Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / c500566d9fb70472a17f62b4cf1689062c52e095 / . / docs / FAQ.html
blob: d62ffd715ab8fcaa9e94418e50e121ec448716ce [file] [log] [blame]
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
3<html>
4<head>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]5 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]6 <title>LLVM: Frequently Asked Questions</title>
Misha Brukman7ce62cc2004-06-01 18:51:03 +0000[diff] [blame]7 <style type="text/css">
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]8 @import url("llvm.css");
9 .question { font-weight: bold }
10 .answer { margin-left: 2em }
11 </style>
12</head>
13<body>
John Criswellc310f622003-10-13 16:13:06 +0000[diff] [blame]14
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]15<div class="doc_title">
16 LLVM: Frequently Asked Questions
17</div>
18
19<ol>
20 <li><a href="#license">License</a>
21 <ol>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]22 <li>Why are the LLVM source code and the front-end distributed under
23 different licenses?</li>
24
25 <li>Does the University of Illinois Open Source License really qualify as an
26 "open source" license?</li>
27
28 <li>Can I modify LLVM source code and redistribute the modified source?</li>
29
30 <li>Can I modify LLVM source code and redistribute binaries or other tools
31 based on it, without redistributing the source?</li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]32 </ol></li>
33
34 <li><a href="#source">Source code</a>
35 <ol>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]36 <li>In what language is LLVM written?</li>
37
38 <li>How portable is the LLVM source code?</li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]39 </ol></li>
40
41 <li><a href="#build">Build Problems</a>
42 <ol>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]43 <li>When I run configure, it finds the wrong C compiler.</li>
44
45 <li>The <tt>configure</tt> script finds the right C compiler, but it uses
46 the LLVM linker from a previous build. What do I do?</li>
47
48 <li>When creating a dynamic library, I get a strange GLIBC error.</li>
49
50 <li>I've updated my source tree from Subversion, and now my build is trying
51 to use a file/directory that doesn't exist.</li>
52
53 <li>I've modified a Makefile in my source tree, but my build tree keeps
54 using the old version. What do I do?</li>
55
56 <li>I've upgraded to a new version of LLVM, and I get strange build
57 errors.</li>
58
59 <li>I've built LLVM and am testing it, but the tests freeze.</li>
60
61 <li>Why do test results differ when I perform different types of
62 builds?</li>
63
64 <li>Compiling LLVM with GCC 3.3.2 fails, what should I do?</li>
65
66 <li>Compiling LLVM with GCC succeeds, but the resulting tools do not work,
67 what can be wrong?</li>
68
69 <li>When I use the test suite, all of the C Backend tests fail. What is
70 wrong?</li>
71
72 <li>After Subversion update, rebuilding gives the error "No rule to make
73 target".</li>
74
75 <li><a href="#llvmc">The <tt>llvmc</tt> program gives me errors/doesn't
76 work.</a></li>
Bill Wendlingf1594022009-04-07 18:51:13 +0000[diff] [blame]77
78 <li><a href="#srcdir-objdir">When I compile LLVM-GCC with srcdir == objdir,
79 it fails. Why?</a></li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]80 </ol></li>
John Criswell76c1e382003-11-18 16:08:49 +0000[diff] [blame]81
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]82 <li><a href="#felangs">Source Languages</a>
83 <ol>
84 <li><a href="#langs">What source languages are supported?</a></li>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]85
Gordon Henriksene5079052008-02-22 21:55:51 +0000[diff] [blame]86 <li><a href="#langirgen">I'd like to write a self-hosting LLVM compiler. How
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]87 should I interface with the LLVM middle-end optimizers and back-end code
88 generators?</a></li>
89
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]90 <li><a href="#langhlsupp">What support is there for higher level source
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]91 language constructs for building a compiler?</a></li>
92
Reid Spencere00906f2006-08-10 20:15:58 +0000[diff] [blame]93 <li><a href="GetElementPtr.html">I don't understand the GetElementPtr
94 instruction. Help!</a></li>
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]95 </ol>
96
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]97 <li><a href="#cfe">Using the GCC Front End</a>
John Criswell76c1e382003-11-18 16:08:49 +0000[diff] [blame]98 <ol>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]99 <li>When I compile software that uses a configure script, the configure
100 script thinks my system has all of the header files and libraries it is
101 testing for. How do I get configure to work correctly?</li>
John Criswell76c1e382003-11-18 16:08:49 +0000[diff] [blame]102
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]103 <li>When I compile code using the LLVM GCC front end, it complains that it
104 cannot find libcrtend.a?</li>
Tanya Lattner14fc5c12005-04-25 20:36:56 +0000[diff] [blame]105
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]106 <li>How can I disable all optimizations when compiling code using the LLVM
107 GCC front end?</li>
Tanya Lattner14fc5c12005-04-25 20:36:56 +0000[diff] [blame]108
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]109 <li><a href="#translatecxx">Can I use LLVM to convert C++ code to C
110 code?</a></li>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]111
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]112 <li><a href="#platformindependent">Can I compile C or C++ code to
113 platform-independent LLVM bitcode?</a></li>
John Criswell76c1e382003-11-18 16:08:49 +0000[diff] [blame]114 </ol>
115 </li>
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]116
117 <li><a href="#cfe_code">Questions about code generated by the GCC front-end</a>
118 <ol>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]119 <li><a href="#iosinit">What is this <tt>llvm.global_ctors</tt> and
Chris Lattnerc50bbc92004-03-29 19:14:35 +0000[diff] [blame]120 <tt>_GLOBAL__I__tmp_webcompile...</tt> stuff that happens when I
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]121 #include &lt;iostream&gt;?</a></li>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]122
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]123 <li><a href="#codedce">Where did all of my code go??</a></li>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]124
125 <li><a href="#undef">What is this "<tt>undef</tt>" thing that shows up in
126 my code?</a></li>
Chris Lattner2c6f9f72009-06-30 17:10:19 +0000[diff] [blame]127
128 <li><a href="#callconvwrong">Why does instcombine + simplifycfg turn
129 a call to a function with a mismatched calling convention into "unreachable"?
