefer that users file bugs
695against the analyzer when it flags false positives. There is also active
696discussion of allowing users in the future to selectively silence specific
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]697analyzer warnings (some of which can already be done using <a
Chris Lattner2c2a3852011-12-19 07:49:23 +0000[diff] [blame]698href="#analyzer_annotations">annotations</a>).</li>
Ted Kremenek2fb11eb2010-08-24 18:12:35 +0000[diff] [blame]699
700</ul>
701
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]702<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
703<h3 id="precompiledheaders">Precompiled Headers</h3>
704<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
705
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]706<p><a href="http://en.wikipedia.org/wiki/Precompiled_header">Precompiled
707headers</a> are a general approach employed by many compilers to reduce
708compilation time. The underlying motivation of the approach is that it is
709common for the same (and often large) header files to be included by
710multiple source files. Consequently, compile times can often be greatly improved
711by caching some of the (redundant) work done by a compiler to process headers.
712Precompiled header files, which represent one of many ways to implement
713this optimization, are literally files that represent an on-disk cache that
714contains the vital information necessary to reduce some of the work
715needed to process a corresponding header file. While details of precompiled
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]716headers vary between compilers, precompiled headers have been shown to be
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]717highly effective at speeding up program compilation on systems with very large
718system headers (e.g., Mac OS/X).</p>
719
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]720<h4>Generating a PCH File</h4>
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]721
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]722<p>To generate a PCH file using Clang, one invokes Clang with
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]723the <b><tt>-x <i>&lt;language&gt;</i>-header</tt></b> option. This mirrors the
724interface in GCC for generating PCH files:</p>
725
726<pre>
727 $ gcc -x c-header test.h -o test.h.gch
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]728 $ clang -x c-header test.h -o test.h.pch
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]729</pre>
730
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]731<h4>Using a PCH File</h4>
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]732
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]733<p>A PCH file can then be used as a prefix header when a
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]734<b><tt>-include</tt></b> option is passed to <tt>clang</tt>:</p>
735
736<pre>
737 $ clang -include test.h test.c -o test
738</pre>
739
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]740<p>The <tt>clang</tt> driver will first check if a PCH file for <tt>test.h</tt>
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]741is available; if so, the contents of <tt>test.h</tt> (and the files it includes)
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]742will be processed from the PCH file. Otherwise, Clang falls back to
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]743directly processing the content of <tt>test.h</tt>. This mirrors the behavior of
744GCC.</p>
745
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]746<p><b>NOTE:</b> Clang does <em>not</em> automatically use PCH files
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]747for headers that are directly included within a source file. For example:</p>
748
749<pre>
Chris Lattnere42ec542009-06-13 20:35:58 +0000[diff] [blame]750 $ clang -x c-header test.h -o test.h.pch
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]751 $ cat test.c
752 #include "test.h"
753 $ clang test.c -o test
754</pre>
755
Douglas Gregorf4d59532009-06-03 22:37:00 +0000[diff] [blame]756<p>In this example, <tt>clang</tt> will not automatically use the PCH file for
Chris Lattner5c3074f2009-04-20 04:37:38 +0000[diff] [blame]757<tt>test.h</tt> since <tt>test.h</tt> was included directly in the source file
758and not specified on the command line using <tt>-include</tt>.</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]759
Douglas Gregore650c8c2009-07-07 00:12:59 +0000[diff] [blame]760<h4>Relocatable PCH Files</h4>
761<p>It is sometimes necessary to build a precompiled header from headers that
762are not yet in their final, installed locations. For example, one might build a
763precompiled header within the build tree that is then meant to be installed
764alongside the headers. Clang permits the creation of "relocatable" precompiled
765headers, which are built with a given path (into the build directory) and can
766later be used from an installed location.</p>
767
768<p>To build a relocatable precompiled header, place your headers into a
769subdirectory whose structure mimics the installed location. For example, if you
770want to build a precompiled header for the header <code>mylib.h</code> that
771will be installed into <code>/usr/include</code>, create a subdirectory
772<code>build/usr/include</code> and place the header <code>mylib.h</code> into
773that subdirectory. If <code>mylib.h</code> depends on other headers, then
774they can be stored within <code>build/usr/include</code> in a way that mimics
775the installed location.</p>
776
777<p>Building a relocatable precompiled header requires two additional arguments.
