1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
5 <title>LLVM Programmer's Manual</title>
6 <link rel="stylesheet" href="llvm.css" type="text/css">
10 <div class="doc_title">
11 LLVM Programmer's Manual
15 <li><a href="#introduction">Introduction</a></li>
16 <li><a href="#general">General Information</a>
18 <li><a href="#stl">The C++ Standard Template Library</a></li>
20 <li>The <tt>-time-passes</tt> option</li>
21 <li>How to use the LLVM Makefile system</li>
22 <li>How to write a regression test</li>
27 <li><a href="#apis">Important and useful LLVM APIs</a>
29 <li><a href="#isa">The <tt>isa<></tt>, <tt>cast<></tt>
30 and <tt>dyn_cast<></tt> templates</a> </li>
31 <li><a href="#DEBUG">The <tt>DEBUG()</tt> macro and <tt>-debug</tt>
34 <li><a href="#DEBUG_TYPE">Fine grained debug info with <tt>DEBUG_TYPE</tt>
35 and the <tt>-debug-only</tt> option</a> </li>
38 <li><a href="#Statistic">The <tt>Statistic</tt> class & <tt>-stats</tt>
41 <li>The <tt>InstVisitor</tt> template
42 <li>The general graph API
44 <li><a href="#ViewGraph">Viewing graphs while debugging code</a></li>
47 <li><a href="#datastructure">Picking the Right Data Structure for a Task</a>
49 <li><a href="#ds_sequential">Sequential Containers (std::vector, std::list, etc)</a>
51 <li><a href="#dss_fixedarrays">Fixed Size Arrays</a></li>
52 <li><a href="#dss_heaparrays">Heap Allocated Arrays</a></li>
53 <li><a href="#dss_smallvector">"llvm/ADT/SmallVector.h"</a></li>
54 <li><a href="#dss_vector"><vector></a></li>
55 <li><a href="#dss_deque"><deque></a></li>
56 <li><a href="#dss_list"><list></a></li>
57 <li><a href="#dss_ilist">llvm/ADT/ilist</a></li>
59 <li><a href="#ds_set">Set-Like Containers (std::set, SmallSet, SetVector, etc)</a>
61 <li><a href="#dss_sortedvectorset">A sorted 'vector'</a></li>
62 <li><a href="#dss_smallset">"llvm/ADT/SmallSet.h"</a></li>
63 <li><a href="#dss_smallptrset">"llvm/ADT/SmallPtrSet.h"</a></li>
64 <li><a href="#dss_FoldingSet">"llvm/ADT/FoldingSet.h"</a></li>
65 <li><a href="#dss_set"><set></a></li>
66 <li><a href="#dss_setvector">"llvm/ADT/SetVector.h"</a></li>
67 <li><a href="#dss_otherset">Other Options</a></li>
69 <li><a href="#ds_map">Map-Like Containers (std::map, DenseMap, etc)</a></li>
72 <li><a href="#common">Helpful Hints for Common Operations</a>
74 <li><a href="#inspection">Basic Inspection and Traversal Routines</a>
76 <li><a href="#iterate_function">Iterating over the <tt>BasicBlock</tt>s
77 in a <tt>Function</tt></a> </li>
78 <li><a href="#iterate_basicblock">Iterating over the <tt>Instruction</tt>s
79 in a <tt>BasicBlock</tt></a> </li>
80 <li><a href="#iterate_institer">Iterating over the <tt>Instruction</tt>s
81 in a <tt>Function</tt></a> </li>
82 <li><a href="#iterate_convert">Turning an iterator into a
83 class pointer</a> </li>
84 <li><a href="#iterate_complex">Finding call sites: a more
85 complex example</a> </li>
86 <li><a href="#calls_and_invokes">Treating calls and invokes
87 the same way</a> </li>
88 <li><a href="#iterate_chains">Iterating over def-use &
89 use-def chains</a> </li>
92 <li><a href="#simplechanges">Making simple changes</a>
94 <li><a href="#schanges_creating">Creating and inserting new
95 <tt>Instruction</tt>s</a> </li>
96 <li><a href="#schanges_deleting">Deleting <tt>Instruction</tt>s</a> </li>
97 <li><a href="#schanges_replacing">Replacing an <tt>Instruction</tt>
98 with another <tt>Value</tt></a> </li>
102 <li>Working with the Control Flow Graph
104 <li>Accessing predecessors and successors of a <tt>BasicBlock</tt>
112 <li><a href="#advanced">Advanced Topics</a>
114 <li><a href="#TypeResolve">LLVM Type Resolution</a>
116 <li><a href="#BuildRecType">Basic Recursive Type Construction</a></li>
117 <li><a href="#refineAbstractTypeTo">The <tt>refineAbstractTypeTo</tt> method</a></li>
118 <li><a href="#PATypeHolder">The PATypeHolder Class</a></li>
119 <li><a href="#AbstractTypeUser">The AbstractTypeUser Class</a></li>
122 <li><a href="#SymbolTable">The <tt>SymbolTable</tt> class </a></li>
125 <li><a href="#coreclasses">The Core LLVM Class Hierarchy Reference</a>
127 <li><a href="#Type">The <tt>Type</tt> class</a> </li>
128 <li><a href="#Value">The <tt>Value</tt> class</a>
130 <li><a href="#User">The <tt>User</tt> class</a>
132 <li><a href="#Instruction">The <tt>Instruction</tt> class</a>
134 <li><a href="#GetElementPtrInst">The <tt>GetElementPtrInst</tt> class</a></li>
137 <li><a href="#Module">The <tt>Module</tt> class</a></li>
138 <li><a href="#Constant">The <tt>Constant</tt> class</a>
140 <li><a href="#GlobalValue">The <tt>GlobalValue</tt> class</a>
142 <li><a href="#BasicBlock">The <tt>BasicBlock</tt>class</a></li>
143 <li><a href="#Function">The <tt>Function</tt> class</a></li>
144 <li><a href="#GlobalVariable">The <tt>GlobalVariable</tt> class</a></li>
151 <li><a href="#Argument">The <tt>Argument</tt> class</a></li>
158 <div class="doc_author">
159 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>,
160 <a href="mailto:dhurjati@cs.uiuc.edu">Dinakar Dhurjati</a>,
161 <a href="mailto:jstanley@cs.uiuc.edu">Joel Stanley</a>, and
162 <a href="mailto:rspencer@x10sys.com">Reid Spencer</a></p>
165 <!-- *********************************************************************** -->
166 <div class="doc_section">
167 <a name="introduction">Introduction </a>
169 <!-- *********************************************************************** -->
171 <div class="doc_text">
173 <p>This document is meant to highlight some of the important classes and
174 interfaces available in the LLVM source-base. This manual is not
175 intended to explain what LLVM is, how it works, and what LLVM code looks
176 like. It assumes that you know the basics of LLVM and are interested
177 in writing transformations or otherwise analyzing or manipulating the
180 <p>This document should get you oriented so that you can find your
181 way in the continuously growing source code that makes up the LLVM
182 infrastructure. Note that this manual is not intended to serve as a
183 replacement for reading the source code, so if you think there should be
184 a method in one of these classes to do something, but it's not listed,
185 check the source. Links to the <a href="/doxygen/">doxygen</a> sources
186 are provided to make this as easy as possible.</p>
188 <p>The first section of this document describes general information that is
189 useful to know when working in the LLVM infrastructure, and the second describes
190 the Core LLVM classes. In the future this manual will be extended with
191 information describing how to use extension libraries, such as dominator
192 information, CFG traversal routines, and useful utilities like the <tt><a
193 href="/doxygen/InstVisitor_8h-source.html">InstVisitor</a></tt> template.</p>
197 <!-- *********************************************************************** -->
198 <div class="doc_section">
199 <a name="general">General Information</a>
201 <!-- *********************************************************************** -->
203 <div class="doc_text">
205 <p>This section contains general information that is useful if you are working
206 in the LLVM source-base, but that isn't specific to any particular API.</p>
210 <!-- ======================================================================= -->
211 <div class="doc_subsection">
212 <a name="stl">The C++ Standard Template Library</a>
215 <div class="doc_text">
217 <p>LLVM makes heavy use of the C++ Standard Template Library (STL),
218 perhaps much more than you are used to, or have seen before. Because of
219 this, you might want to do a little background reading in the
220 techniques used and capabilities of the library. There are many good
221 pages that discuss the STL, and several books on the subject that you
222 can get, so it will not be discussed in this document.</p>
224 <p>Here are some useful links:</p>
228 <li><a href="http://www.dinkumware.com/refxcpp.html">Dinkumware C++ Library
229 reference</a> - an excellent reference for the STL and other parts of the
230 standard C++ library.</li>
232 <li><a href="http://www.tempest-sw.com/cpp/">C++ In a Nutshell</a> - This is an
233 O'Reilly book in the making. It has a decent
235 Reference that rivals Dinkumware's, and is unfortunately no longer free since the book has been
238 <li><a href="http://www.parashift.com/c++-faq-lite/">C++ Frequently Asked
241 <li><a href="http://www.sgi.com/tech/stl/">SGI's STL Programmer's Guide</a> -
243 href="http://www.sgi.com/tech/stl/stl_introduction.html">Introduction to the
246 <li><a href="http://www.research.att.com/%7Ebs/C++.html">Bjarne Stroustrup's C++
249 <li><a href="http://64.78.49.204/">
250 Bruce Eckel's Thinking in C++, 2nd ed. Volume 2 Revision 4.0 (even better, get
255 <p>You are also encouraged to take a look at the <a
256 href="CodingStandards.html">LLVM Coding Standards</a> guide which focuses on how
257 to write maintainable code more than where to put your curly braces.</p>
261 <!-- ======================================================================= -->
262 <div class="doc_subsection">
263 <a name="stl">Other useful references</a>
266 <div class="doc_text">
269 <li><a href="http://www.psc.edu/%7Esemke/cvs_branches.html">CVS
270 Branch and Tag Primer</a></li>
271 <li><a href="http://www.fortran-2000.com/ArnaudRecipes/sharedlib.html">Using
272 static and shared libraries across platforms</a></li>
277 <!-- *********************************************************************** -->
278 <div class="doc_section">
279 <a name="apis">Important and useful LLVM APIs</a>
281 <!-- *********************************************************************** -->
283 <div class="doc_text">
285 <p>Here we highlight some LLVM APIs that are generally useful and good to
286 know about when writing transformations.</p>
290 <!-- ======================================================================= -->
291 <div class="doc_subsection">
292 <a name="isa">The <tt>isa<></tt>, <tt>cast<></tt> and
293 <tt>dyn_cast<></tt> templates</a>
296 <div class="doc_text">
298 <p>The LLVM source-base makes extensive use of a custom form of RTTI.
299 These templates have many similarities to the C++ <tt>dynamic_cast<></tt>
300 operator, but they don't have some drawbacks (primarily stemming from
301 the fact that <tt>dynamic_cast<></tt> only works on classes that
302 have a v-table). Because they are used so often, you must know what they
303 do and how they work. All of these templates are defined in the <a
304 href="/doxygen/Casting_8h-source.html"><tt>llvm/Support/Casting.h</tt></a>
305 file (note that you very rarely have to include this file directly).</p>
308 <dt><tt>isa<></tt>: </dt>
310 <dd><p>The <tt>isa<></tt> operator works exactly like the Java
311 "<tt>instanceof</tt>" operator. It returns true or false depending on whether
312 a reference or pointer points to an instance of the specified class. This can
313 be very useful for constraint checking of various sorts (example below).</p>
316 <dt><tt>cast<></tt>: </dt>
318 <dd><p>The <tt>cast<></tt> operator is a "checked cast" operation. It
319 converts a pointer or reference from a base class to a derived cast, causing
320 an assertion failure if it is not really an instance of the right type. This
321 should be used in cases where you have some information that makes you believe
322 that something is of the right type. An example of the <tt>isa<></tt>
323 and <tt>cast<></tt> template is:</p>
325 <div class="doc_code">
327 static bool isLoopInvariant(const <a href="#Value">Value</a> *V, const Loop *L) {
328 if (isa<<a href="#Constant">Constant</a>>(V) || isa<<a href="#Argument">Argument</a>>(V) || isa<<a href="#GlobalValue">GlobalValue</a>>(V))
331 // <i>Otherwise, it must be an instruction...</i>
332 return !L->contains(cast<<a href="#Instruction">Instruction</a>>(V)->getParent());
337 <p>Note that you should <b>not</b> use an <tt>isa<></tt> test followed
338 by a <tt>cast<></tt>, for that use the <tt>dyn_cast<></tt>
343 <dt><tt>dyn_cast<></tt>:</dt>
345 <dd><p>The <tt>dyn_cast<></tt> operator is a "checking cast" operation.