130 Why not make the verifier reject it?</a></li>
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]131 </ol>
132 </li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]133</ol>
134
Chris Lattner7911ce22004-05-23 21:07:27 +0000[diff] [blame]135<div class="doc_author">
Reid Spencer05fe4b02006-03-14 05:39:39 +0000[diff] [blame]136 <p>Written by <a href="http://llvm.org">The LLVM Team</a></p>
Chris Lattner7911ce22004-05-23 21:07:27 +0000[diff] [blame]137</div>
138
139
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]140<!-- *********************************************************************** -->
141<div class="doc_section">
142 <a name="license">License</a>
143</div>
144<!-- *********************************************************************** -->
145
146<div class="question">
147<p>Why are the LLVM source code and the front-end distributed under different
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]148 licenses?</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]149</div>
150
151<div class="answer">
152<p>The C/C++ front-ends are based on GCC and must be distributed under the GPL.
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]153 Our aim is to distribute LLVM source code under a <em>much less
154 restrictive</em> license, in particular one that does not compel users who
155 distribute tools based on modifying the source to redistribute the modified
156 source code as well.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]157</div>
158
159<div class="question">
160<p>Does the University of Illinois Open Source License really qualify as an
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]161 "open source" license?</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]162</div>
163
164<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]165<p>Yes, the license
166 is <a href="http://www.opensource.org/licenses/UoI-NCSA.php">certified</a> by
167 the Open Source Initiative (OSI).</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]168</div>
169
170<div class="question">
171<p>Can I modify LLVM source code and redistribute the modified source?</p>
172</div>
173
174<div class="answer">
175<p>Yes. The modified source distribution must retain the copyright notice and
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]176 follow the three bulletted conditions listed in
177 the <a href="http://llvm.org/svn/llvm-project/llvm/trunk/LICENSE.TXT">LLVM
178 license</a>.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]179</div>
180
181<div class="question">
182<p>Can I modify LLVM source code and redistribute binaries or other tools based
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]183 on it, without redistributing the source?</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]184</div>
185
186<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]187<p>Yes. This is why we distribute LLVM under a less restrictive license than
188 GPL, as explained in the first question above.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]189</div>
190
191<!-- *********************************************************************** -->
192<div class="doc_section">
193 <a name="source">Source Code</a>
194</div>
195<!-- *********************************************************************** -->
196
197<div class="question">
198<p>In what language is LLVM written?</p>
199</div>
200
201<div class="answer">
202<p>All of the LLVM tools and libraries are written in C++ with extensive use of
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]203 the STL.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]204</div>
205
206<div class="question">
207<p>How portable is the LLVM source code?</p>
208</div>
209
210<div class="answer">
211<p>The LLVM source code should be portable to most modern UNIX-like operating
212systems. Most of the code is written in standard C++ with operating system
213services abstracted to a support library. The tools required to build and test
214LLVM have been ported to a plethora of platforms.</p>
215
216<p>Some porting problems may exist in the following areas:</p>
217
218<ul>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]219 <li>The GCC front end code is not as portable as the LLVM suite, so it may not
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]220 compile as well on unsupported platforms.</li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]221
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]222 <li>The LLVM build system relies heavily on UNIX shell tools, like the Bourne
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]223 Shell and sed. Porting to systems without these tools (MacOS 9, Plan 9)
224 will require more effort.</li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]225</ul>
226
227</div>
228
229<!-- *********************************************************************** -->
230<div class="doc_section">
231 <a name="build">Build Problems</a>
232</div>
233<!-- *********************************************************************** -->
234
235<div class="question">
236<p>When I run configure, it finds the wrong C compiler.</p>
237</div>
238
239<div class="answer">
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]240<p>The <tt>configure</tt> script attempts to locate first <tt>gcc</tt> and then
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]241 <tt>cc</tt>, unless it finds compiler paths set in <tt>CC</tt>
242 and <tt>CXX</tt> for the C and C++ compiler, respectively.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]243
244<p>If <tt>configure</tt> finds the wrong compiler, either adjust your
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]245 <tt>PATH</tt> environment variable or set <tt>CC</tt> and <tt>CXX</tt>
246 explicitly.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]247
248</div>
249
250<div class="question">
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]251<p>The <tt>configure</tt> script finds the right C compiler, but it uses the
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]252 LLVM linker from a previous build. What do I do?</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]253</div>
254
255<div class="answer">
256<p>The <tt>configure</tt> script uses the <tt>PATH</tt> to find executables, so
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]257 if it's grabbing the wrong linker/assembler/etc, there are two ways to fix
258 it:</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]259
260<ol>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]261 <li><p>Adjust your <tt>PATH</tt> environment variable so that the correct
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]262 program appears first in the <tt>PATH</tt>. This may work, but may not be
263 convenient when you want them <i>first</i> in your path for other
264 work.</p></li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]265
266 <li><p>Run <tt>configure</tt> with an alternative <tt>PATH</tt> that is
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]267 correct. In a Borne compatible shell, the syntax would be:</p>
268
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]269<pre class="doc_code">
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]270% PATH=[the path without the bad program] ./configure ...