778First, pass the <code>--relocatable-pch</code> flag to indicate that the
779resulting PCH file should be relocatable. Second, pass
780<code>-isysroot /path/to/build</code>, which makes all includes for your
781library relative to the build directory. For example:</p>
782
783<pre>
784 # clang -x c-header --relocatable-pch -isysroot /path/to/build /path/to/build/mylib.h mylib.h.pch
785</pre>
786
787<p>When loading the relocatable PCH file, the various headers used in the PCH
788file are found from the system header root. For example, <code>mylib.h</code>
789can be found in <code>/usr/include/mylib.h</code>. If the headers are installed
790in some other system root, the <code>-isysroot</code> option can be used provide
791a different system root from which the headers will be based. For example,
792<code>-isysroot /Developer/SDKs/MacOSX10.4u.sdk</code> will look for
793<code>mylib.h</code> in
794<code>/Developer/SDKs/MacOSX10.4u.sdk/usr/include/mylib.h</code>.</p>
795
796<p>Relocatable precompiled headers are intended to be used in a limited number
797of cases where the compilation environment is tightly controlled and the
798precompiled header cannot be generated after headers have been installed.
799Relocatable precompiled headers also have some performance impact, because
800the difference in location between the header locations at PCH build time vs.
801at the time of PCH use requires one of the PCH optimizations,
802<code>stat()</code> caching, to be disabled. However, this change is only
803likely to affect PCH files that reference a large number of headers.</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]804
Mike Stump53664ca2009-12-14 23:53:10 +0000[diff] [blame]805<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
806<h3 id="codegen">Controlling Code Generation</h3>
807<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
808
809<p>Clang provides a number of ways to control code generation. The options are listed below.</p>
810
811<!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -->
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]812<dl>
Mike Stump53664ca2009-12-14 23:53:10 +0000[diff] [blame]813<dt id="opt_fcatch-undefined-behavior"><b>-fcatch-undefined-behavior</b>: Turn
814on runtime code generation to check for undefined behavior.</dt>
815
816<dd>This option, which defaults to off, controls whether or not Clang
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]817adds runtime checks for undefined runtime behavior. If a check fails,
Mike Stump53664ca2009-12-14 23:53:10 +0000[diff] [blame]818<tt>__builtin_trap()</tt> is used to indicate failure.
819The checks are:
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]820<ul>
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]821<li>Subscripting where the static type of one operand is a variable
Mike Stump88b2a172009-12-16 03:25:12 +0000[diff] [blame]822 which is decayed from an array type and the other operand is
823 greater than the size of the array or less than zero.</li>
Mike Stump53664ca2009-12-14 23:53:10 +0000[diff] [blame]824<li>Shift operators where the amount shifted is greater or equal to the
825 promoted bit-width of the left-hand-side or less than zero.</li>
Mike Stump8f6a3ed2009-12-16 03:18:14 +0000[diff] [blame]826<li>If control flow reaches __builtin_unreachable.
827<li>When llvm implements more __builtin_object_size support, reads and
828 writes for objects that __builtin_object_size indicates we aren't
829 accessing valid memory. Bit-fields and vectors are not yet checked.
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]830</ul>
Mike Stump53664ca2009-12-14 23:53:10 +0000[diff] [blame]831</dd>
832
Nuno Lopesaa526242009-12-17 10:00:52 +0000[diff] [blame]833<dt id="opt_fno-assume-sane-operator-new"><b>-fno-assume-sane-operator-new</b>:
834Don't assume that the C++'s new operator is sane.</dt>
Nuno Lopesb23f20d2009-12-17 10:15:49 +0000[diff] [blame]835<dd>This option tells the compiler to do not assume that C++'s global new
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]836operator will always return a pointer that does not
Nuno Lopesaa526242009-12-17 10:00:52 +0000[diff] [blame]837alias any other pointer when the function returns.</dd>
Evan Chengfda026b2011-04-08 22:18:01 +0000[diff] [blame]838
Evan Cheng4cf4b592011-04-08 22:34:21 +0000[diff] [blame]839<dt id="opt_ftrap-function"><b>-ftrap-function=[name]</b>: Instruct code
Evan Chengfda026b2011-04-08 22:18:01 +0000[diff] [blame]840generator to emit a function call to the specified function name for
841<tt>__builtin_trap()</tt>.</dt>
842
843<dd>LLVM code generator translates <tt>__builtin_trap()</tt> to a trap
844instruction if it is supported by the target ISA. Otherwise, the builtin is
845translated into a call to <tt>abort</tt>. If this option is set, then the code
846generator will always lower the builtin to a call to the specified function
847regardless of whether the target ISA has a trap instruction. This option is
Chris Lattner02e266b2011-04-08 22:58:43 +0000[diff] [blame]848useful for environments (e.g. deeply embedded) where a trap cannot be properly
849handled, or when some custom behavior is desired.</dd>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]850</dl>
Nuno Lopesaa526242009-12-17 10:00:52 +0000[diff] [blame]851
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]852<!-- ======================================================================= -->
853<h2 id="c">C Language Features</h2>
854<!-- ======================================================================= -->
855
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]856<p>The support for standard C in clang is feature-complete except for the C99
857floating-point pragmas.</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]858
859<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]860<h3 id="c_ext">Extensions supported by clang</h3>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]861<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
862
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]863<p>See <a href="LanguageExtensions.html">clang language extensions</a>.</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]864
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]865<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
866<h3 id="c_modes">Differences between various standard modes</h3>
867<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]868
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]869<p>clang supports the -std option, which changes what language mode clang uses.