346 It checks to see if the operand is of the specified type, and if so, returns a
347 pointer to it (this operator does not work with references). If the operand is
348 not of the correct type, a null pointer is returned. Thus, this works very
349 much like the <tt>dynamic_cast<></tt> operator in C++, and should be
350 used in the same circumstances. Typically, the <tt>dyn_cast<></tt>
351 operator is used in an <tt>if</tt> statement or some other flow control
352 statement like this:</p>
354 <div class="doc_code">
356 if (<a href="#AllocationInst">AllocationInst</a> *AI = dyn_cast<<a href="#AllocationInst">AllocationInst</a>>(Val)) {
362 <p>This form of the <tt>if</tt> statement effectively combines together a call
363 to <tt>isa<></tt> and a call to <tt>cast<></tt> into one
364 statement, which is very convenient.</p>
366 <p>Note that the <tt>dyn_cast<></tt> operator, like C++'s
367 <tt>dynamic_cast<></tt> or Java's <tt>instanceof</tt> operator, can be
368 abused. In particular, you should not use big chained <tt>if/then/else</tt>
369 blocks to check for lots of different variants of classes. If you find
370 yourself wanting to do this, it is much cleaner and more efficient to use the
371 <tt>InstVisitor</tt> class to dispatch over the instruction type directly.</p>
375 <dt><tt>cast_or_null<></tt>: </dt>
377 <dd><p>The <tt>cast_or_null<></tt> operator works just like the
378 <tt>cast<></tt> operator, except that it allows for a null pointer as an
379 argument (which it then propagates). This can sometimes be useful, allowing
380 you to combine several null checks into one.</p></dd>
382 <dt><tt>dyn_cast_or_null<></tt>: </dt>
384 <dd><p>The <tt>dyn_cast_or_null<></tt> operator works just like the
385 <tt>dyn_cast<></tt> operator, except that it allows for a null pointer
386 as an argument (which it then propagates). This can sometimes be useful,
387 allowing you to combine several null checks into one.</p></dd>
391 <p>These five templates can be used with any classes, whether they have a
392 v-table or not. To add support for these templates, you simply need to add
393 <tt>classof</tt> static methods to the class you are interested casting
394 to. Describing this is currently outside the scope of this document, but there
395 are lots of examples in the LLVM source base.</p>
399 <!-- ======================================================================= -->
400 <div class="doc_subsection">
401 <a name="DEBUG">The <tt>DEBUG()</tt> macro and <tt>-debug</tt> option</a>
404 <div class="doc_text">
406 <p>Often when working on your pass you will put a bunch of debugging printouts
407 and other code into your pass. After you get it working, you want to remove
408 it, but you may need it again in the future (to work out new bugs that you run
411 <p> Naturally, because of this, you don't want to delete the debug printouts,
412 but you don't want them to always be noisy. A standard compromise is to comment
413 them out, allowing you to enable them if you need them in the future.</p>
415 <p>The "<tt><a href="/doxygen/Debug_8h-source.html">llvm/Support/Debug.h</a></tt>"
416 file provides a macro named <tt>DEBUG()</tt> that is a much nicer solution to
417 this problem. Basically, you can put arbitrary code into the argument of the
418 <tt>DEBUG</tt> macro, and it is only executed if '<tt>opt</tt>' (or any other
419 tool) is run with the '<tt>-debug</tt>' command line argument:</p>
421 <div class="doc_code">
423 DOUT << "I am here!\n";
427 <p>Then you can run your pass like this:</p>
429 <div class="doc_code">
431 $ opt < a.bc > /dev/null -mypass
432 <i><no output></i>
433 $ opt < a.bc > /dev/null -mypass -debug
438 <p>Using the <tt>DEBUG()</tt> macro instead of a home-brewed solution allows you
439 to not have to create "yet another" command line option for the debug output for
440 your pass. Note that <tt>DEBUG()</tt> macros are disabled for optimized builds,
441 so they do not cause a performance impact at all (for the same reason, they
442 should also not contain side-effects!).</p>
444 <p>One additional nice thing about the <tt>DEBUG()</tt> macro is that you can
445 enable or disable it directly in gdb. Just use "<tt>set DebugFlag=0</tt>" or
446 "<tt>set DebugFlag=1</tt>" from the gdb if the program is running. If the
447 program hasn't been started yet, you can always just run it with
452 <!-- _______________________________________________________________________ -->
453 <div class="doc_subsubsection">
454 <a name="DEBUG_TYPE">Fine grained debug info with <tt>DEBUG_TYPE</tt> and
455 the <tt>-debug-only</tt> option</a>
458 <div class="doc_text">
460 <p>Sometimes you may find yourself in a situation where enabling <tt>-debug</tt>
461 just turns on <b>too much</b> information (such as when working on the code
462 generator). If you want to enable debug information with more fine-grained
463 control, you define the <tt>DEBUG_TYPE</tt> macro and the <tt>-debug</tt> only
464 option as follows:</p>
466 <div class="doc_code">
468 DOUT << "No debug type\n";
470 #define DEBUG_TYPE "foo"
471 DOUT << "'foo' debug type\n";
473 #define DEBUG_TYPE "bar"
474 DOUT << "'bar' debug type\n";
476 #define DEBUG_TYPE ""
477 DOUT << "No debug type (2)\n";
481 <p>Then you can run your pass like this:</p>
483 <div class="doc_code">
485 $ opt < a.bc > /dev/null -mypass
486 <i><no output></i>
487 $ opt < a.bc > /dev/null -mypass -debug
492 $ opt < a.bc > /dev/null -mypass -debug-only=foo
494 $ opt < a.bc > /dev/null -mypass -debug-only=bar
499 <p>Of course, in practice, you should only set <tt>DEBUG_TYPE</tt> at the top of
500 a file, to specify the debug type for the entire module (if you do this before
501 you <tt>#include "llvm/Support/Debug.h"</tt>, you don't have to insert the ugly
502 <tt>#undef</tt>'s). Also, you should use names more meaningful than "foo" and
503 "bar", because there is no system in place to ensure that names do not
504 conflict. If two different modules use the same string, they will all be turned
505 on when the name is specified. This allows, for example, all debug information
506 for instruction scheduling to be enabled with <tt>-debug-type=InstrSched</tt>,
507 even if the source lives in multiple files.</p>
511 <!-- ======================================================================= -->
512 <div class="doc_subsection">
513 <a name="Statistic">The <tt>Statistic</tt> class & <tt>-stats</tt>
517 <div class="doc_text">
520 href="/doxygen/Statistic_8h-source.html">llvm/ADT/Statistic.h</a></tt>" file
521 provides a class named <tt>Statistic</tt> that is used as a unified way to
522 keep track of what the LLVM compiler is doing and how effective various
523 optimizations are. It is useful to see what optimizations are contributing to
524 making a particular program run faster.</p>
526 <p>Often you may run your pass on some big program, and you're interested to see
527 how many times it makes a certain transformation. Although you can do this with
528 hand inspection, or some ad-hoc method, this is a real pain and not very useful
529 for big programs. Using the <tt>Statistic</tt> class makes it very easy to
530 keep track of this information, and the calculated information is presented in a
531 uniform manner with the rest of the passes being executed.</p>
533 <p>There are many examples of <tt>Statistic</tt> uses, but the basics of using
534 it are as follows:</p>
537 <li><p>Define your statistic like this:</p>
539 <div class="doc_code">
541 #define <a href="#DEBUG_TYPE">DEBUG_TYPE</a> "mypassname" <i>// This goes before any #includes.</i>
542 STATISTIC(NumXForms, "The # of times I did stuff");
546 <p>The <tt>STATISTIC</tt> macro defines a static variable, whose name is
547 specified by the first argument. The pass name is taken from the DEBUG_TYPE
548 macro, and the description is taken from the second argument. The variable
549 defined ("NumXForms" in this case) acts like an unsigned integer.</p></li>
551 <li><p>Whenever you make a transformation, bump the counter:</p>
553 <div class="doc_code">
555 ++NumXForms; // <i>I did stuff!</i>
562 <p>That's all you have to do. To get '<tt>opt</tt>' to print out the
563 statistics gathered, use the '<tt>-stats</tt>' option:</p>
565 <div class="doc_code">
567 $ opt -stats -mypassname < program.bc > /dev/null
568 <i>... statistics output ...</i>
572 <p> When running <tt>gccas</tt> on a C file from the SPEC benchmark
573 suite, it gives a report that looks like this:</p>
575 <div class="doc_code">
577 7646 bytecodewriter - Number of normal instructions
578 725 bytecodewriter - Number of oversized instructions
579 129996 bytecodewriter - Number of bytecode bytes written
580 2817 raise - Number of insts DCEd or constprop'd
581 3213 raise - Number of cast-of-self removed
582 5046 raise - Number of expression trees converted
583 75 raise - Number of other getelementptr's formed
584 138 raise - Number of load/store peepholes
585 42 deadtypeelim - Number of unused typenames removed from symtab
586 392 funcresolve - Number of varargs functions resolved
587 27 globaldce - Number of global variables removed
588 2 adce - Number of basic blocks removed
589 134 cee - Number of branches revectored
590 49 cee - Number of setcc instruction eliminated
591 532 gcse - Number of loads removed
592 2919 gcse - Number of instructions removed
593 86 indvars - Number of canonical indvars added
594 87 indvars - Number of aux indvars removed
595 25 instcombine - Number of dead inst eliminate
596 434 instcombine - Number of insts combined
597 248 licm - Number of load insts hoisted
598 1298 licm - Number of insts hoisted to a loop pre-header
599 3 licm - Number of insts hoisted to multiple loop preds (bad, no loop pre-header)
600 75 mem2reg - Number of alloca's promoted
601 1444 cfgsimplify - Number of blocks simplified
605 <p>Obviously, with so many optimizations, having a unified framework for this
606 stuff is very nice. Making your pass fit well into the framework makes it more
607 maintainable and useful.</p>
611 <!-- ======================================================================= -->
612 <div class="doc_subsection">
613 <a name="ViewGraph">Viewing graphs while debugging code</a>
616 <div class="doc_text">
618 <p>Several of the important data structures in LLVM are graphs: for example
619 CFGs made out of LLVM <a href="#BasicBlock">BasicBlock</a>s, CFGs made out of
620 LLVM <a href="CodeGenerator.html#machinebasicblock">MachineBasicBlock</a>s, and
621 <a href="CodeGenerator.html#selectiondag_intro">Instruction Selection
622 DAGs</a>. In many cases, while debugging various parts of the compiler, it is
623 nice to instantly visualize these graphs.</p>
625 <p>LLVM provides several callbacks that are available in a debug build to do
626 exactly that. If you call the <tt>Function::viewCFG()</tt> method, for example,
627 the current LLVM tool will pop up a window containing the CFG for the function
628 where each basic block is a node in the graph, and each node contains the
629 instructions in the block. Similarly, there also exists
630 <tt>Function::viewCFGOnly()</tt> (does not include the instructions), the
631 <tt>MachineFunction::viewCFG()</tt> and <tt>MachineFunction::viewCFGOnly()</tt>,
632 and the <tt>SelectionDAG::viewGraph()</tt> methods. Within GDB, for example,
633 you can usually use something like <tt>call DAG.viewGraph()</tt> to pop
634 up a window. Alternatively, you can sprinkle calls to these functions in your
635 code in places you want to debug.</p>
637 <p>Getting this to work requires a small amount of configuration. On Unix
638 systems with X11, install the <a href="http://www.graphviz.org">graphviz</a>
639 toolkit, and make sure 'dot' and 'gv' are in your path. If you are running on
640 Mac OS/X, download and install the Mac OS/X <a
641 href="http://www.pixelglow.com/graphviz/">Graphviz program</a>, and add
642 <tt>/Applications/Graphviz.app/Contents/MacOS/</tt> (or whereever you install
643 it) to your path. Once in your system and path are set up, rerun the LLVM
644 configure script and rebuild LLVM to enable this functionality.</p>
646 <p><tt>SelectionDAG</tt> has been extended to make it easier to locate
647 <i>interesting</i> nodes in large complex graphs. From gdb, if you
648 <tt>call DAG.setGraphColor(<i>node</i>, "<i>color</i>")</tt>, then the
649 next <tt>call DAG.viewGraph()</tt> would hilight the node in the
650 specified color (choices of colors can be found at <a
651 href="http://www.graphviz.org/doc/info/colors.html">colors</a>.) More
652 complex node attributes can be provided with <tt>call
653 DAG.setGraphAttrs(<i>node</i>, "<i>attributes</i>")</tt> (choices can be
654 found at <a href="http://www.graphviz.org/doc/info/attrs.html">Graph
655 Attributes</a>.) If you want to restart and clear all the current graph
656 attributes, then you can <tt>call DAG.clearGraphAttrs()</tt>. </p>
660 <!-- *********************************************************************** -->
661 <div class="doc_section">
662 <a name="datastructure">Picking the Right Data Structure for a Task</a>
664 <!-- *********************************************************************** -->
666 <div class="doc_text">
668 <p>LLVM has a plethora of datastructures in the <tt>llvm/ADT/</tt> directory,
669 and we commonly use STL datastructures. This section describes the tradeoffs
670 you should consider when you pick one.</p>
673 The first step is a choose your own adventure: do you want a sequential
674 container, a set-like container, or a map-like container? The most important
675 thing when choosing a container is the algorithmic properties of how you plan to
676 access the container. Based on that, you should use:</p>
679 <li>a <a href="#ds_map">map-like</a> container if you need efficient lookup
680 of an value based on another value. Map-like containers also support
681 efficient queries for containment (whether a key is in the map). Map-like
682 containers generally do not support efficient reverse mapping (values to
683 keys). If you need that, use two maps. Some map-like containers also
684 support efficient iteration through the keys in sorted order. Map-like
685 containers are the most expensive sort, only use them if you need one of
686 these capabilities.</li>
688 <li>a <a href="#ds_set">set-like</a> container if you need to put a bunch of
689 stuff into a container that automatically eliminates duplicates. Some
690 set-like containers support efficient iteration through the elements in
691 sorted order. Set-like containers are more expensive than sequential
695 <li>a <a href="#ds_sequential">sequential</a> container provides
696 the most efficient way to add elements and keeps track of the order they are
697 added to the collection. They permit duplicates and support efficient
698 iteration, but do not support efficient lookup based on a key.