271</pre>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]272
273 <p>This is still somewhat inconvenient, but it allows <tt>configure</tt>
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]274 to do its work without having to adjust your <tt>PATH</tt>
275 permanently.</p></li>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]276</ol>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]277</div>
278
279<div class="question">
280<p>When creating a dynamic library, I get a strange GLIBC error.</p>
281</div>
282
283<div class="answer">
284<p>Under some operating systems (i.e. Linux), libtool does not work correctly if
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]285 GCC was compiled with the --disable-shared option. To work around this,
286 install your own version of GCC that has shared libraries enabled by
287 default.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]288</div>
289
290<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]291<p>I've updated my source tree from Subversion, and now my build is trying to
292 use a file/directory that doesn't exist.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]293</div>
294
295<div class="answer">
296<p>You need to re-run configure in your object directory. When new Makefiles
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]297 are added to the source tree, they have to be copied over to the object tree
298 in order to be used by the build.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]299</div>
300
301<div class="question">
302<p>I've modified a Makefile in my source tree, but my build tree keeps using the
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]303 old version. What do I do?</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]304</div>
305
306<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]307<p>If the Makefile already exists in your object tree, you can just run the
308 following command in the top level directory of your object tree:</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]309
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]310<pre class="doc_code">
311% ./config.status &lt;relative path to Makefile&gt;
312</pre>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]313
314<p>If the Makefile is new, you will have to modify the configure script to copy
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]315 it over.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]316</div>
317
318<div class="question">
319<p>I've upgraded to a new version of LLVM, and I get strange build errors.</p>
320</div>
321
322<div class="answer">
323
324<p>Sometimes, changes to the LLVM source code alters how the build system works.
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]325 Changes in libtool, autoconf, or header file dependencies are especially
326 prone to this sort of problem.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]327
328<p>The best thing to try is to remove the old files and re-build. In most
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]329 cases, this takes care of the problem. To do this, just type <tt>make
330 clean</tt> and then <tt>make</tt> in the directory that fails to build.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]331</div>
332
333<div class="question">
334<p>I've built LLVM and am testing it, but the tests freeze.</p>
335</div>
336
337<div class="answer">
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]338<p>This is most likely occurring because you built a profile or release
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]339 (optimized) build of LLVM and have not specified the same information on the
340 <tt>gmake</tt> command line.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]341
342<p>For example, if you built LLVM with the command:</p>
343
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]344<pre class="doc_code">
345% gmake ENABLE_PROFILING=1
346</pre>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]347
348<p>...then you must run the tests with the following commands:</p>
349
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]350<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]351% cd llvm/test
352% gmake ENABLE_PROFILING=1
353</pre>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]354</div>
355
356<div class="question">
357<p>Why do test results differ when I perform different types of builds?</p>
358</div>
359
360<div class="answer">
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]361<p>The LLVM test suite is dependent upon several features of the LLVM tools and
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]362 libraries.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]363
364<p>First, the debugging assertions in code are not enabled in optimized or
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]365 profiling builds. Hence, tests that used to fail may pass.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]366
367<p>Second, some tests may rely upon debugging options or behavior that is only
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]368 available in the debug build. These tests will fail in an optimized or
369 profile build.</p>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]370</div>
371
Chris Lattner8a0b9242003-12-08 05:43:19 +0000[diff] [blame]372<div class="question">
Chris Lattner306acee2003-12-22 04:06:12 +0000[diff] [blame]373<p>Compiling LLVM with GCC 3.3.2 fails, what should I do?</p>
Chris Lattner8a0b9242003-12-08 05:43:19 +0000[diff] [blame]374</div>
375
376<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]377<p>This is <a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13392">a bug in
378 GCC</a>, and affects projects other than LLVM. Try upgrading or downgrading
379 your GCC.</p>
Chris Lattner8a0b9242003-12-08 05:43:19 +0000[diff] [blame]380</div>
381
John Criswelld1799612004-03-29 20:23:11 +0000[diff] [blame]382<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]383<p>Compiling LLVM with GCC succeeds, but the resulting tools do not work, what
384 can be wrong?</p>
Gabor Greif54820ce2009-03-02 19:08:05 +0000[diff] [blame]385</div>
386
387<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]388<p>Several versions of GCC have shown a weakness in miscompiling the LLVM
389 codebase. Please consult your compiler version (<tt>gcc --version</tt>) to
390 find out whether it is <a href="GettingStarted.html#brokengcc">broken</a>.