870The supported modes for C are c89, gnu89, c94, c99, gnu99 and various aliases
871for those modes. If no -std option is specified, clang defaults to gnu99 mode.
872</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]873
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]874<p>Differences between all c* and gnu* modes:</p>
875<ul>
876<li>c* modes define "__STRICT_ANSI__".</li>
Eli Friedman26fa0ed2009-05-27 23:02:57 +0000[diff] [blame]877<li>Target-specific defines not prefixed by underscores, like "linux", are
878defined in gnu* modes.</li>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]879<li>Trigraphs default to being off in gnu* modes; they can be enabled by the
880-trigraphs option.</li>
881<li>The parser recognizes "asm" and "typeof" as keywords in gnu* modes; the
882variants "__asm__" and "__typeof__" are recognized in all modes.</li>
Eli Friedmanb0ac5452009-05-16 23:17:30 +0000[diff] [blame]883<li>The Apple "blocks" extension is recognized by default in gnu* modes
884on some platforms; it can be enabled in any mode with the "-fblocks"
885option.</li>
Chris Lattner741be6a2011-06-15 00:41:40 +0000[diff] [blame]886<li>Arrays that are VLA's according to the standard, but which can be constant
887 folded by the frontend are treated as fixed size arrays. This occurs for
888 things like "int X[(1, 2)];", which is technically a VLA. c* modes are
889 strictly compliant and treat these as VLAs.</li>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]890</ul>
891
892<p>Differences between *89 and *99 modes:</p>
893<ul>
894<li>The *99 modes default to implementing "inline" as specified in C99, while
895the *89 modes implement the GNU version. This can be overridden for individual
896functions with the __gnu_inline__ attribute.</li>
Eli Friedman26fa0ed2009-05-27 23:02:57 +0000[diff] [blame]897<li>Digraphs are not recognized in c89 mode.</li>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]898<li>The scope of names defined inside a "for", "if", "switch", "while", or "do"
899statement is different. (example: "if ((struct x {int x;}*)0) {}".)</li>
900<li>__STDC_VERSION__ is not defined in *89 modes.</li>
Eli Friedman26fa0ed2009-05-27 23:02:57 +0000[diff] [blame]901<li>"inline" is not recognized as a keyword in c89 mode.</li>
902<li>"restrict" is not recognized as a keyword in *89 modes.</li>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]903<li>Commas are allowed in integer constant expressions in *99 modes.</li>
904<li>Arrays which are not lvalues are not implicitly promoted to pointers in
905*89 modes.</li>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]906<li>Some warnings are different.</li>
907</ul>
908
909<p>c94 mode is identical to c89 mode except that digraphs are enabled in
910c94 mode (FIXME: And __STDC_VERSION__ should be defined!).</p>
911
912<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
913<h3 id="c_unimpl_gcc">GCC extensions not implemented yet</h3>
914<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
915
916<p>clang tries to be compatible with gcc as much as possible, but some gcc
917extensions are not implemented yet:</p>
918
919<ul>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]920
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]921<li>clang does not support #pragma weak
Eli Friedman4da92552009-06-02 08:21:31 +0000[diff] [blame]922(<a href="http://llvm.org/bugs/show_bug.cgi?id=3679">bug 3679</a>). Due to
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]923the uses described in the bug, this is likely to be implemented at some
924point, at least partially.</li>
925
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]926<li>clang does not support decimal floating point types (_Decimal32 and
927friends) or fixed-point types (_Fract and friends); nobody has expressed
928interest in these features yet, so it's hard to say when they will be
929implemented.</li>
930
931<li>clang does not support nested functions; this is a complex feature which
932is infrequently used, so it is unlikely to be implemented anytime soon.</li>
933
Daniel Dunbar05fa6292009-05-12 23:12:07 +0000[diff] [blame]934<li>clang does not support global register variables, this is unlikely
Eli Friedmanca130bd2009-06-12 20:04:25 +0000[diff] [blame]935to be implemented soon because it requires additional LLVM backend support.