704 Once the proper catagory of container is determined, you can fine tune the
705 memory use, constant factors, and cache behaviors of access by intelligently
706 picking a member of the catagory. Note that constant factors and cache behavior
707 can be a big deal. If you have a vector that usually only contains a few
708 elements (but could contain many), for example, it's much better to use
709 <a href="#dss_smallvector">SmallVector</a> than <a href="#dss_vector">vector</a>
710 . Doing so avoids (relatively) expensive malloc/free calls, which dwarf the
711 cost of adding the elements to the container. </p>
715 <!-- ======================================================================= -->
716 <div class="doc_subsection">
717 <a name="ds_sequential">Sequential Containers (std::vector, std::list, etc)</a>
720 <div class="doc_text">
721 There are a variety of sequential containers available for you, based on your
722 needs. Pick the first in this section that will do what you want.
725 <!-- _______________________________________________________________________ -->
726 <div class="doc_subsubsection">
727 <a name="dss_fixedarrays">Fixed Size Arrays</a>
730 <div class="doc_text">
731 <p>Fixed size arrays are very simple and very fast. They are good if you know
732 exactly how many elements you have, or you have a (low) upper bound on how many
736 <!-- _______________________________________________________________________ -->
737 <div class="doc_subsubsection">
738 <a name="dss_heaparrays">Heap Allocated Arrays</a>
741 <div class="doc_text">
742 <p>Heap allocated arrays (new[] + delete[]) are also simple. They are good if
743 the number of elements is variable, if you know how many elements you will need
744 before the array is allocated, and if the array is usually large (if not,
745 consider a <a href="#dss_smallvector">SmallVector</a>). The cost of a heap
746 allocated array is the cost of the new/delete (aka malloc/free). Also note that
747 if you are allocating an array of a type with a constructor, the constructor and
748 destructors will be run for every element in the array (resizable vectors only
749 construct those elements actually used).</p>
752 <!-- _______________________________________________________________________ -->
753 <div class="doc_subsubsection">
754 <a name="dss_smallvector">"llvm/ADT/SmallVector.h"</a>
757 <div class="doc_text">
758 <p><tt>SmallVector<Type, N></tt> is a simple class that looks and smells
759 just like <tt>vector<Type></tt>:
760 it supports efficient iteration, lays out elements in memory order (so you can
761 do pointer arithmetic between elements), supports efficient push_back/pop_back
762 operations, supports efficient random access to its elements, etc.</p>
764 <p>The advantage of SmallVector is that it allocates space for
765 some number of elements (N) <b>in the object itself</b>. Because of this, if
766 the SmallVector is dynamically smaller than N, no malloc is performed. This can
767 be a big win in cases where the malloc/free call is far more expensive than the
768 code that fiddles around with the elements.</p>
770 <p>This is good for vectors that are "usually small" (e.g. the number of
771 predecessors/successors of a block is usually less than 8). On the other hand,
772 this makes the size of the SmallVector itself large, so you don't want to
773 allocate lots of them (doing so will waste a lot of space). As such,
774 SmallVectors are most useful when on the stack.</p>
776 <p>SmallVector also provides a nice portable and efficient replacement for
781 <!-- _______________________________________________________________________ -->
782 <div class="doc_subsubsection">
783 <a name="dss_vector"><vector></a>
786 <div class="doc_text">
788 std::vector is well loved and respected. It is useful when SmallVector isn't:
789 when the size of the vector is often large (thus the small optimization will
790 rarely be a benefit) or if you will be allocating many instances of the vector
791 itself (which would waste space for elements that aren't in the container).
792 vector is also useful when interfacing with code that expects vectors :).
796 <!-- _______________________________________________________________________ -->
797 <div class="doc_subsubsection">
798 <a name="dss_deque"><deque></a>
801 <div class="doc_text">
802 <p>std::deque is, in some senses, a generalized version of std::vector. Like
803 std::vector, it provides constant time random access and other similar
804 properties, but it also provides efficient access to the front of the list. It
805 does not guarantee continuity of elements within memory.</p>
807 <p>In exchange for this extra flexibility, std::deque has significantly higher
808 constant factor costs than std::vector. If possible, use std::vector or
809 something cheaper.</p>
812 <!-- _______________________________________________________________________ -->
813 <div class="doc_subsubsection">
814 <a name="dss_list"><list></a>
817 <div class="doc_text">
818 <p>std::list is an extremely inefficient class that is rarely useful.
819 It performs a heap allocation for every element inserted into it, thus having an
820 extremely high constant factor, particularly for small data types. std::list
821 also only supports bidirectional iteration, not random access iteration.</p>
823 <p>In exchange for this high cost, std::list supports efficient access to both
824 ends of the list (like std::deque, but unlike std::vector or SmallVector). In
825 addition, the iterator invalidation characteristics of std::list are stronger
826 than that of a vector class: inserting or removing an element into the list does
827 not invalidate iterator or pointers to other elements in the list.</p>
830 <!-- _______________________________________________________________________ -->
831 <div class="doc_subsubsection">
832 <a name="dss_ilist">llvm/ADT/ilist</a>
835 <div class="doc_text">
836 <p><tt>ilist<T></tt> implements an 'intrusive' doubly-linked list. It is
837 intrusive, because it requires the element to store and provide access to the
838 prev/next pointers for the list.</p>
840 <p>ilist has the same drawbacks as std::list, and additionally requires an
841 ilist_traits implementation for the element type, but it provides some novel
842 characteristics. In particular, it can efficiently store polymorphic objects,
843 the traits class is informed when an element is inserted or removed from the
844 list, and ilists are guaranteed to support a constant-time splice operation.
847 <p>These properties are exactly what we want for things like Instructions and
848 basic blocks, which is why these are implemented with ilists.</p>
851 <!-- _______________________________________________________________________ -->
852 <div class="doc_subsubsection">
853 <a name="dss_other">Other options</a>
856 <div class="doc_text">
857 <p>Other STL containers are available, such as std::string.</p>
859 <p>There are also various STL adapter classes such as std::queue,
860 std::priority_queue, std::stack, etc. These provide simplified access to an
861 underlying container but don't affect the cost of the container itself.</p>
866 <!-- ======================================================================= -->
867 <div class="doc_subsection">
868 <a name="ds_set">Set-Like Containers (std::set, SmallSet, SetVector, etc)</a>
871 <div class="doc_text">
873 <p>Set-like containers are useful when you need to canonicalize multiple values
874 into a single representation. There are several different choices for how to do
875 this, providing various trade-offs.</p>
880 <!-- _______________________________________________________________________ -->
881 <div class="doc_subsubsection">
882 <a name="dss_sortedvectorset">A sorted 'vector'</a>
885 <div class="doc_text">
887 <p>If you intend to insert a lot of elements, then do a lot of queries, one
888 great approach is to use a vector (or other sequential container), and then use
889 std::sort+std::unique to remove duplicates. This approach works really well if
890 your usage pattern has these two distinct phases (insert then query), and,
891 coupled with a good choice of <a href="#ds_sequential">sequential container</a>
892 can provide the several nice properties: the result data is contiguous in memory
893 (good for cache locality), has few allocations, is easy to address (iterators in
894 the final vector are just indices or pointers), and can be efficiently queried
895 with a standard binary search.</p>
899 <!-- _______________________________________________________________________ -->
900 <div class="doc_subsubsection">
901 <a name="dss_smallset">"llvm/ADT/SmallSet.h"</a>
904 <div class="doc_text">
906 <p>If you have a set-like datastructure that is usually small and whose elements
907 are reasonably small, a <tt>SmallSet<Type, N> is a good choice. This set
908 has space for N elements in place (thus, if the set is dynamically smaller than
909 N, no malloc traffic is required) and access them with a simple linear search.
910 When the set grows beyond 'N', it allocates a more expensive representation that
911 guarantees efficient access (for most types, it falls back to std::set, but for
912 pointers it uses something far better, see <a
913 href="#dss_smallptrset">SmallPtrSet</a>).</p>
915 <p>The magic of this class is that it handles small sets extremely efficiently,
916 but gracefully handles extremely large sets without loss of efficiency. The
917 drawback is that the interface is quite small: it supports insertion, queries
918 and erasing, but does not support iteration.</p>
922 <!-- _______________________________________________________________________ -->
923 <div class="doc_subsubsection">
924 <a name="dss_smallptrset">"llvm/ADT/SmallPtrSet.h"</a>
927 <div class="doc_text">
929 <p>SmallPtrSet has all the advantages of SmallSet (and a SmallSet of pointers is
930 transparently implemented with a SmallPtrSet), but also suports iterators. If
931 more than 'N' allocations are performed, a single quadratically
932 probed hash table is allocated and grows as needed, providing extremely
933 efficient access (constant time insertion/deleting/queries with low constant
934 factors) and is very stingy with malloc traffic.</p>
936 <p>Note that, unlike std::set, the iterators of SmallPtrSet are invalidated
937 whenever an insertion occurs. Also, the values visited by the iterators are not
938 visited in sorted order.</p>
942 <!-- _______________________________________________________________________ -->
943 <div class="doc_subsubsection">
944 <a name="dss_FoldingSet">"llvm/ADT/FoldingSet.h"</a>
947 <div class="doc_text">
950 FoldingSet is an aggregate class that is really good at uniquing
951 expensive-to-create or polymorphic objects. It is a combination of a chained
952 hash table with intrusive links (uniqued objects are required to inherit from
953 FoldingSetNode) that uses SmallVector as part of its ID process.</p>
955 <p>Consider a case where you want to implement a "getorcreate_foo" method for
956 a complex object (for example, a node in the code generator). The client has a
957 description of *what* it wants to generate (it knows the opcode and all the
958 operands), but we don't want to 'new' a node, then try inserting it into a set
959 only to find out it already exists (at which point we would have to delete it
960 and return the node that already exists).
963 <p>To support this style of client, FoldingSet perform a query with a
964 FoldingSetNodeID (which wraps SmallVector) that can be used to describe the
965 element that we want to query for. The query either returns the element
966 matching the ID or it returns an opaque ID that indicates where insertion should
969 <p>Because FoldingSet uses intrusive links, it can support polymorphic objects
970 in the set (for example, you can have SDNode instances mixed with LoadSDNodes).
971 Because the elements are individually allocated, pointers to the elements are
972 stable: inserting or removing elements does not invalidate any pointers to other
978 <!-- _______________________________________________________________________ -->
979 <div class="doc_subsubsection">
980 <a name="dss_set"><set></a>
983 <div class="doc_text">
985 <p>std::set is a reasonable all-around set class, which is good at many things
986 but great at nothing. std::set use a allocates memory for every single element
987 inserted (thus it is very malloc intensive) and typically stores three pointers
988 with every element (thus adding a large amount of per-element space overhead).
989 It offers guaranteed log(n) performance, which is not particularly fast.
992 <p>The advantages of std::set is that its iterators are stable (deleting or
993 inserting an element from the set does not affect iterators or pointers to other
994 elements) and that iteration over the set is guaranteed to be in sorted order.
995 If the elements in the set are large, then the relative overhead of the pointers
996 and malloc traffic is not a big deal, but if the elements of the set are small,
997 std::set is almost never a good choice.</p>
1001 <!-- _______________________________________________________________________ -->
1002 <div class="doc_subsubsection">
1003 <a name="dss_setvector">"llvm/ADT/SetVector.h"</a>
1006 <div class="doc_text">
1007 <p>LLVM's SetVector<Type> is actually a combination of a set along with
1008 a <a href="#ds_sequential">Sequential Container</a>. The important property
1009 that this provides is efficient insertion with uniquing (duplicate elements are
1010 ignored) with iteration support. It implements this by inserting elements into
1011 both a set-like container and the sequential container, using the set-like
1012 container for uniquing and the sequential container for iteration.
1015 <p>The difference between SetVector and other sets is that the order of
1016 iteration is guaranteed to match the order of insertion into the SetVector.
1017 This property is really important for things like sets of pointers. Because
1018 pointer values are non-deterministic (e.g. vary across runs of the program on
1019 different machines), iterating over the pointers in a std::set or other set will
1020 not be in a well-defined order.</p>
1023 The drawback of SetVector is that it requires twice as much space as a normal
1024 set and has the sum of constant factors from the set-like container and the
1025 sequential container that it uses. Use it *only* if you need to iterate over
1026 the elements in a deterministic order. SetVector is also expensive to delete
1027 elements out of (linear time).