391 If so, your only option is to upgrade GCC to a known good version.</p>
Gabor Greif54820ce2009-03-02 19:08:05 +0000[diff] [blame]392</div>
393
394<div class="question">
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]395<p>After Subversion update, rebuilding gives the error "No rule to make
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]396 target".</p>
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]397</div>
398
399<div class="answer">
400<p>If the error is of the form:</p>
401
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]402<pre class="doc_code">
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]403gmake[2]: *** No rule to make target `/path/to/somefile', needed by
404`/path/to/another/file.d'.<br>
405Stop.
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]406</pre>
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]407
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]408<p>This may occur anytime files are moved within the Subversion repository or
409 removed entirely. In this case, the best solution is to erase all
410 <tt>.d</tt> files, which list dependencies for source files, and rebuild:</p>
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]411
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]412<pre class="doc_code">
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]413% cd $LLVM_OBJ_DIR
414% rm -f `find . -name \*\.d`
415% gmake
416</pre>
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]417
418<p>In other cases, it may be necessary to run <tt>make clean</tt> before
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]419 rebuilding.</p>
Misha Brukman1739aec2004-09-09 16:36:47 +0000[diff] [blame]420</div>
421
Bill Wendlingf1594022009-04-07 18:51:13 +0000[diff] [blame]422<div class="question">
423<p><a name="llvmc">The <tt>llvmc</tt> program gives me errors/doesn't
424 work.</a></p>
Bill Wendling174d5782007-05-29 09:35:34 +0000[diff] [blame]425</div>
426
427<div class="answer">
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]428<p><tt>llvmc</tt> is experimental and isn't really supported. We suggest
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]429 using <tt>llvm-gcc</tt> instead.</p>
Bill Wendling174d5782007-05-29 09:35:34 +0000[diff] [blame]430</div>
431
Bill Wendlingf1594022009-04-07 18:51:13 +0000[diff] [blame]432<div class="question">
Bill Wendlinga0bd8102009-04-07 18:54:06 +0000[diff] [blame]433<p><a name="srcdir-objdir">When I compile LLVM-GCC with srcdir == objdir, it
Bill Wendlingf1594022009-04-07 18:51:13 +0000[diff] [blame]434 fails. Why?</a></p>
435</div>
436
437<div class="answer">
438<p>The <tt>GNUmakefile</tt> in the top-level directory of LLVM-GCC is a special
439 <tt>Makefile</tt> used by Apple to invoke the <tt>build_gcc</tt> script after
440 setting up a special environment. This has the unforunate side-effect that
441 trying to build LLVM-GCC with srcdir == objdir in a "non-Apple way" invokes
442 the <tt>GNUmakefile</tt> instead of <tt>Makefile</tt>. Because the
443 environment isn't set up correctly to do this, the build fails.</p>
444
445<p>People not building LLVM-GCC the "Apple way" need to build LLVM-GCC with
446 srcdir != objdir, or simply remove the GNUmakefile entirely.</p>
447
448<p>We regret the inconvenience.</p>
449</div>
450
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]451<!-- *********************************************************************** -->
Reid Spencerc87f4972006-04-26 15:46:53 +0000[diff] [blame]452<div class="doc_section"><a name="felangs">Source Languages</a></div>
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]453
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]454<div class="question">
455<p><a name="langs">What source languages are supported?</a></p>
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]456</div>
Gordon Henriksen58366822008-02-22 20:58:29 +0000[diff] [blame]457
Gordon Henriksen58366822008-02-22 20:58:29 +0000[diff] [blame]458<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]459<p>LLVM currently has full support for C and C++ source languages. These are
460 available through a special version of GCC that LLVM calls the
461 <a href="#cfe">C Front End</a></p>
462
463<p>There is an incomplete version of a Java front end available in the
464 <tt>java</tt> module. There is no documentation on this yet so you'll need to
465 download the code, compile it, and try it.</p>
466
467<p>The PyPy developers are working on integrating LLVM into the PyPy backend so
468 that PyPy language can translate to LLVM.</p>
Gordon Henriksen58366822008-02-22 20:58:29 +0000[diff] [blame]469</div>
470
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]471<div class="question">
472<p><a name="langirgen">I'd like to write a self-hosting LLVM compiler. How
473 should I interface with the LLVM middle-end optimizers and back-end code
474 generators?</a></p>
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]475</div>
Chris Lattner33bef482006-08-15 00:43:35 +0000[diff] [blame]476
Chris Lattner33bef482006-08-15 00:43:35 +0000[diff] [blame]477<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]478<p>Your compiler front-end will communicate with LLVM by creating a module in
479 the LLVM intermediate representation (IR) format. Assuming you want to write
480 your language's compiler in the language itself (rather than C++), there are