936</li>
Daniel Dunbar05fa6292009-05-12 23:12:07 +0000[diff] [blame]937
938<li>clang does not support static initialization of flexible array
939members. This appears to be a rarely used extension, but could be
940implemented pending user demand.</li>
941
Eli Friedmanca130bd2009-06-12 20:04:25 +0000[diff] [blame]942<li>clang does not support __builtin_va_arg_pack/__builtin_va_arg_pack_len.
943This is used rarely, but in some potentially interesting places, like the
944glibc headers, so it may be implemented pending user demand. Note that
945because clang pretends to be like GCC 4.2, and this extension was introduced
946in 4.3, the glibc headers will not try to use this extension with clang at
947the moment.</li>
948
949<li>clang does not support the gcc extension for forward-declaring function
950parameters; this has not showed up in any real-world code yet, though, so it
951might never be implemented.</li>
952
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]953</ul>
954
955<p>This is not a complete list; if you find an unsupported extension
956missing from this list, please send an e-mail to cfe-dev. This list
957currently excludes C++; see <a href="#cxx">C++ Language Features</a>.
958Also, this list does not include bugs in mostly-implemented features; please
959see the <a href="http://llvm.org/bugs/buglist.cgi?quicksearch=product%3Aclang+component%3A-New%2BBugs%2CAST%2CBasic%2CDriver%2CHeaders%2CLLVM%2BCodeGen%2Cparser%2Cpreprocessor%2CSemantic%2BAnalyzer">
960bug tracker</a> for known existing bugs (FIXME: Is there a section for
961bug-reporting guidelines somewhere?).</p>
962
963<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
964<h3 id="c_unsupp_gcc">Intentionally unsupported GCC extensions</h3>
965<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
966
Eli Friedman0b326002009-06-12 20:11:05 +0000[diff] [blame]967<ul>
968
Eli Friedman2f3fca02009-06-12 20:08:48 +0000[diff] [blame]969<li>clang does not support the gcc extension that allows variable-length arrays
Chris Lattnerd462b6a2011-01-24 03:47:59 +0000[diff] [blame]970in structures. This is for a few reasons: one, it is tricky
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]971to implement, two, the extension is completely undocumented, and three, the
Chris Lattnerb9d511c2010-10-28 02:20:32 +0000[diff] [blame]972extension appears to be rarely used. Note that clang <em>does</em> support
973flexible array members (arrays with a zero or unspecified size at the end of
974a structure).</li>
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]975
Eli Friedmanca130bd2009-06-12 20:04:25 +0000[diff] [blame]976<li>clang does not have an equivalent to gcc's "fold"; this means that
977clang doesn't accept some constructs gcc might accept in contexts where a
Eli Friedman248f9732011-06-13 23:12:01 +0000[diff] [blame]978constant expression is required, like "x-x" where x is a variable.</li>
Eli Friedmanca130bd2009-06-12 20:04:25 +0000[diff] [blame]979
Eli Friedmanca130bd2009-06-12 20:04:25 +0000[diff] [blame]980<li>clang does not support __builtin_apply and friends; this extension is
981extremely obscure and difficult to implement reliably.</li>
982
Eli Friedman0b326002009-06-12 20:11:05 +0000[diff] [blame]983</ul>
984
Eli Friedman660a5ac2009-04-28 18:48:34 +0000[diff] [blame]985<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
986<h3 id="c_ms">Microsoft extensions</h3>
987<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
988
989<p>clang has some experimental support for extensions from
990Microsoft Visual C++; to enable it, use the -fms-extensions command-line
Eli Friedmana1821b52009-06-08 06:21:03 +0000[diff] [blame]991option. This is the default for Windows targets. Note that the
992support is incomplete; enabling Microsoft extensions will silently drop
993certain constructs (including __declspec and Microsoft-style asm statements).