1032 <!-- _______________________________________________________________________ -->
1033 <div class="doc_subsubsection">
1034 <a name="dss_otherset">Other Options</a>
1037 <div class="doc_text">
1040 The STL provides several other options, such as std::multiset and the various
1041 "hash_set" like containers (whether from C++TR1 or from the SGI library).</p>
1043 <p>std::multiset is useful if you're not interested in elimination of
1044 duplicates, but has all the drawbacks of std::set. A sorted vector or some
1045 other approach is almost always better.</p>
1047 <p>The various hash_set implementations (exposed portably by
1048 "llvm/ADT/hash_set") is a standard chained hashtable. This algorithm is malloc
1049 intensive like std::set (performing an allocation for each element inserted,
1050 thus having really high constant factors) but (usually) provides O(1)
1051 insertion/deletion of elements. This can be useful if your elements are large
1052 (thus making the constant-factor cost relatively low). Element iteration does
1053 not visit elements in a useful order.</p>
1057 <!-- ======================================================================= -->
1058 <div class="doc_subsection">
1059 <a name="ds_map">Map-Like Containers (std::map, DenseMap, etc)</a>
1062 <div class="doc_text">
1073 <!-- *********************************************************************** -->
1074 <div class="doc_section">
1075 <a name="common">Helpful Hints for Common Operations</a>
1077 <!-- *********************************************************************** -->
1079 <div class="doc_text">
1081 <p>This section describes how to perform some very simple transformations of
1082 LLVM code. This is meant to give examples of common idioms used, showing the
1083 practical side of LLVM transformations. <p> Because this is a "how-to" section,
1084 you should also read about the main classes that you will be working with. The
1085 <a href="#coreclasses">Core LLVM Class Hierarchy Reference</a> contains details
1086 and descriptions of the main classes that you should know about.</p>
1090 <!-- NOTE: this section should be heavy on example code -->
1091 <!-- ======================================================================= -->
1092 <div class="doc_subsection">
1093 <a name="inspection">Basic Inspection and Traversal Routines</a>
1096 <div class="doc_text">
1098 <p>The LLVM compiler infrastructure have many different data structures that may
1099 be traversed. Following the example of the C++ standard template library, the
1100 techniques used to traverse these various data structures are all basically the
1101 same. For a enumerable sequence of values, the <tt>XXXbegin()</tt> function (or
1102 method) returns an iterator to the start of the sequence, the <tt>XXXend()</tt>
1103 function returns an iterator pointing to one past the last valid element of the
1104 sequence, and there is some <tt>XXXiterator</tt> data type that is common
1105 between the two operations.</p>
1107 <p>Because the pattern for iteration is common across many different aspects of
1108 the program representation, the standard template library algorithms may be used
1109 on them, and it is easier to remember how to iterate. First we show a few common
1110 examples of the data structures that need to be traversed. Other data
1111 structures are traversed in very similar ways.</p>
1115 <!-- _______________________________________________________________________ -->
1116 <div class="doc_subsubsection">
1117 <a name="iterate_function">Iterating over the </a><a
1118 href="#BasicBlock"><tt>BasicBlock</tt></a>s in a <a
1119 href="#Function"><tt>Function</tt></a>
1122 <div class="doc_text">
1124 <p>It's quite common to have a <tt>Function</tt> instance that you'd like to
1125 transform in some way; in particular, you'd like to manipulate its
1126 <tt>BasicBlock</tt>s. To facilitate this, you'll need to iterate over all of
1127 the <tt>BasicBlock</tt>s that constitute the <tt>Function</tt>. The following is
1128 an example that prints the name of a <tt>BasicBlock</tt> and the number of
1129 <tt>Instruction</tt>s it contains:</p>
1131 <div class="doc_code">
1133 // <i>func is a pointer to a Function instance</i>
1134 for (Function::iterator i = func->begin(), e = func->end(); i != e; ++i)
1135 // <i>Print out the name of the basic block if it has one, and then the</i>
1136 // <i>number of instructions that it contains</i>
1137 llvm::cerr << "Basic block (name=" << i->getName() << ") has "
1138 << i->size() << " instructions.\n";
1142 <p>Note that i can be used as if it were a pointer for the purposes of
1143 invoking member functions of the <tt>Instruction</tt> class. This is
1144 because the indirection operator is overloaded for the iterator
1145 classes. In the above code, the expression <tt>i->size()</tt> is
1146 exactly equivalent to <tt>(*i).size()</tt> just like you'd expect.</p>
1150 <!-- _______________________________________________________________________ -->
1151 <div class="doc_subsubsection">
1152 <a name="iterate_basicblock">Iterating over the </a><a
1153 href="#Instruction"><tt>Instruction</tt></a>s in a <a
1154 href="#BasicBlock"><tt>BasicBlock</tt></a>
1157 <div class="doc_text">
1159 <p>Just like when dealing with <tt>BasicBlock</tt>s in <tt>Function</tt>s, it's
1160 easy to iterate over the individual instructions that make up
1161 <tt>BasicBlock</tt>s. Here's a code snippet that prints out each instruction in
1162 a <tt>BasicBlock</tt>:</p>
1164 <div class="doc_code">
1166 // <i>blk is a pointer to a BasicBlock instance</i>
1167 for (BasicBlock::iterator i = blk->begin(), e = blk->end(); i != e; ++i)
1168 // <i>The next statement works since operator<<(ostream&,...)</i>
1169 // <i>is overloaded for Instruction&</i>
1170 llvm::cerr << *i << "\n";
1174 <p>However, this isn't really the best way to print out the contents of a
1175 <tt>BasicBlock</tt>! Since the ostream operators are overloaded for virtually
1176 anything you'll care about, you could have just invoked the print routine on the
1177 basic block itself: <tt>llvm::cerr << *blk << "\n";</tt>.</p>
1181 <!-- _______________________________________________________________________ -->
1182 <div class="doc_subsubsection">
1183 <a name="iterate_institer">Iterating over the </a><a
1184 href="#Instruction"><tt>Instruction</tt></a>s in a <a
1185 href="#Function"><tt>Function</tt></a>
1188 <div class="doc_text">
1190 <p>If you're finding that you commonly iterate over a <tt>Function</tt>'s
1191 <tt>BasicBlock</tt>s and then that <tt>BasicBlock</tt>'s <tt>Instruction</tt>s,
1192 <tt>InstIterator</tt> should be used instead. You'll need to include <a
1193 href="/doxygen/InstIterator_8h-source.html"><tt>llvm/Support/InstIterator.h</tt></a>,
1194 and then instantiate <tt>InstIterator</tt>s explicitly in your code. Here's a
1195 small example that shows how to dump all instructions in a function to the standard error stream:<p>
1197 <div class="doc_code">
1199 #include "<a href="/doxygen/InstIterator_8h-source.html">llvm/Support/InstIterator.h</a>"
1201 // <i>F is a ptr to a Function instance</i>
1202 for (inst_iterator i = inst_begin(F), e = inst_end(F); i != e; ++i)
1203 llvm::cerr << *i << "\n";
1207 <p>Easy, isn't it? You can also use <tt>InstIterator</tt>s to fill a
1208 worklist with its initial contents. For example, if you wanted to
1209 initialize a worklist to contain all instructions in a <tt>Function</tt>
1210 F, all you would need to do is something like:</p>
1212 <div class="doc_code">
1214 std::set<Instruction*> worklist;
1215 worklist.insert(inst_begin(F), inst_end(F));
1219 <p>The STL set <tt>worklist</tt> would now contain all instructions in the
1220 <tt>Function</tt> pointed to by F.</p>
1224 <!-- _______________________________________________________________________ -->
1225 <div class="doc_subsubsection">
1226 <a name="iterate_convert">Turning an iterator into a class pointer (and
1230 <div class="doc_text">
1232 <p>Sometimes, it'll be useful to grab a reference (or pointer) to a class
1233 instance when all you've got at hand is an iterator. Well, extracting
1234 a reference or a pointer from an iterator is very straight-forward.
1235 Assuming that <tt>i</tt> is a <tt>BasicBlock::iterator</tt> and <tt>j</tt>
1236 is a <tt>BasicBlock::const_iterator</tt>:</p>
1238 <div class="doc_code">
1240 Instruction& inst = *i; // <i>Grab reference to instruction reference</i>
1241 Instruction* pinst = &*i; // <i>Grab pointer to instruction reference</i>
1242 const Instruction& inst = *j;
1246 <p>However, the iterators you'll be working with in the LLVM framework are
1247 special: they will automatically convert to a ptr-to-instance type whenever they
1248 need to. Instead of dereferencing the iterator and then taking the address of
1249 the result, you can simply assign the iterator to the proper pointer type and
1250 you get the dereference and address-of operation as a result of the assignment
1251 (behind the scenes, this is a result of overloading casting mechanisms). Thus
1252 the last line of the last example,</p>
1254 <div class="doc_code">
1256 Instruction* pinst = &*i;
1260 <p>is semantically equivalent to</p>
1262 <div class="doc_code">
1264 Instruction* pinst = i;
1268 <p>It's also possible to turn a class pointer into the corresponding iterator,
1269 and this is a constant time operation (very efficient). The following code
1270 snippet illustrates use of the conversion constructors provided by LLVM
1271 iterators. By using these, you can explicitly grab the iterator of something
1272 without actually obtaining it via iteration over some structure:</p>
1274 <div class="doc_code">
1276 void printNextInstruction(Instruction* inst) {
1277 BasicBlock::iterator it(inst);
1278 ++it; // <i>After this line, it refers to the instruction after *inst</i>
1279 if (it != inst->getParent()->end()) llvm::cerr << *it << "\n";
1286 <!--_______________________________________________________________________-->
1287 <div class="doc_subsubsection">
1288 <a name="iterate_complex">Finding call sites: a slightly more complex
1292 <div class="doc_text">
1294 <p>Say that you're writing a FunctionPass and would like to count all the
1295 locations in the entire module (that is, across every <tt>Function</tt>) where a
1296 certain function (i.e., some <tt>Function</tt>*) is already in scope. As you'll
1297 learn later, you may want to use an <tt>InstVisitor</tt> to accomplish this in a
1298 much more straight-forward manner, but this example will allow us to explore how
1299 you'd do it if you didn't have <tt>InstVisitor</tt> around. In pseudocode, this
1300 is what we want to do:</p>
1302 <div class="doc_code">
1304 initialize callCounter to zero
1305 for each Function f in the Module
1306 for each BasicBlock b in f
1307 for each Instruction i in b
1308 if (i is a CallInst and calls the given function)
1309 increment callCounter
1313 <p>And the actual code is (remember, because we're writing a
1314 <tt>FunctionPass</tt>, our <tt>FunctionPass</tt>-derived class simply has to
1315 override the <tt>runOnFunction</tt> method):</p>
1317 <div class="doc_code">
1319 Function* targetFunc = ...;
1321 class OurFunctionPass : public FunctionPass {
1323 OurFunctionPass(): callCounter(0) { }
1325 virtual runOnFunction(Function& F) {
1326 for (Function::iterator b = F.begin(), be = F.end(); b != be; ++b) {
1327 for (BasicBlock::iterator i = b->begin(); ie = b->end(); i != ie; ++i) {
1328 if (<a href="#CallInst">CallInst</a>* callInst = <a href="#isa">dyn_cast</a><<a
1329 href="#CallInst">CallInst</a>>(&*i)) {
1330 // <i>We know we've encountered a call instruction, so we</i>
1331 // <i>need to determine if it's a call to the</i>
1332 // <i>function pointed to by m_func or not</i>
1334 if (callInst->getCalledFunction() == targetFunc)
1342 unsigned callCounter;
1349 <!--_______________________________________________________________________-->
1350 <div class="doc_subsubsection">
1351 <a name="calls_and_invokes">Treating calls and invokes the same way</a>
1354 <div class="doc_text">
1356 <p>You may have noticed that the previous example was a bit oversimplified in
1357 that it did not deal with call sites generated by 'invoke' instructions. In
1358 this, and in other situations, you may find that you want to treat
1359 <tt>CallInst</tt>s and <tt>InvokeInst</tt>s the same way, even though their
1360 most-specific common base class is <tt>Instruction</tt>, which includes lots of
1361 less closely-related things. For these cases, LLVM provides a handy wrapper
1363 href="http://llvm.org/doxygen/classllvm_1_1CallSite.html"><tt>CallSite</tt></a>.
1364 It is essentially a wrapper around an <tt>Instruction</tt> pointer, with some
1365 methods that provide functionality common to <tt>CallInst</tt>s and
1366 <tt>InvokeInst</tt>s.</p>
1368 <p>This class has "value semantics": it should be passed by value, not by
1369 reference and it should not be dynamically allocated or deallocated using
1370 <tt>operator new</tt> or <tt>operator delete</tt>. It is efficiently copyable,
1371 assignable and constructable, with costs equivalents to that of a bare pointer.
1372 If you look at its definition, it has only a single pointer member.</p>
1376 <!--_______________________________________________________________________-->
1377 <div class="doc_subsubsection">
1378 <a name="iterate_chains">Iterating over def-use & use-def chains</a>
1381 <div class="doc_text">
1383 <p>Frequently, we might have an instance of the <a
1384 href="/doxygen/classllvm_1_1Value.html">Value Class</a> and we want to
1385 determine which <tt>User</tt>s use the <tt>Value</tt>. The list of all
1386 <tt>User</tt>s of a particular <tt>Value</tt> is called a <i>def-use</i> chain.