481 3 major ways to tackle generating LLVM IR from a front-end:</p>
482
483<ul>
484 <li><strong>Call into the LLVM libraries code using your language's FFI
485 (foreign function interface).</strong>
486
487 <ul>
488 <li><em>for:</em> best tracks changes to the LLVM IR, .ll syntax, and .bc
489 format</li>
490
491 <li><em>for:</em> enables running LLVM optimization passes without a
492 emit/parse overhead</li>
493
494 <li><em>for:</em> adapts well to a JIT context</li>
495
496 <li><em>against:</em> lots of ugly glue code to write</li>
497 </ul></li>
498
499 <li> <strong>Emit LLVM assembly from your compiler's native language.</strong>
500 <ul>
501 <li><em>for:</em> very straightforward to get started</li>
502
503 <li><em>against:</em> the .ll parser is slower than the bitcode reader
504 when interfacing to the middle end</li>
505
506 <li><em>against:</em> you'll have to re-engineer the LLVM IR object model
507 and asm writer in your language</li>
508
509 <li><em>against:</em> it may be harder to track changes to the IR</li>
510 </ul></li>
511
512 <li><strong>Emit LLVM bitcode from your compiler's native language.</strong>
513
514 <ul>
515 <li><em>for:</em> can use the more-efficient bitcode reader when
516 interfacing to the middle end</li>
517
518 <li><em>against:</em> you'll have to re-engineer the LLVM IR object
519 model and bitcode writer in your language</li>
520
521 <li><em>against:</em> it may be harder to track changes to the IR</li>
522 </ul></li>
523</ul>
524
525<p>If you go with the first option, the C bindings in include/llvm-c should help
526 a lot, since most languages have strong support for interfacing with C. The
527 most common hurdle with calling C from managed code is interfacing with the
528 garbage collector. The C interface was designed to require very little memory
529 management, and so is straightforward in this regard.</p>
530</div>
531
532<div class="question">
533<p><a name="langhlsupp">What support is there for a higher level source language
534 constructs for building a compiler?</a></p>
535</div>
536
537<div class="answer">
538<p>Currently, there isn't much. LLVM supports an intermediate representation
539 which is useful for code representation but will not support the high level
540 (abstract syntax tree) representation needed by most compilers. There are no
541 facilities for lexical nor semantic analysis. There is, however, a <i>mostly
542 implemented</i> configuration-driven
543 <a href="CompilerDriver.html">compiler driver</a> which simplifies the task
544 of running optimizations, linking, and executable generation.</p>
545</div>
546
547<div class="question">
548<p><a name="getelementptr">I don't understand the GetElementPtr
549 instruction. Help!</a></p>
550</div>
551
552<div class="answer">
553<p>See <a href="GetElementPtr.html">The Often Misunderstood GEP
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]554 Instruction</a>.</p>
Chris Lattner33bef482006-08-15 00:43:35 +0000[diff] [blame]555</div>
556
Reid Spencer501bfee2006-04-26 14:52:19 +0000[diff] [blame]557<!-- *********************************************************************** -->
558<div class="doc_section">
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]559 <a name="cfe">Using the GCC Front End</a>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]560</div>
561
562<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]563<p>When I compile software that uses a configure script, the configure script
564 thinks my system has all of the header files and libraries it is testing for.
565 How do I get configure to work correctly?</p>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]566</div>
567
568<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]569<p>The configure script is getting things wrong because the LLVM linker allows
570 symbols to be undefined at link time (so that they can be resolved during JIT
571 or translation to the C back end). That is why configure thinks your system
572 "has everything."</p>
573
574<p>To work around this, perform the following steps:</p>
575
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]576<ol>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]577 <li>Make sure the CC and CXX environment variables contains the full path to
578 the LLVM GCC front end.</li>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]579
Reid Spencer434262a2007-02-09 15:59:08 +0000[diff] [blame]580 <li>Make sure that the regular C compiler is first in your PATH. </li>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]581
Reid Spencer434262a2007-02-09 15:59:08 +0000[diff] [blame]582 <li>Add the string "-Wl,-native" to your CFLAGS environment variable.</li>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]583</ol>
584
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]585<p>This will allow the <tt>llvm-ld</tt> linker to create a native code
586 executable instead of shell script that runs the JIT. Creating native code
587 requires standard linkage, which in turn will allow the configure script to
588 find out if code is not linking on your system because the feature isn't
589 available on your system.</p>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]590</div>
591
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]592<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]593<p>When I compile code using the LLVM GCC front end, it complains that it cannot
594 find libcrtend.a.