994</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]995
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]996<ul>
Douglas Gregord3b227d2010-12-14 16:21:49 +0000[diff] [blame]997<li>clang allows setting _MSC_VER with -fmsc-version=. It defaults to 1300 which
Michael J. Spencerdae4ac42010-10-21 05:21:48 +0000[diff] [blame]998is the same as Visual C/C++ 2003. Any number is supported and can greatly affect
999what Windows SDK and c++stdlib headers clang can compile. This option will be
1000removed when clang supports the full set of MS extensions required for these
1001headers.</li>
1002
Daniel Dunbar05fa6292009-05-12 23:12:07 +0000[diff] [blame]1003<li>clang does not support the Microsoft extension where anonymous
Daniel Dunbar869e0542009-05-13 00:23:53 +0000[diff] [blame]1004record members can be declared using user defined typedefs.</li>
Daniel Dunbar05fa6292009-05-12 23:12:07 +0000[diff] [blame]1005
Daniel Dunbar9375ed12009-05-13 21:40:49 +0000[diff] [blame]1006<li>clang supports the Microsoft "#pragma pack" feature for
1007controlling record layout. GCC also contains support for this feature,
1008however where MSVC and GCC are incompatible clang follows the MSVC
1009definition.</li>
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]1010</ul>
Daniel Dunbar9375ed12009-05-13 21:40:49 +0000[diff] [blame]1011
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]1012<!-- ======================================================================= -->
Richard Smithbdb10182011-11-21 20:54:59 +0000[diff] [blame]1013<h2 id="cxx">C++ Language Features</h2>
1014<!-- ======================================================================= -->
1015
1016<p>clang fully implements all of standard C++98 except for exported templates
1017(which were removed in C++11), and
1018<a href="http://clang.llvm.org/cxx_status.html">many C++11 features</a> are also
1019implemented.</p>
1020
1021<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
1022<h3 id="cxx_implimits">Controlling implementation limits</h3>
1023<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
1024
1025<p><b>-fconstexpr-depth=N</b>: Sets the limit for recursive constexpr function
1026invocations to N. The default is 512.</p>
1027
1028<p><b>-ftemplate-depth=N</b>: Sets the limit for recursively nested template
1029instantiations to N. The default is 1024.</p>
1030
1031<!-- ======================================================================= -->
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]1032<h2 id="target_features">Target-Specific Features and Limitations</h2>
1033<!-- ======================================================================= -->
1034
1035
1036<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
1037<h3 id="target_arch">CPU Architectures Features and Limitations</h3>
1038<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
1039
1040<!-- ======================== -->
1041<h4 id="target_arch_x86">X86</h4>
1042<!-- ======================== -->
Daniel Dunbarbcaf7aa2010-09-19 19:26:59 +0000[diff] [blame]1043
1044<p>The support for X86 (both 32-bit and 64-bit) is considered stable on Darwin
1045(Mac OS/X), Linux, FreeBSD, and Dragonfly BSD: it has been tested to correctly
1046compile many large C, C++, Objective-C, and Objective-C++ codebases.</p>
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]1047
NAKAMURA Takumiddf68ff2011-04-05 04:31:31 +0000[diff] [blame]1048<p>On x86_64-mingw32, passing i128(by value) is incompatible to Microsoft x64
1049calling conversion. You might need to tweak WinX86_64ABIInfo::classify()
1050in lib/CodeGen/TargetInfo.cpp.</p>
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]1051
Eli Friedman3b658d32009-06-08 05:12:39 +0000[diff] [blame]1052<!-- ======================== -->
1053<h4 id="target_arch_arm">ARM</h4>
1054<!-- ======================== -->
Daniel Dunbarbcaf7aa2010-09-19 19:26:59 +0000[diff] [blame]1055
1056<p>The support for ARM (specifically ARMv6 and ARMv7) is considered stable on
1057Darwin (iOS): it has been tested to correctly compile many large C, C++,
Bob Wilson4ea8dfa2011-01-10 17:55:14 +0000[diff] [blame]1058Objective-C, and Objective-C++ codebases. Clang only supports a limited number
1059of ARM architectures. It does not yet fully support ARMv5, for example.</p>
Eli Friedman3b658d32009-06-08 05:12:39 +0000[diff] [blame]1060
1061<!-- ======================== -->
1062<h4 id="target_arch_other">Other platforms</h4>
1063<!-- ======================== -->
1064clang currently contains some support for PPC and Sparc; however, significant
1065pieces of code generation are still missing, and they haven't undergone
1066significant testing.