1387 For example, let's say we have a <tt>Function*</tt> named <tt>F</tt> to a
1388 particular function <tt>foo</tt>. Finding all of the instructions that
1389 <i>use</i> <tt>foo</tt> is as simple as iterating over the <i>def-use</i> chain
1392 <div class="doc_code">
1396 for (Value::use_iterator i = F->use_begin(), e = F->use_end(); i != e; ++i)
1397 if (Instruction *Inst = dyn_cast<Instruction>(*i)) {
1398 llvm::cerr << "F is used in instruction:\n";
1399 llvm::cerr << *Inst << "\n";
1404 <p>Alternately, it's common to have an instance of the <a
1405 href="/doxygen/classllvm_1_1User.html">User Class</a> and need to know what
1406 <tt>Value</tt>s are used by it. The list of all <tt>Value</tt>s used by a
1407 <tt>User</tt> is known as a <i>use-def</i> chain. Instances of class
1408 <tt>Instruction</tt> are common <tt>User</tt>s, so we might want to iterate over
1409 all of the values that a particular instruction uses (that is, the operands of
1410 the particular <tt>Instruction</tt>):</p>
1412 <div class="doc_code">
1414 Instruction* pi = ...;
1416 for (User::op_iterator i = pi->op_begin(), e = pi->op_end(); i != e; ++i) {
1424 def-use chains ("finding all users of"): Value::use_begin/use_end
1425 use-def chains ("finding all values used"): User::op_begin/op_end [op=operand]
1430 <!-- ======================================================================= -->
1431 <div class="doc_subsection">
1432 <a name="simplechanges">Making simple changes</a>
1435 <div class="doc_text">
1437 <p>There are some primitive transformation operations present in the LLVM
1438 infrastructure that are worth knowing about. When performing
1439 transformations, it's fairly common to manipulate the contents of basic
1440 blocks. This section describes some of the common methods for doing so
1441 and gives example code.</p>
1445 <!--_______________________________________________________________________-->
1446 <div class="doc_subsubsection">
1447 <a name="schanges_creating">Creating and inserting new
1448 <tt>Instruction</tt>s</a>
1451 <div class="doc_text">
1453 <p><i>Instantiating Instructions</i></p>
1455 <p>Creation of <tt>Instruction</tt>s is straight-forward: simply call the
1456 constructor for the kind of instruction to instantiate and provide the necessary
1457 parameters. For example, an <tt>AllocaInst</tt> only <i>requires</i> a
1458 (const-ptr-to) <tt>Type</tt>. Thus:</p>
1460 <div class="doc_code">
1462 AllocaInst* ai = new AllocaInst(Type::IntTy);
1466 <p>will create an <tt>AllocaInst</tt> instance that represents the allocation of
1467 one integer in the current stack frame, at runtime. Each <tt>Instruction</tt>
1468 subclass is likely to have varying default parameters which change the semantics
1469 of the instruction, so refer to the <a
1470 href="/doxygen/classllvm_1_1Instruction.html">doxygen documentation for the subclass of
1471 Instruction</a> that you're interested in instantiating.</p>
1473 <p><i>Naming values</i></p>
1475 <p>It is very useful to name the values of instructions when you're able to, as
1476 this facilitates the debugging of your transformations. If you end up looking
1477 at generated LLVM machine code, you definitely want to have logical names
1478 associated with the results of instructions! By supplying a value for the
1479 <tt>Name</tt> (default) parameter of the <tt>Instruction</tt> constructor, you
1480 associate a logical name with the result of the instruction's execution at
1481 runtime. For example, say that I'm writing a transformation that dynamically
1482 allocates space for an integer on the stack, and that integer is going to be
1483 used as some kind of index by some other code. To accomplish this, I place an
1484 <tt>AllocaInst</tt> at the first point in the first <tt>BasicBlock</tt> of some
1485 <tt>Function</tt>, and I'm intending to use it within the same
1486 <tt>Function</tt>. I might do:</p>
1488 <div class="doc_code">
1490 AllocaInst* pa = new AllocaInst(Type::IntTy, 0, "indexLoc");
1494 <p>where <tt>indexLoc</tt> is now the logical name of the instruction's
1495 execution value, which is a pointer to an integer on the runtime stack.</p>
1497 <p><i>Inserting instructions</i></p>
1499 <p>There are essentially two ways to insert an <tt>Instruction</tt>
1500 into an existing sequence of instructions that form a <tt>BasicBlock</tt>:</p>
1503 <li>Insertion into an explicit instruction list
1505 <p>Given a <tt>BasicBlock* pb</tt>, an <tt>Instruction* pi</tt> within that
1506 <tt>BasicBlock</tt>, and a newly-created instruction we wish to insert
1507 before <tt>*pi</tt>, we do the following: </p>
1509 <div class="doc_code">
1511 BasicBlock *pb = ...;
1512 Instruction *pi = ...;
1513 Instruction *newInst = new Instruction(...);
1515 pb->getInstList().insert(pi, newInst); // <i>Inserts newInst before pi in pb</i>
1519 <p>Appending to the end of a <tt>BasicBlock</tt> is so common that
1520 the <tt>Instruction</tt> class and <tt>Instruction</tt>-derived
1521 classes provide constructors which take a pointer to a
1522 <tt>BasicBlock</tt> to be appended to. For example code that
1525 <div class="doc_code">
1527 BasicBlock *pb = ...;
1528 Instruction *newInst = new Instruction(...);
1530 pb->getInstList().push_back(newInst); // <i>Appends newInst to pb</i>
1536 <div class="doc_code">
1538 BasicBlock *pb = ...;
1539 Instruction *newInst = new Instruction(..., pb);
1543 <p>which is much cleaner, especially if you are creating
1544 long instruction streams.</p></li>
1546 <li>Insertion into an implicit instruction list
1548 <p><tt>Instruction</tt> instances that are already in <tt>BasicBlock</tt>s
1549 are implicitly associated with an existing instruction list: the instruction
1550 list of the enclosing basic block. Thus, we could have accomplished the same
1551 thing as the above code without being given a <tt>BasicBlock</tt> by doing:
1554 <div class="doc_code">
1556 Instruction *pi = ...;
1557 Instruction *newInst = new Instruction(...);
1559 pi->getParent()->getInstList().insert(pi, newInst);
1563 <p>In fact, this sequence of steps occurs so frequently that the
1564 <tt>Instruction</tt> class and <tt>Instruction</tt>-derived classes provide
1565 constructors which take (as a default parameter) a pointer to an
1566 <tt>Instruction</tt> which the newly-created <tt>Instruction</tt> should
1567 precede. That is, <tt>Instruction</tt> constructors are capable of
1568 inserting the newly-created instance into the <tt>BasicBlock</tt> of a
1569 provided instruction, immediately before that instruction. Using an
1570 <tt>Instruction</tt> constructor with a <tt>insertBefore</tt> (default)
1571 parameter, the above code becomes:</p>
1573 <div class="doc_code">
1575 Instruction* pi = ...;
1576 Instruction* newInst = new Instruction(..., pi);
1580 <p>which is much cleaner, especially if you're creating a lot of
1581 instructions and adding them to <tt>BasicBlock</tt>s.</p></li>
1586 <!--_______________________________________________________________________-->
1587 <div class="doc_subsubsection">
1588 <a name="schanges_deleting">Deleting <tt>Instruction</tt>s</a>
1591 <div class="doc_text">
1593 <p>Deleting an instruction from an existing sequence of instructions that form a
1594 <a href="#BasicBlock"><tt>BasicBlock</tt></a> is very straight-forward. First,
1595 you must have a pointer to the instruction that you wish to delete. Second, you
1596 need to obtain the pointer to that instruction's basic block. You use the
1597 pointer to the basic block to get its list of instructions and then use the
1598 erase function to remove your instruction. For example:</p>
1600 <div class="doc_code">
1602 <a href="#Instruction">Instruction</a> *I = .. ;
1603 <a href="#BasicBlock">BasicBlock</a> *BB = I->getParent();
1605 BB->getInstList().erase(I);
1611 <!--_______________________________________________________________________-->
1612 <div class="doc_subsubsection">
1613 <a name="schanges_replacing">Replacing an <tt>Instruction</tt> with another
1617 <div class="doc_text">
1619 <p><i>Replacing individual instructions</i></p>
1621 <p>Including "<a href="/doxygen/BasicBlockUtils_8h-source.html">llvm/Transforms/Utils/BasicBlockUtils.h</a>"
1622 permits use of two very useful replace functions: <tt>ReplaceInstWithValue</tt>
1623 and <tt>ReplaceInstWithInst</tt>.</p>
1625 <h4><a name="schanges_deleting">Deleting <tt>Instruction</tt>s</a></h4>
1628 <li><tt>ReplaceInstWithValue</tt>
1630 <p>This function replaces all uses (within a basic block) of a given
1631 instruction with a value, and then removes the original instruction. The
1632 following example illustrates the replacement of the result of a particular
1633 <tt>AllocaInst</tt> that allocates memory for a single integer with a null
1634 pointer to an integer.</p>
1636 <div class="doc_code">
1638 AllocaInst* instToReplace = ...;
1639 BasicBlock::iterator ii(instToReplace);
1641 ReplaceInstWithValue(instToReplace->getParent()->getInstList(), ii,
1642 Constant::getNullValue(PointerType::get(Type::IntTy)));
1645 <li><tt>ReplaceInstWithInst</tt>
1647 <p>This function replaces a particular instruction with another
1648 instruction. The following example illustrates the replacement of one
1649 <tt>AllocaInst</tt> with another.</p>
1651 <div class="doc_code">
1653 AllocaInst* instToReplace = ...;
1654 BasicBlock::iterator ii(instToReplace);
1656 ReplaceInstWithInst(instToReplace->getParent()->getInstList(), ii,
1657 new AllocaInst(Type::IntTy, 0, "ptrToReplacedInt"));
1661 <p><i>Replacing multiple uses of <tt>User</tt>s and <tt>Value</tt>s</i></p>
1663 <p>You can use <tt>Value::replaceAllUsesWith</tt> and
1664 <tt>User::replaceUsesOfWith</tt> to change more than one use at a time. See the
1665 doxygen documentation for the <a href="/doxygen/classllvm_1_1Value.html">Value Class</a>
1666 and <a href="/doxygen/classllvm_1_1User.html">User Class</a>, respectively, for more
1669 <!-- Value::replaceAllUsesWith User::replaceUsesOfWith Point out:
1670 include/llvm/Transforms/Utils/ especially BasicBlockUtils.h with:
1671 ReplaceInstWithValue, ReplaceInstWithInst -->
1675 <!-- *********************************************************************** -->
1676 <div class="doc_section">
1677 <a name="advanced">Advanced Topics</a>
1679 <!-- *********************************************************************** -->
1681 <div class="doc_text">
1683 This section describes some of the advanced or obscure API's that most clients
1684 do not need to be aware of. These API's tend manage the inner workings of the
1685 LLVM system, and only need to be accessed in unusual circumstances.
1689 <!-- ======================================================================= -->
1690 <div class="doc_subsection">
1691 <a name="TypeResolve">LLVM Type Resolution</a>
1694 <div class="doc_text">
1697 The LLVM type system has a very simple goal: allow clients to compare types for
1698 structural equality with a simple pointer comparison (aka a shallow compare).
1699 This goal makes clients much simpler and faster, and is used throughout the LLVM
1704 Unfortunately achieving this goal is not a simple matter. In particular,
1705 recursive types and late resolution of opaque types makes the situation very
1706 difficult to handle. Fortunately, for the most part, our implementation makes
1707 most clients able to be completely unaware of the nasty internal details. The
1708 primary case where clients are exposed to the inner workings of it are when
1709 building a recursive type. In addition to this case, the LLVM bytecode reader,
1710 assembly parser, and linker also have to be aware of the inner workings of this
1715 For our purposes below, we need three concepts. First, an "Opaque Type" is
1716 exactly as defined in the <a href="LangRef.html#t_opaque">language
1717 reference</a>. Second an "Abstract Type" is any type which includes an
1718 opaque type as part of its type graph (for example "<tt>{ opaque, i32 }</tt>").
1719 Third, a concrete type is a type that is not an abstract type (e.g. "<tt>{ i32,
1725 <!-- ______________________________________________________________________ -->
1726 <div class="doc_subsubsection">
1727 <a name="BuildRecType">Basic Recursive Type Construction</a>
1730 <div class="doc_text">
1733 Because the most common question is "how do I build a recursive type with LLVM",
1734 we answer it now and explain it as we go. Here we include enough to cause this
1735 to be emitted to an output .ll file:
1738 <div class="doc_code">
1740 %mylist = type { %mylist*, i32 }
1745 To build this, use the following LLVM APIs:
1748 <div class="doc_code">
1750 // <i>Create the initial outer struct</i>
1751 <a href="#PATypeHolder">PATypeHolder</a> StructTy = OpaqueType::get();
1752 std::vector<const Type*> Elts;
1753 Elts.push_back(PointerType::get(StructTy));
1754 Elts.push_back(Type::IntTy);
1755 StructType *NewSTy = StructType::get(Elts);
1757 // <i>At this point, NewSTy = "{ opaque*, i32 }". Tell VMCore that</i>
1758 // <i>the struct and the opaque type are actually the same.</i>
1759 cast<OpaqueType>(StructTy.get())-><a href="#refineAbstractTypeTo">refineAbstractTypeTo</a>(NewSTy);
1761 // <i>NewSTy is potentially invalidated, but StructTy (a <a href="#PATypeHolder">PATypeHolder</a>) is</i>
1762 // <i>kept up-to-date</i>
1763 NewSTy = cast<StructType>(StructTy.get());
1765 // <i>Add a name for the type to the module symbol table (optional)</i>
1766 MyModule->addTypeName("mylist", NewSTy);
1771 This code shows the basic approach used to build recursive types: build a
1772 non-recursive type using 'opaque', then use type unification to close the cycle.
1773 The type unification step is performed by the <tt><a
1774 href="#refineAbstractTypeTo">refineAbstractTypeTo</a></tt> method, which is
1775 described next. After that, we describe the <a
1776 href="#PATypeHolder">PATypeHolder class</a>.
1781 <!-- ______________________________________________________________________ -->
1782 <div class="doc_subsubsection">
1783 <a name="refineAbstractTypeTo">The <tt>refineAbstractTypeTo</tt> method</a>
1786 <div class="doc_text">
1788 The <tt>refineAbstractTypeTo</tt> method starts the type unification process.
1789 While this method is actually a member of the DerivedType class, it is most
1790 often used on OpaqueType instances. Type unification is actually a recursive
1791 process. After unification, types can become structurally isomorphic to
1792 existing types, and all duplicates are deleted (to preserve pointer equality).