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]595</p>
596</div>
597
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]598<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]599<p>The only way this can happen is if you haven't installed the runtime
600 library. To correct this, do:</p>
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]601
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]602<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]603% cd llvm/runtime
604% make clean ; make install-bytecode
Reid Spencerf96eb572004-12-15 00:14:01 +0000[diff] [blame]605</pre>
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]606</div>
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]607
Tanya Lattner14fc5c12005-04-25 20:36:56 +0000[diff] [blame]608<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]609<p>How can I disable all optimizations when compiling code using the LLVM GCC
610 front end?</p>
Tanya Lattner14fc5c12005-04-25 20:36:56 +0000[diff] [blame]611</div>
612
613<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]614<p>Passing "-Wa,-disable-opt -Wl,-disable-opt" will disable *all* cleanup and
615 optimizations done at the llvm level, leaving you with the truly horrible
616 code that you desire.</p>
Tanya Lattner14fc5c12005-04-25 20:36:56 +0000[diff] [blame]617</div>
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]618
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]619
620<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]621<p><a name="translatecxx">Can I use LLVM to convert C++ code to C code?</a></p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]622</div>
623
624<div class="answer">
625<p>Yes, you can use LLVM to convert code from any language LLVM supports to C.
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]626 Note that the generated C code will be very low level (all loops are lowered
627 to gotos, etc) and not very pretty (comments are stripped, original source
628 formatting is totally lost, variables are renamed, expressions are
629 regrouped), so this may not be what you're looking for. Also, there are
630 several limitations noted below.<p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]631
632<p>Use commands like this:</p>
633
634<ol>
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]635 <li><p>Compile your program as normal with llvm-g++:</p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]636
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]637<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]638% llvm-g++ x.cpp -o program
639</pre>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]640
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]641 <p>or:</p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]642
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]643<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]644% llvm-g++ a.cpp -c
645% llvm-g++ b.cpp -c
646% llvm-g++ a.o b.o -o program
647</pre>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]648
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]649 <p>With llvm-gcc3, this will generate program and program.bc. The .bc
650 file is the LLVM version of the program all linked together.</p></li>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]651
Bill Wendlinge9a6c352007-09-22 09:54:47 +0000[diff] [blame]652 <li><p>Convert the LLVM code to C code, using the LLC tool with the C
653 backend:</p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]654
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]655<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]656% llc -march=c program.bc -o program.c
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]657</pre></li>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]658
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]659 <li><p>Finally, compile the C file:</p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]660
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]661<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]662% cc x.c
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]663</pre></li>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]664
665</ol>
666
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]667<p>Using LLVM does not eliminate the need for C++ library support. If you use
668 the llvm-g++ front-end, the generated code will depend on g++'s C++ support
669 libraries in the same way that code generated from g++ would. If you use
670 another C++ front-end, the generated code will depend on whatever library
671 that front-end would normally require.</p>
Chris Lattnerb495fb02006-08-31 04:26:31 +0000[diff] [blame]672
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]673<p>If you are working on a platform that does not provide any C++ libraries, you
674 may be able to manually compile libstdc++ to LLVM bitcode, statically link it
675 into your program, then use the commands above to convert the whole result
676 into C code. Alternatively, you might compile the libraries and your
677 application into two different chunks of C code and link them.</p>
Chris Lattnerb495fb02006-08-31 04:26:31 +0000[diff] [blame]678
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]679<p>Note that, by default, the C back end does not support exception handling.
680 If you want/need it for a certain program, you can enable it by passing
681 "-enable-correct-eh-support" to the llc program. The resultant code will use
682 setjmp/longjmp to implement exception support that is relatively slow, and
683 not C++-ABI-conforming on most platforms, but otherwise correct.</p>
Dan Gohmand5b455f2009-01-25 16:04:50 +0000[diff] [blame]684
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]685<p>Also, there are a number of other limitations of the C backend that cause it
686 to produce code that does not fully conform to the C++ ABI on most
687 platforms. Some of the C++ programs in LLVM's test suite are known to fail
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]688 when compiled with the C back end because of ABI incompatibilities with
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]689 standard C++ libraries.</p>
Chris Lattneraf7fd202006-07-19 18:19:59 +0000[diff] [blame]690</div>
691
Dan Gohmancfbcd592009-02-10 17:26:53 +0000[diff] [blame]692<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]693<p><a name="platformindependent">Can I compile C or C++ code to
694 platform-independent LLVM bitcode?</a></p>
Dan Gohmancfbcd592009-02-10 17:26:53 +0000[diff] [blame]695</div>
696