1067
Eli Friedman3b658d32009-06-08 05:12:39 +0000[diff] [blame]1068<p>clang contains limited support for the MSP430 embedded processor, but both
1069the clang support and the LLVM backend support are highly experimental.
1070
1071<p>Other platforms are completely unsupported at the moment. Adding the
1072minimal support needed for parsing and semantic analysis on a new platform
1073is quite easy; see lib/Basic/Targets.cpp in the clang source tree. This level
1074of support is also sufficient for conversion to LLVM IR for simple programs.
1075Proper support for conversion to LLVM IR requires adding code to
1076lib/CodeGen/CGCall.cpp at the moment; this is likely to change soon, though.
1077Generating assembly requires a suitable LLVM backend.
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]1078
1079<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
1080<h3 id="target_os">Operating System Features and Limitations</h3>
1081<!-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = -->
1082
1083<!-- ======================================= -->
1084<h4 id="target_os_darwin">Darwin (Mac OS/X)</h4>
1085<!-- ======================================= -->
1086
1087<p>No __thread support, 64-bit ObjC support requires SL tools.</p>
1088
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]1089<!-- ======================================= -->
1090<h4 id="target_os_win32">Windows</h4>
1091<!-- ======================================= -->
1092
1093<p>Experimental supports are on Cygming.</p>
1094
1095<h5>Cygwin</h5>
1096
1097<p>Clang works on Cygwin-1.7.</p>
1098
1099<h5>MinGW32</h5>
1100
1101<p>Clang works on some mingw32 distributions.
1102Clang assumes directories as below;</p>
1103
1104<ul>
1105<li><tt>C:/mingw/include</tt></li>
1106<li><tt>C:/mingw/lib</tt></li>
1107<li><tt>C:/mingw/lib/gcc/mingw32/4.[3-5].0/include/c++</tt></li>
1108</ul>
1109
NAKAMURA Takumi0e53df62011-11-17 07:54:13 +0000[diff] [blame]1110<p>On MSYS, a few tests might fail.</p>
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]1111
1112<h5>MinGW-w64</h5>
1113
Douglas Gregordad879a2011-08-30 00:40:12 +0000[diff] [blame]1114<p>For 32-bit (i686-w64-mingw32), and 64-bit (x86_64-w64-mingw32), Clang assumes as below;<p>
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]1115
1116<ul>
Douglas Gregordad879a2011-08-30 00:40:12 +0000[diff] [blame]1117<li><tt>GCC versions 4.5.0 to 4.5.3, 4.6.0 to 4.6.2, or 4.7.0 (for the C++ header search path)</tt></li>
1118<li><tt>some_directory/bin/gcc.exe</tt></li>
1119<li><tt>some_directory/bin/clang.exe</tt></li>
1120<li><tt>some_directory/bin/clang++.exe</tt></li>
1121<li><tt>some_directory/bin/../include/c++/GCC_version</tt></li>
1122<li><tt>some_directory/bin/../include/c++/GCC_version/x86_64-w64-mingw32</tt></li>
1123<li><tt>some_directory/bin/../include/c++/GCC_version/i686-w64-mingw32</tt></li>
1124<li><tt>some_directory/bin/../include/c++/GCC_version/backward</tt></li>
1125<li><tt>some_directory/bin/../x86_64-w64-mingw32/include</tt></li>
1126<li><tt>some_directory/bin/../i686-w64-mingw32/include</tt></li>
1127<li><tt>some_directory/bin/../include</tt></li>
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]1128</ul>
1129
Douglas Gregordad879a2011-08-30 00:40:12 +0000[diff] [blame]1130<p>This directory layout is standard for any toolchain you will find on the official <a href="mingw-w64.sourceforge.net">MinGW-w64 website</a>.
1131
1132<p>Clang expects the GCC executable &quot;gcc.exe&quot; compiled for i686-w64-mingw32 (or x86_64-w64-mingw32) to be present on PATH.</p>
1133
1134<p><a href="http://llvm.org/bugs/show_bug.cgi?id=9072">Some tests might fail</a>
1135on x86_64-w64-mingw32.</p>
NAKAMURA Takumia75fdb22011-04-04 15:02:41 +0000[diff] [blame]1136
Chris Lattnercf17d9d2009-04-20 04:23:09 +0000[diff] [blame]1137</div>
1138</body>
1139</html>