1796 In the example above, the OpaqueType object is definitely deleted.
1797 Additionally, if there is an "{ \2*, i32}" type already created in the system,
1798 the pointer and struct type created are <b>also</b> deleted. Obviously whenever
1799 a type is deleted, any "Type*" pointers in the program are invalidated. As
1800 such, it is safest to avoid having <i>any</i> "Type*" pointers to abstract types
1801 live across a call to <tt>refineAbstractTypeTo</tt> (note that non-abstract
1802 types can never move or be deleted). To deal with this, the <a
1803 href="#PATypeHolder">PATypeHolder</a> class is used to maintain a stable
1804 reference to a possibly refined type, and the <a
1805 href="#AbstractTypeUser">AbstractTypeUser</a> class is used to update more
1806 complex datastructures.
1811 <!-- ______________________________________________________________________ -->
1812 <div class="doc_subsubsection">
1813 <a name="PATypeHolder">The PATypeHolder Class</a>
1816 <div class="doc_text">
1818 PATypeHolder is a form of a "smart pointer" for Type objects. When VMCore
1819 happily goes about nuking types that become isomorphic to existing types, it
1820 automatically updates all PATypeHolder objects to point to the new type. In the
1821 example above, this allows the code to maintain a pointer to the resultant
1822 resolved recursive type, even though the Type*'s are potentially invalidated.
1826 PATypeHolder is an extremely light-weight object that uses a lazy union-find
1827 implementation to update pointers. For example the pointer from a Value to its
1828 Type is maintained by PATypeHolder objects.
1833 <!-- ______________________________________________________________________ -->
1834 <div class="doc_subsubsection">
1835 <a name="AbstractTypeUser">The AbstractTypeUser Class</a>
1838 <div class="doc_text">
1841 Some data structures need more to perform more complex updates when types get
1842 resolved. The <a href="#SymbolTable">SymbolTable</a> class, for example, needs
1843 move and potentially merge type planes in its representation when a pointer
1847 To support this, a class can derive from the AbstractTypeUser class. This class
1848 allows it to get callbacks when certain types are resolved. To register to get
1849 callbacks for a particular type, the DerivedType::{add/remove}AbstractTypeUser
1850 methods can be called on a type. Note that these methods only work for <i>
1851 abstract</i> types. Concrete types (those that do not include any opaque
1852 objects) can never be refined.
1857 <!-- ======================================================================= -->
1858 <div class="doc_subsection">
1859 <a name="SymbolTable">The <tt>SymbolTable</tt> class</a>
1862 <div class="doc_text">
1863 <p>This class provides a symbol table that the <a
1864 href="#Function"><tt>Function</tt></a> and <a href="#Module">
1865 <tt>Module</tt></a> classes use for naming definitions. The symbol table can
1866 provide a name for any <a href="#Value"><tt>Value</tt></a>.
1867 <tt>SymbolTable</tt> is an abstract data type. It hides the data it contains
1868 and provides access to it through a controlled interface.</p>
1870 <p>Note that the <tt>SymbolTable</tt> class should not be directly accessed
1871 by most clients. It should only be used when iteration over the symbol table
1872 names themselves are required, which is very special purpose. Note that not
1874 <a href="#Value">Value</a>s have names, and those without names (i.e. they have
1875 an empty name) do not exist in the symbol table.
1878 <p>To use the <tt>SymbolTable</tt> well, you need to understand the
1879 structure of the information it holds. The class contains two
1880 <tt>std::map</tt> objects. The first, <tt>pmap</tt>, is a map of
1881 <tt>Type*</tt> to maps of name (<tt>std::string</tt>) to <tt>Value*</tt>.
1882 Thus, Values are stored in two-dimensions and accessed by <tt>Type</tt> and
1885 <p>The interface of this class provides three basic types of operations:
1887 <li><em>Accessors</em>. Accessors provide read-only access to information
1888 such as finding a value for a name with the
1889 <a href="#SymbolTable_lookup">lookup</a> method.</li>
1890 <li><em>Mutators</em>. Mutators allow the user to add information to the
1891 <tt>SymbolTable</tt> with methods like
1892 <a href="#SymbolTable_insert"><tt>insert</tt></a>.</li>
1893 <li><em>Iterators</em>. Iterators allow the user to traverse the content
1894 of the symbol table in well defined ways, such as the method
1895 <a href="#SymbolTable_plane_begin"><tt>plane_begin</tt></a>.</li>
1900 <dt><tt>Value* lookup(const Type* Ty, const std::string& name) const</tt>:
1902 <dd>The <tt>lookup</tt> method searches the type plane given by the
1903 <tt>Ty</tt> parameter for a <tt>Value</tt> with the provided <tt>name</tt>.
1904 If a suitable <tt>Value</tt> is not found, null is returned.</dd>
1906 <dt><tt>bool isEmpty() const</tt>:</dt>
1907 <dd>This function returns true if both the value and types maps are
1913 <dt><tt>void insert(Value *Val)</tt>:</dt>
1914 <dd>This method adds the provided value to the symbol table. The Value must
1915 have both a name and a type which are extracted and used to place the value
1916 in the correct type plane under the value's name.</dd>
1918 <dt><tt>void insert(const std::string& Name, Value *Val)</tt>:</dt>
1919 <dd> Inserts a constant or type into the symbol table with the specified
1920 name. There can be a many to one mapping between names and constants
1923 <dt><tt>void remove(Value* Val)</tt>:</dt>
1924 <dd> This method removes a named value from the symbol table. The
1925 type and name of the Value are extracted from \p N and used to
1926 lookup the Value in the correct type plane. If the Value is
1927 not in the symbol table, this method silently ignores the
1930 <dt><tt>Value* remove(const std::string& Name, Value *Val)</tt>:</dt>
1931 <dd> Remove a constant or type with the specified name from the
1934 <dt><tt>Value *remove(const value_iterator& It)</tt>:</dt>
1935 <dd> Removes a specific value from the symbol table.
1936 Returns the removed value.</dd>
1938 <dt><tt>bool strip()</tt>:</dt>
1939 <dd> This method will strip the symbol table of its names leaving
1940 the type and values. </dd>
1942 <dt><tt>void clear()</tt>:</dt>
1943 <dd>Empty the symbol table completely.</dd>
1947 <p>The following functions describe three types of iterators you can obtain
1948 the beginning or end of the sequence for both const and non-const. It is
1949 important to keep track of the different kinds of iterators. There are
1950 three idioms worth pointing out:</p>
1953 <tr><th>Units</th><th>Iterator</th><th>Idiom</th></tr>
1955 <td align="left">Planes Of name/Value maps</td><td>PI</td>
1956 <td align="left"><pre><tt>
1957 for (SymbolTable::plane_const_iterator PI = ST.plane_begin(),
1958 PE = ST.plane_end(); PI != PE; ++PI ) {
1959 PI->first // <i>This is the Type* of the plane</i>
1960 PI->second // <i>This is the SymbolTable::ValueMap of name/Value pairs</i>
1965 <td align="left">name/Value pairs in a plane</td><td>VI</td>
1966 <td align="left"><pre><tt>
1967 for (SymbolTable::value_const_iterator VI = ST.value_begin(SomeType),
1968 VE = ST.value_end(SomeType); VI != VE; ++VI ) {
1969 VI->first // <i>This is the name of the Value</i>
1970 VI->second // <i>This is the Value* value associated with the name</i>
1976 <p>Using the recommended iterator names and idioms will help you avoid
1977 making mistakes. Of particular note, make sure that whenever you use
1978 value_begin(SomeType) that you always compare the resulting iterator
1979 with value_end(SomeType) not value_end(SomeOtherType) or else you
1980 will loop infinitely.</p>
1984 <dt><tt>plane_iterator plane_begin()</tt>:</dt>
1985 <dd>Get an iterator that starts at the beginning of the type planes.
1986 The iterator will iterate over the Type/ValueMap pairs in the
1989 <dt><tt>plane_const_iterator plane_begin() const</tt>:</dt>
1990 <dd>Get a const_iterator that starts at the beginning of the type
1991 planes. The iterator will iterate over the Type/ValueMap pairs
1992 in the type planes. </dd>
1994 <dt><tt>plane_iterator plane_end()</tt>:</dt>
1995 <dd>Get an iterator at the end of the type planes. This serves as
1996 the marker for end of iteration over the type planes.</dd>
1998 <dt><tt>plane_const_iterator plane_end() const</tt>:</dt>
1999 <dd>Get a const_iterator at the end of the type planes. This serves as
2000 the marker for end of iteration over the type planes.</dd>
2002 <dt><tt>value_iterator value_begin(const Type *Typ)</tt>:</dt>
2003 <dd>Get an iterator that starts at the beginning of a type plane.
2004 The iterator will iterate over the name/value pairs in the type plane.
2005 Note: The type plane must already exist before using this.</dd>
2007 <dt><tt>value_const_iterator value_begin(const Type *Typ) const</tt>:</dt>
2008 <dd>Get a const_iterator that starts at the beginning of a type plane.
2009 The iterator will iterate over the name/value pairs in the type plane.
2010 Note: The type plane must already exist before using this.</dd>
2012 <dt><tt>value_iterator value_end(const Type *Typ)</tt>:</dt>
2013 <dd>Get an iterator to the end of a type plane. This serves as the marker
2014 for end of iteration of the type plane.
2015 Note: The type plane must already exist before using this.</dd>
2017 <dt><tt>value_const_iterator value_end(const Type *Typ) const</tt>:</dt>
2018 <dd>Get a const_iterator to the end of a type plane. This serves as the
2019 marker for end of iteration of the type plane.
2020 Note: the type plane must already exist before using this.</dd>
2022 <dt><tt>plane_const_iterator find(const Type* Typ ) const</tt>:</dt>
2023 <dd>This method returns a plane_const_iterator for iteration over
2024 the type planes starting at a specific plane, given by \p Ty.</dd>
2026 <dt><tt>plane_iterator find( const Type* Typ </tt>:</dt>
2027 <dd>This method returns a plane_iterator for iteration over the
2028 type planes starting at a specific plane, given by \p Ty.</dd>
2035 <!-- *********************************************************************** -->
2036 <div class="doc_section">
2037 <a name="coreclasses">The Core LLVM Class Hierarchy Reference </a>
2039 <!-- *********************************************************************** -->
2041 <div class="doc_text">
2042 <p><tt>#include "<a href="/doxygen/Type_8h-source.html">llvm/Type.h</a>"</tt>
2043 <br>doxygen info: <a href="/doxygen/classllvm_1_1Type.html">Type Class</a></p>
2045 <p>The Core LLVM classes are the primary means of representing the program
2046 being inspected or transformed. The core LLVM classes are defined in
2047 header files in the <tt>include/llvm/</tt> directory, and implemented in
2048 the <tt>lib/VMCore</tt> directory.</p>
2052 <!-- ======================================================================= -->
2053 <div class="doc_subsection">
2054 <a name="Type">The <tt>Type</tt> class and Derived Types</a>
2057 <div class="doc_text">
2059 <p><tt>Type</tt> is a superclass of all type classes. Every <tt>Value</tt> has
2060 a <tt>Type</tt>. <tt>Type</tt> cannot be instantiated directly but only
2061 through its subclasses. Certain primitive types (<tt>VoidType</tt>,
2062 <tt>LabelType</tt>, <tt>FloatType</tt> and <tt>DoubleType</tt>) have hidden
2063 subclasses. They are hidden because they offer no useful functionality beyond
2064 what the <tt>Type</tt> class offers except to distinguish themselves from
2065 other subclasses of <tt>Type</tt>.</p>
2066 <p>All other types are subclasses of <tt>DerivedType</tt>. Types can be
2067 named, but this is not a requirement. There exists exactly
2068 one instance of a given shape at any one time. This allows type equality to
2069 be performed with address equality of the Type Instance. That is, given two
2070 <tt>Type*</tt> values, the types are identical if the pointers are identical.
2074 <!-- _______________________________________________________________________ -->
2075 <div class="doc_subsubsection">
2076 <a name="m_Value">Important Public Methods</a>
2079 <div class="doc_text">
2082 <li><tt>bool isInteger() const</tt>: Returns true for any integer type.</li>
2084 <li><tt>bool isFloatingPoint()</tt>: Return true if this is one of the two
2085 floating point types.</li>
2087 <li><tt>bool isAbstract()</tt>: Return true if the type is abstract (contains
2088 an OpaqueType anywhere in its definition).</li>
2090 <li><tt>bool isSized()</tt>: Return true if the type has known size. Things
2091 that don't have a size are abstract types, labels and void.</li>
2096 <!-- _______________________________________________________________________ -->
2097 <div class="doc_subsubsection">
2098 <a name="m_Value">Important Derived Types</a>
2100 <div class="doc_text">
2102 <dt><tt>IntegerType</tt></dt>
2103 <dd>Subclass of DerivedType that represents integer types of any bit width.
2104 Any bit width between <tt>IntegerType::MIN_INT_BITS</tt> (1) and
2105 <tt>IntegerType::MAX_INT_BITS</tt> (~8 million) can be represented.