697<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]698<p>No. C and C++ are inherently platform-dependent languages. The most obvious
699 example of this is the preprocessor. A very common way that C code is made
700 portable is by using the preprocessor to include platform-specific code. In
701 practice, information about other platforms is lost after preprocessing, so
702 the result is inherently dependent on the platform that the preprocessing was
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]703 targeting.</p>
Dan Gohmancfbcd592009-02-10 17:26:53 +0000[diff] [blame]704
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]705<p>Another example is <tt>sizeof</tt>. It's common for <tt>sizeof(long)</tt> to
706 vary between platforms. In most C front-ends, <tt>sizeof</tt> is expanded to
Misha Brukman7e0fc8a2009-04-10 20:48:27 +0000[diff] [blame]707 a constant immediately, thus hard-wiring a platform-specific detail.</p>
Dan Gohmancfbcd592009-02-10 17:26:53 +0000[diff] [blame]708
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]709<p>Also, since many platforms define their ABIs in terms of C, and since LLVM is
710 lower-level than C, front-ends currently must emit platform-specific IR in
711 order to have the result conform to the platform ABI.</p>
Dan Gohmancfbcd592009-02-10 17:26:53 +0000[diff] [blame]712</div>
713
Chris Lattnercc33d702003-11-19 05:53:12 +0000[diff] [blame]714<!-- *********************************************************************** -->
715<div class="doc_section">
716 <a name="cfe_code">Questions about code generated by the GCC front-end</a>
717</div>
718
Misha Brukman237dc2a2004-12-03 23:58:18 +0000[diff] [blame]719<div class="question">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]720<p><a name="iosinit">What is this <tt>llvm.global_ctors</tt> and
721 <tt>_GLOBAL__I__tmp_webcompile...</tt> stuff that happens when I <tt>#include
722 &lt;iostream&gt;</tt>?</a></p>
Misha Brukman237dc2a2004-12-03 23:58:18 +0000[diff] [blame]723</div>
Chris Lattnerc50bbc92004-03-29 19:14:35 +0000[diff] [blame]724
725<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]726<p>If you <tt>#include</tt> the <tt>&lt;iostream&gt;</tt> header into a C++
727 translation unit, the file will probably use
728 the <tt>std::cin</tt>/<tt>std::cout</tt>/... global objects. However, C++
729 does not guarantee an order of initialization between static objects in
730 different translation units, so if a static ctor/dtor in your .cpp file
731 used <tt>std::cout</tt>, for example, the object would not necessarily be
732 automatically initialized before your use.</p>
Chris Lattnerc50bbc92004-03-29 19:14:35 +0000[diff] [blame]733
Misha Brukman237dc2a2004-12-03 23:58:18 +0000[diff] [blame]734<p>To make <tt>std::cout</tt> and friends work correctly in these scenarios, the
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]735 STL that we use declares a static object that gets created in every
736 translation unit that includes <tt>&lt;iostream&gt;</tt>. This object has a
737 static constructor and destructor that initializes and destroys the global
738 iostream objects before they could possibly be used in the file. The code
739 that you see in the .ll file corresponds to the constructor and destructor
740 registration code.
Chris Lattnerc50bbc92004-03-29 19:14:35 +0000[diff] [blame]741</p>
742
Misha Brukman237dc2a2004-12-03 23:58:18 +0000[diff] [blame]743<p>If you would like to make it easier to <b>understand</b> the LLVM code
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]744 generated by the compiler in the demo page, consider using <tt>printf()</tt>
745 instead of <tt>iostream</tt>s to print values.</p>
Chris Lattnerc50bbc92004-03-29 19:14:35 +0000[diff] [blame]746</div>
747
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]748<!--=========================================================================-->
749
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]750<div class="question">
751<p><a name="codedce">Where did all of my code go??</a></p>
752</div>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]753
754<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]755<p>If you are using the LLVM demo page, you may often wonder what happened to
756 all of the code that you typed in. Remember that the demo script is running
757 the code through the LLVM optimizers, so if your code doesn't actually do
758 anything useful, it might all be deleted.</p>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]759
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]760<p>To prevent this, make sure that the code is actually needed. For example, if
761 you are computing some expression, return the value from the function instead
762 of leaving it in a local variable. If you really want to constrain the
763 optimizer, you can read from and assign to <tt>volatile</tt> global
764 variables.</p>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]765</div>
766
767<!--=========================================================================-->
768
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]769<div class="question">
770<p><a name="undef">What is this "<tt>undef</tt>" thing that shows up in my
Bill Wendlingb1a61bd2009-04-07 18:52:30 +0000[diff] [blame]771 code?</a></p>
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]772</div>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]773
774<div class="answer">
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]775<p><a href="LangRef.html#undef"><tt>undef</tt></a> is the LLVM way of
776 representing a value that is not defined. You can get these if you do not
777 initialize a variable before you use it. For example, the C function:</p>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]778
Misha Brukmana54d4b22008-12-17 18:11:40 +0000[diff] [blame]779<pre class="doc_code">
Bill Wendlingd6a68eb2007-05-29 09:24:33 +0000[diff] [blame]780int X() { int i; return i; }
781</pre>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]782
Bill Wendling290235f2009-04-07 18:40:56 +0000[diff] [blame]783<p>Is compiled to "<tt>ret i32 undef</tt>" because "<tt>i</tt>" never has a
784 value specified for it.</p>
Chris Lattner5a53c5d2005-02-25 20:30:21 +0000[diff] [blame]785</div>
786
Chris Lattner2c6f9f72009-06-30 17:10:19 +0000[diff] [blame]787<!--=========================================================================-->
788
789<div class="question">
790<p><a name="callconvwrong">Why does instcombine + simplifycfg turn
791 a call to a function with a mismatched calling convention into "unreachable"?