2107 <li><tt>static const IntegerType* get(unsigned NumBits)</tt>: get an integer
2108 type of a specific bit width.</li>
2109 <li><tt>unsigned getBitWidth() const</tt>: Get the bit width of an integer
2113 <dt><tt>SequentialType</tt></dt>
2114 <dd>This is subclassed by ArrayType and PointerType
2116 <li><tt>const Type * getElementType() const</tt>: Returns the type of each
2117 of the elements in the sequential type. </li>
2120 <dt><tt>ArrayType</tt></dt>
2121 <dd>This is a subclass of SequentialType and defines the interface for array
2124 <li><tt>unsigned getNumElements() const</tt>: Returns the number of
2125 elements in the array. </li>
2128 <dt><tt>PointerType</tt></dt>
2129 <dd>Subclass of SequentialType for pointer types.</dd>
2130 <dt><tt>PackedType</tt></dt>
2131 <dd>Subclass of SequentialType for packed (vector) types. A
2132 packed type is similar to an ArrayType but is distinguished because it is
2133 a first class type wherease ArrayType is not. Packed types are used for
2134 vector operations and are usually small vectors of of an integer or floating
2136 <dt><tt>StructType</tt></dt>
2137 <dd>Subclass of DerivedTypes for struct types.</dd>
2138 <dt><tt>FunctionType</tt></dt>
2139 <dd>Subclass of DerivedTypes for function types.
2141 <li><tt>bool isVarArg() const</tt>: Returns true if its a vararg
2143 <li><tt> const Type * getReturnType() const</tt>: Returns the
2144 return type of the function.</li>
2145 <li><tt>const Type * getParamType (unsigned i)</tt>: Returns
2146 the type of the ith parameter.</li>
2147 <li><tt> const unsigned getNumParams() const</tt>: Returns the
2148 number of formal parameters.</li>
2151 <dt><tt>OpaqueType</tt></dt>
2152 <dd>Sublcass of DerivedType for abstract types. This class
2153 defines no content and is used as a placeholder for some other type. Note
2154 that OpaqueType is used (temporarily) during type resolution for forward
2155 references of types. Once the referenced type is resolved, the OpaqueType
2156 is replaced with the actual type. OpaqueType can also be used for data
2157 abstraction. At link time opaque types can be resolved to actual types
2158 of the same name.</dd>
2162 <!-- ======================================================================= -->
2163 <div class="doc_subsection">
2164 <a name="Value">The <tt>Value</tt> class</a>
2169 <p><tt>#include "<a href="/doxygen/Value_8h-source.html">llvm/Value.h</a>"</tt>
2171 doxygen info: <a href="/doxygen/classllvm_1_1Value.html">Value Class</a></p>
2173 <p>The <tt>Value</tt> class is the most important class in the LLVM Source
2174 base. It represents a typed value that may be used (among other things) as an
2175 operand to an instruction. There are many different types of <tt>Value</tt>s,
2176 such as <a href="#Constant"><tt>Constant</tt></a>s,<a
2177 href="#Argument"><tt>Argument</tt></a>s. Even <a
2178 href="#Instruction"><tt>Instruction</tt></a>s and <a
2179 href="#Function"><tt>Function</tt></a>s are <tt>Value</tt>s.</p>
2181 <p>A particular <tt>Value</tt> may be used many times in the LLVM representation
2182 for a program. For example, an incoming argument to a function (represented
2183 with an instance of the <a href="#Argument">Argument</a> class) is "used" by
2184 every instruction in the function that references the argument. To keep track
2185 of this relationship, the <tt>Value</tt> class keeps a list of all of the <a
2186 href="#User"><tt>User</tt></a>s that is using it (the <a
2187 href="#User"><tt>User</tt></a> class is a base class for all nodes in the LLVM
2188 graph that can refer to <tt>Value</tt>s). This use list is how LLVM represents
2189 def-use information in the program, and is accessible through the <tt>use_</tt>*
2190 methods, shown below.</p>
2192 <p>Because LLVM is a typed representation, every LLVM <tt>Value</tt> is typed,
2193 and this <a href="#Type">Type</a> is available through the <tt>getType()</tt>
2194 method. In addition, all LLVM values can be named. The "name" of the
2195 <tt>Value</tt> is a symbolic string printed in the LLVM code:</p>
2197 <div class="doc_code">
2199 %<b>foo</b> = add i32 1, 2
2203 <p><a name="#nameWarning">The name of this instruction is "foo".</a> <b>NOTE</b>
2204 that the name of any value may be missing (an empty string), so names should
2205 <b>ONLY</b> be used for debugging (making the source code easier to read,
2206 debugging printouts), they should not be used to keep track of values or map
2207 between them. For this purpose, use a <tt>std::map</tt> of pointers to the
2208 <tt>Value</tt> itself instead.</p>
2210 <p>One important aspect of LLVM is that there is no distinction between an SSA
2211 variable and the operation that produces it. Because of this, any reference to
2212 the value produced by an instruction (or the value available as an incoming
2213 argument, for example) is represented as a direct pointer to the instance of
2215 represents this value. Although this may take some getting used to, it
2216 simplifies the representation and makes it easier to manipulate.</p>
2220 <!-- _______________________________________________________________________ -->
2221 <div class="doc_subsubsection">
2222 <a name="m_Value">Important Public Members of the <tt>Value</tt> class</a>
2225 <div class="doc_text">
2228 <li><tt>Value::use_iterator</tt> - Typedef for iterator over the
2230 <tt>Value::use_const_iterator</tt> - Typedef for const_iterator over
2232 <tt>unsigned use_size()</tt> - Returns the number of users of the
2234 <tt>bool use_empty()</tt> - Returns true if there are no users.<br>
2235 <tt>use_iterator use_begin()</tt> - Get an iterator to the start of
2237 <tt>use_iterator use_end()</tt> - Get an iterator to the end of the
2239 <tt><a href="#User">User</a> *use_back()</tt> - Returns the last
2240 element in the list.
2241 <p> These methods are the interface to access the def-use
2242 information in LLVM. As with all other iterators in LLVM, the naming
2243 conventions follow the conventions defined by the <a href="#stl">STL</a>.</p>
2245 <li><tt><a href="#Type">Type</a> *getType() const</tt>
2246 <p>This method returns the Type of the Value.</p>
2248 <li><tt>bool hasName() const</tt><br>
2249 <tt>std::string getName() const</tt><br>
2250 <tt>void setName(const std::string &Name)</tt>
2251 <p> This family of methods is used to access and assign a name to a <tt>Value</tt>,
2252 be aware of the <a href="#nameWarning">precaution above</a>.</p>
2254 <li><tt>void replaceAllUsesWith(Value *V)</tt>
2256 <p>This method traverses the use list of a <tt>Value</tt> changing all <a
2257 href="#User"><tt>User</tt>s</a> of the current value to refer to
2258 "<tt>V</tt>" instead. For example, if you detect that an instruction always
2259 produces a constant value (for example through constant folding), you can
2260 replace all uses of the instruction with the constant like this:</p>
2262 <div class="doc_code">
2264 Inst->replaceAllUsesWith(ConstVal);
2272 <!-- ======================================================================= -->
2273 <div class="doc_subsection">
2274 <a name="User">The <tt>User</tt> class</a>
2277 <div class="doc_text">
2280 <tt>#include "<a href="/doxygen/User_8h-source.html">llvm/User.h</a>"</tt><br>
2281 doxygen info: <a href="/doxygen/classllvm_1_1User.html">User Class</a><br>
2282 Superclass: <a href="#Value"><tt>Value</tt></a></p>
2284 <p>The <tt>User</tt> class is the common base class of all LLVM nodes that may
2285 refer to <a href="#Value"><tt>Value</tt></a>s. It exposes a list of "Operands"
2286 that are all of the <a href="#Value"><tt>Value</tt></a>s that the User is
2287 referring to. The <tt>User</tt> class itself is a subclass of
2290 <p>The operands of a <tt>User</tt> point directly to the LLVM <a
2291 href="#Value"><tt>Value</tt></a> that it refers to. Because LLVM uses Static
2292 Single Assignment (SSA) form, there can only be one definition referred to,
2293 allowing this direct connection. This connection provides the use-def
2294 information in LLVM.</p>
2298 <!-- _______________________________________________________________________ -->
2299 <div class="doc_subsubsection">
2300 <a name="m_User">Important Public Members of the <tt>User</tt> class</a>
2303 <div class="doc_text">
2305 <p>The <tt>User</tt> class exposes the operand list in two ways: through
2306 an index access interface and through an iterator based interface.</p>
2309 <li><tt>Value *getOperand(unsigned i)</tt><br>
2310 <tt>unsigned getNumOperands()</tt>
2311 <p> These two methods expose the operands of the <tt>User</tt> in a
2312 convenient form for direct access.</p></li>
2314 <li><tt>User::op_iterator</tt> - Typedef for iterator over the operand
2316 <tt>op_iterator op_begin()</tt> - Get an iterator to the start of
2317 the operand list.<br>
2318 <tt>op_iterator op_end()</tt> - Get an iterator to the end of the
2320 <p> Together, these methods make up the iterator based interface to
2321 the operands of a <tt>User</tt>.</p></li>
2326 <!-- ======================================================================= -->
2327 <div class="doc_subsection">
2328 <a name="Instruction">The <tt>Instruction</tt> class</a>
2331 <div class="doc_text">
2333 <p><tt>#include "</tt><tt><a
2334 href="/doxygen/Instruction_8h-source.html">llvm/Instruction.h</a>"</tt><br>
2335 doxygen info: <a href="/doxygen/classllvm_1_1Instruction.html">Instruction Class</a><br>
2336 Superclasses: <a href="#User"><tt>User</tt></a>, <a
2337 href="#Value"><tt>Value</tt></a></p>
2339 <p>The <tt>Instruction</tt> class is the common base class for all LLVM
2340 instructions. It provides only a few methods, but is a very commonly used
2341 class. The primary data tracked by the <tt>Instruction</tt> class itself is the
2342 opcode (instruction type) and the parent <a
2343 href="#BasicBlock"><tt>BasicBlock</tt></a> the <tt>Instruction</tt> is embedded
2344 into. To represent a specific type of instruction, one of many subclasses of
2345 <tt>Instruction</tt> are used.</p>
2347 <p> Because the <tt>Instruction</tt> class subclasses the <a
2348 href="#User"><tt>User</tt></a> class, its operands can be accessed in the same
2349 way as for other <a href="#User"><tt>User</tt></a>s (with the
2350 <tt>getOperand()</tt>/<tt>getNumOperands()</tt> and
2351 <tt>op_begin()</tt>/<tt>op_end()</tt> methods).</p> <p> An important file for
2352 the <tt>Instruction</tt> class is the <tt>llvm/Instruction.def</tt> file. This
2353 file contains some meta-data about the various different types of instructions
2354 in LLVM. It describes the enum values that are used as opcodes (for example
2355 <tt>Instruction::Add</tt> and <tt>Instruction::ICmp</tt>), as well as the
2356 concrete sub-classes of <tt>Instruction</tt> that implement the instruction (for
2357 example <tt><a href="#BinaryOperator">BinaryOperator</a></tt> and <tt><a
2358 href="#CmpInst">CmpInst</a></tt>). Unfortunately, the use of macros in
2359 this file confuses doxygen, so these enum values don't show up correctly in the
2360 <a href="/doxygen/classllvm_1_1Instruction.html">doxygen output</a>.</p>
2364 <!-- _______________________________________________________________________ -->
2365 <div class="doc_subsubsection">
2366 <a name="s_Instruction">Important Subclasses of the <tt>Instruction</tt>
2369 <div class="doc_text">
2371 <li><tt><a name="BinaryOperator">BinaryOperator</a></tt>
2372 <p>This subclasses represents all two operand instructions whose operands
2373 must be the same type, except for the comparison instructions.</p></li>
2374 <li><tt><a name="CastInst">CastInst</a></tt>
2375 <p>This subclass is the parent of the 12 casting instructions. It provides
2376 common operations on cast instructions.</p>
2377 <li><tt><a name="CmpInst">CmpInst</a></tt>
2378 <p>This subclass respresents the two comparison instructions,
2379 <a href="LangRef.html#i_icmp">ICmpInst</a> (integer opreands), and
2380 <a href="LangRef.html#i_fcmp">FCmpInst</a> (floating point operands).</p>
2381 <li><tt><a name="TerminatorInst">TerminatorInst</a></tt>
2382 <p>This subclass is the parent of all terminator instructions (those which
2383 can terminate a block).</p>
2387 <!-- _______________________________________________________________________ -->
2388 <div class="doc_subsubsection">
2389 <a name="m_Instruction">Important Public Members of the <tt>Instruction</tt>
2393 <div class="doc_text">
2396 <li><tt><a href="#BasicBlock">BasicBlock</a> *getParent()</tt>
2397 <p>Returns the <a href="#BasicBlock"><tt>BasicBlock</tt></a> that
2398 this <tt>Instruction</tt> is embedded into.</p></li>
2399 <li><tt>bool mayWriteToMemory()</tt>
2400 <p>Returns true if the instruction writes to memory, i.e. it is a
2401 <tt>call</tt>,<tt>free</tt>,<tt>invoke</tt>, or <tt>store</tt>.</p></li>
2402 <li><tt>unsigned getOpcode()</tt>
2403 <p>Returns the opcode for the <tt>Instruction</tt>.</p></li>
2404 <li><tt><a href="#Instruction">Instruction</a> *clone() const</tt>
2405 <p>Returns another instance of the specified instruction, identical
2406 in all ways to the original except that the instruction has no parent
2407 (ie it's not embedded into a <a href="#BasicBlock"><tt>BasicBlock</tt></a>),
2408 and it has no name</p></li>
2413 <!-- ======================================================================= -->
2414 <div class="doc_subsection">
2415 <a name="BasicBlock">The <tt>BasicBlock</tt> class</a>
2418 <div class="doc_text">
2421 href="/doxygen/BasicBlock_8h-source.html">llvm/BasicBlock.h</a>"</tt><br>
2422 doxygen info: <a href="/doxygen/structllvm_1_1BasicBlock.html">BasicBlock
2424 Superclass: <a href="#Value"><tt>Value</tt></a></p>
2426 <p>This class represents a single entry multiple exit section of the code,
2427 commonly known as a basic block by the compiler community. The
2428 <tt>BasicBlock</tt> class maintains a list of <a
2429 href="#Instruction"><tt>Instruction</tt></a>s, which form the body of the block.