792 Why not make the verifier reject it?</a></p>
793</div>
794
795<div class="answer">
796<p>This is a common problem run into by authors of front-ends that are using
797custom calling conventions: you need to make sure to set the right calling
798convention on both the function and on each call to the function. For example,
799this code:</p>
800
801<pre class="doc_code">
802define fastcc void @foo() {
803 ret void
804}
805define void @bar() {
806 call void @foo( )
807 ret void
808}
809</pre>
810
811<p>Is optimized to:</p>
812
813<pre class="doc_code">
814define fastcc void @foo() {
815 ret void
816}
817define void @bar() {
818 unreachable
819}
820</pre>
821
822<p>... with "opt -instcombine -simplifycfg". This often bites people because
823"all their code disappears". Setting the calling convention on the caller and
824callee is required for indirect calls to work, so people often ask why not make
825the verifier reject this sort of thing.</p>
826
827<p>The answer is that this code has undefined behavior, but it is not illegal.
828If we made it illegal, then every transformation that could potentially create
829this would have to ensure that it doesn't, and there is valid code that can
830create this sort of construct (in dead code). The sorts of things that can
831cause this to happen are fairly contrived, but we still need to accept them.
832Here's an example:</p>
833
834<pre class="doc_code">
835define fastcc void @foo() {
836 ret void
837}
838define internal void @bar(void()* %FP, i1 %cond) {
839 br i1 %cond, label %T, label %F
840T:
841 call void %FP()
842 ret void
843F:
844 call fastcc void %FP()
845 ret void
846}
847define void @test() {
848 %X = or i1 false, false
849 call void @bar(void()* @foo, i1 %X)
850 ret void
851}
852</pre>
853
854<p>In this example, "test" always passes @foo/false into bar, which ensures that
855 it is dynamically called with the right calling conv (thus, the code is
856 perfectly well defined). If you run this through the inliner, you get this
857 (the explicit "or" is there so that the inliner doesn't dead code eliminate
858 a bunch of stuff):
859</p>
860
861<pre class="doc_code">
862define fastcc void @foo() {
863 ret void
864}
865define void @test() {
866 %X = or i1 false, false
867 br i1 %X, label %T.i, label %F.i
868T.i:
869 call void @foo()
870 br label %bar.exit
871F.i:
872 call fastcc void @foo()
873 br label %bar.exit
874bar.exit:
875 ret void
876}
877</pre>
878
879<p>Here you can see that the inlining pass made an undefined call to @foo with
880 the wrong calling convention. We really don't want to make the inliner have
881 to know about this sort of thing, so it needs to be valid code. In this case,
882 dead code elimination can trivially remove the undefined code. However, if %X
883 was an input argument to @test, the inliner would produce this:
884</p>
885
886<pre class="doc_code">
887define fastcc void @foo() {
888 ret void
889}
890
891define void @test(i1 %X) {
892 br i1 %X, label %T.i, label %F.i
893T.i:
894 call void @foo()
895 br label %bar.exit
896F.i:
897 call fastcc void @foo()
898 br label %bar.exit
899bar.exit:
900 ret void
901}
902</pre>
903
904<p>The interesting thing about this is that %X <em>must</em> be false for the
905code to be well-defined, but no amount of dead code elimination will be able to
906delete the broken call as unreachable. However, since instcombine/simplifycfg
907turns the undefined call into unreachable, we end up with a branch on a
908condition that goes to unreachable: a branch to unreachable can never happen, so
909"-inline -instcombine -simplifycfg" is able to produce:</p>
910
911<pre class="doc_code">
912define fastcc void @foo() {
913 ret void
914}
915define void @test(i1 %X) {
916F.i:
917 call fastcc void @foo()
918 ret void
919}
920</pre>
921
922</div>
923
John Criswell6ea30b02003-11-18 16:05:23 +0000[diff] [blame]924<!-- *********************************************************************** -->
John Criswellc310f622003-10-13 16:13:06 +0000[diff] [blame]925
926<hr>
Misha Brukman7ce62cc2004-06-01 18:51:03 +0000[diff] [blame]927<address>
928 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]929 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukman7ce62cc2004-06-01 18:51:03 +0000[diff] [blame]930 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]931 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Misha Brukman7ce62cc2004-06-01 18:51:03 +0000[diff] [blame]932
Reid Spencer05fe4b02006-03-14 05:39:39 +0000[diff] [blame]933 <a href="http://llvm.org">LLVM Compiler Infrastructure</a><br>
Misha Brukmana6538852003-11-06 21:55:44 +0000[diff] [blame]934 Last modified: $Date$
Misha Brukman7ce62cc2004-06-01 18:51:03 +0000[diff] [blame]935</address>
John Criswellf08c5d82003-10-24 22:48:20 +0000[diff] [blame]936
John Criswellc310f622003-10-13 16:13:06 +0000[diff] [blame]937</body>
938</html>