2430 Matching the language definition, the last element of this list of instructions
2431 is always a terminator instruction (a subclass of the <a
2432 href="#TerminatorInst"><tt>TerminatorInst</tt></a> class).</p>
2434 <p>In addition to tracking the list of instructions that make up the block, the
2435 <tt>BasicBlock</tt> class also keeps track of the <a
2436 href="#Function"><tt>Function</tt></a> that it is embedded into.</p>
2438 <p>Note that <tt>BasicBlock</tt>s themselves are <a
2439 href="#Value"><tt>Value</tt></a>s, because they are referenced by instructions
2440 like branches and can go in the switch tables. <tt>BasicBlock</tt>s have type
2445 <!-- _______________________________________________________________________ -->
2446 <div class="doc_subsubsection">
2447 <a name="m_BasicBlock">Important Public Members of the <tt>BasicBlock</tt>
2451 <div class="doc_text">
2455 <li><tt>BasicBlock(const std::string &Name = "", </tt><tt><a
2456 href="#Function">Function</a> *Parent = 0)</tt>
2458 <p>The <tt>BasicBlock</tt> constructor is used to create new basic blocks for
2459 insertion into a function. The constructor optionally takes a name for the new
2460 block, and a <a href="#Function"><tt>Function</tt></a> to insert it into. If
2461 the <tt>Parent</tt> parameter is specified, the new <tt>BasicBlock</tt> is
2462 automatically inserted at the end of the specified <a
2463 href="#Function"><tt>Function</tt></a>, if not specified, the BasicBlock must be
2464 manually inserted into the <a href="#Function"><tt>Function</tt></a>.</p></li>
2466 <li><tt>BasicBlock::iterator</tt> - Typedef for instruction list iterator<br>
2467 <tt>BasicBlock::const_iterator</tt> - Typedef for const_iterator.<br>
2468 <tt>begin()</tt>, <tt>end()</tt>, <tt>front()</tt>, <tt>back()</tt>,
2469 <tt>size()</tt>, <tt>empty()</tt>
2470 STL-style functions for accessing the instruction list.
2472 <p>These methods and typedefs are forwarding functions that have the same
2473 semantics as the standard library methods of the same names. These methods
2474 expose the underlying instruction list of a basic block in a way that is easy to
2475 manipulate. To get the full complement of container operations (including
2476 operations to update the list), you must use the <tt>getInstList()</tt>
2479 <li><tt>BasicBlock::InstListType &getInstList()</tt>
2481 <p>This method is used to get access to the underlying container that actually
2482 holds the Instructions. This method must be used when there isn't a forwarding
2483 function in the <tt>BasicBlock</tt> class for the operation that you would like
2484 to perform. Because there are no forwarding functions for "updating"
2485 operations, you need to use this if you want to update the contents of a
2486 <tt>BasicBlock</tt>.</p></li>
2488 <li><tt><a href="#Function">Function</a> *getParent()</tt>
2490 <p> Returns a pointer to <a href="#Function"><tt>Function</tt></a> the block is
2491 embedded into, or a null pointer if it is homeless.</p></li>
2493 <li><tt><a href="#TerminatorInst">TerminatorInst</a> *getTerminator()</tt>
2495 <p> Returns a pointer to the terminator instruction that appears at the end of
2496 the <tt>BasicBlock</tt>. If there is no terminator instruction, or if the last
2497 instruction in the block is not a terminator, then a null pointer is
2504 <!-- ======================================================================= -->
2505 <div class="doc_subsection">
2506 <a name="GlobalValue">The <tt>GlobalValue</tt> class</a>
2509 <div class="doc_text">
2512 href="/doxygen/GlobalValue_8h-source.html">llvm/GlobalValue.h</a>"</tt><br>
2513 doxygen info: <a href="/doxygen/classllvm_1_1GlobalValue.html">GlobalValue
2515 Superclasses: <a href="#Constant"><tt>Constant</tt></a>,
2516 <a href="#User"><tt>User</tt></a>, <a href="#Value"><tt>Value</tt></a></p>
2518 <p>Global values (<a href="#GlobalVariable"><tt>GlobalVariable</tt></a>s or <a
2519 href="#Function"><tt>Function</tt></a>s) are the only LLVM values that are
2520 visible in the bodies of all <a href="#Function"><tt>Function</tt></a>s.
2521 Because they are visible at global scope, they are also subject to linking with
2522 other globals defined in different translation units. To control the linking
2523 process, <tt>GlobalValue</tt>s know their linkage rules. Specifically,
2524 <tt>GlobalValue</tt>s know whether they have internal or external linkage, as
2525 defined by the <tt>LinkageTypes</tt> enumeration.</p>
2527 <p>If a <tt>GlobalValue</tt> has internal linkage (equivalent to being
2528 <tt>static</tt> in C), it is not visible to code outside the current translation
2529 unit, and does not participate in linking. If it has external linkage, it is
2530 visible to external code, and does participate in linking. In addition to
2531 linkage information, <tt>GlobalValue</tt>s keep track of which <a
2532 href="#Module"><tt>Module</tt></a> they are currently part of.</p>
2534 <p>Because <tt>GlobalValue</tt>s are memory objects, they are always referred to
2535 by their <b>address</b>. As such, the <a href="#Type"><tt>Type</tt></a> of a
2536 global is always a pointer to its contents. It is important to remember this
2537 when using the <tt>GetElementPtrInst</tt> instruction because this pointer must
2538 be dereferenced first. For example, if you have a <tt>GlobalVariable</tt> (a
2539 subclass of <tt>GlobalValue)</tt> that is an array of 24 ints, type <tt>[24 x
2540 i32]</tt>, then the <tt>GlobalVariable</tt> is a pointer to that array. Although
2541 the address of the first element of this array and the value of the
2542 <tt>GlobalVariable</tt> are the same, they have different types. The
2543 <tt>GlobalVariable</tt>'s type is <tt>[24 x i32]</tt>. The first element's type
2544 is <tt>i32.</tt> Because of this, accessing a global value requires you to
2545 dereference the pointer with <tt>GetElementPtrInst</tt> first, then its elements
2546 can be accessed. This is explained in the <a href="LangRef.html#globalvars">LLVM
2547 Language Reference Manual</a>.</p>
2551 <!-- _______________________________________________________________________ -->
2552 <div class="doc_subsubsection">
2553 <a name="m_GlobalValue">Important Public Members of the <tt>GlobalValue</tt>
2557 <div class="doc_text">
2560 <li><tt>bool hasInternalLinkage() const</tt><br>
2561 <tt>bool hasExternalLinkage() const</tt><br>
2562 <tt>void setInternalLinkage(bool HasInternalLinkage)</tt>
2563 <p> These methods manipulate the linkage characteristics of the <tt>GlobalValue</tt>.</p>
2566 <li><tt><a href="#Module">Module</a> *getParent()</tt>
2567 <p> This returns the <a href="#Module"><tt>Module</tt></a> that the
2568 GlobalValue is currently embedded into.</p></li>
2573 <!-- ======================================================================= -->
2574 <div class="doc_subsection">
2575 <a name="Function">The <tt>Function</tt> class</a>
2578 <div class="doc_text">
2581 href="/doxygen/Function_8h-source.html">llvm/Function.h</a>"</tt><br> doxygen
2582 info: <a href="/doxygen/classllvm_1_1Function.html">Function Class</a><br>
2583 Superclasses: <a href="#GlobalValue"><tt>GlobalValue</tt></a>,
2584 <a href="#Constant"><tt>Constant</tt></a>,
2585 <a href="#User"><tt>User</tt></a>,
2586 <a href="#Value"><tt>Value</tt></a></p>
2588 <p>The <tt>Function</tt> class represents a single procedure in LLVM. It is
2589 actually one of the more complex classes in the LLVM heirarchy because it must
2590 keep track of a large amount of data. The <tt>Function</tt> class keeps track
2591 of a list of <a href="#BasicBlock"><tt>BasicBlock</tt></a>s, a list of formal
2592 <a href="#Argument"><tt>Argument</tt></a>s, and a
2593 <a href="#SymbolTable"><tt>SymbolTable</tt></a>.</p>
2595 <p>The list of <a href="#BasicBlock"><tt>BasicBlock</tt></a>s is the most
2596 commonly used part of <tt>Function</tt> objects. The list imposes an implicit
2597 ordering of the blocks in the function, which indicate how the code will be
2598 layed out by the backend. Additionally, the first <a
2599 href="#BasicBlock"><tt>BasicBlock</tt></a> is the implicit entry node for the
2600 <tt>Function</tt>. It is not legal in LLVM to explicitly branch to this initial
2601 block. There are no implicit exit nodes, and in fact there may be multiple exit
2602 nodes from a single <tt>Function</tt>. If the <a
2603 href="#BasicBlock"><tt>BasicBlock</tt></a> list is empty, this indicates that
2604 the <tt>Function</tt> is actually a function declaration: the actual body of the
2605 function hasn't been linked in yet.</p>
2607 <p>In addition to a list of <a href="#BasicBlock"><tt>BasicBlock</tt></a>s, the
2608 <tt>Function</tt> class also keeps track of the list of formal <a
2609 href="#Argument"><tt>Argument</tt></a>s that the function receives. This
2610 container manages the lifetime of the <a href="#Argument"><tt>Argument</tt></a>
2611 nodes, just like the <a href="#BasicBlock"><tt>BasicBlock</tt></a> list does for
2612 the <a href="#BasicBlock"><tt>BasicBlock</tt></a>s.</p>
2614 <p>The <a href="#SymbolTable"><tt>SymbolTable</tt></a> is a very rarely used
2615 LLVM feature that is only used when you have to look up a value by name. Aside
2616 from that, the <a href="#SymbolTable"><tt>SymbolTable</tt></a> is used
2617 internally to make sure that there are not conflicts between the names of <a
2618 href="#Instruction"><tt>Instruction</tt></a>s, <a
2619 href="#BasicBlock"><tt>BasicBlock</tt></a>s, or <a
2620 href="#Argument"><tt>Argument</tt></a>s in the function body.</p>
2622 <p>Note that <tt>Function</tt> is a <a href="#GlobalValue">GlobalValue</a>
2623 and therefore also a <a href="#Constant">Constant</a>. The value of the function
2624 is its address (after linking) which is guaranteed to be constant.</p>
2627 <!-- _______________________________________________________________________ -->
2628 <div class="doc_subsubsection">
2629 <a name="m_Function">Important Public Members of the <tt>Function</tt>
2633 <div class="doc_text">
2636 <li><tt>Function(const </tt><tt><a href="#FunctionType">FunctionType</a>
2637 *Ty, LinkageTypes Linkage, const std::string &N = "", Module* Parent = 0)</tt>
2639 <p>Constructor used when you need to create new <tt>Function</tt>s to add
2640 the the program. The constructor must specify the type of the function to
2641 create and what type of linkage the function should have. The <a
2642 href="#FunctionType"><tt>FunctionType</tt></a> argument
2643 specifies the formal arguments and return value for the function. The same
2644 <a href="#FunctionTypel"><tt>FunctionType</tt></a> value can be used to
2645 create multiple functions. The <tt>Parent</tt> argument specifies the Module
2646 in which the function is defined. If this argument is provided, the function
2647 will automatically be inserted into that module's list of
2650 <li><tt>bool isExternal()</tt>
2652 <p>Return whether or not the <tt>Function</tt> has a body defined. If the
2653 function is "external", it does not have a body, and thus must be resolved
2654 by linking with a function defined in a different translation unit.</p></li>
2656 <li><tt>Function::iterator</tt> - Typedef for basic block list iterator<br>
2657 <tt>Function::const_iterator</tt> - Typedef for const_iterator.<br>
2659 <tt>begin()</tt>, <tt>end()</tt>
2660 <tt>size()</tt>, <tt>empty()</tt>
2662 <p>These are forwarding methods that make it easy to access the contents of
2663 a <tt>Function</tt> object's <a href="#BasicBlock"><tt>BasicBlock</tt></a>
2666 <li><tt>Function::BasicBlockListType &getBasicBlockList()</tt>
2668 <p>Returns the list of <a href="#BasicBlock"><tt>BasicBlock</tt></a>s. This
2669 is necessary to use when you need to update the list or perform a complex
2670 action that doesn't have a forwarding method.</p></li>
2672 <li><tt>Function::arg_iterator</tt> - Typedef for the argument list
2674 <tt>Function::const_arg_iterator</tt> - Typedef for const_iterator.<br>
2676 <tt>arg_begin()</tt>, <tt>arg_end()</tt>
2677 <tt>arg_size()</tt>, <tt>arg_empty()</tt>
2679 <p>These are forwarding methods that make it easy to access the contents of
2680 a <tt>Function</tt> object's <a href="#Argument"><tt>Argument</tt></a>
2683 <li><tt>Function::ArgumentListType &getArgumentList()</tt>
2685 <p>Returns the list of <a href="#Argument"><tt>Argument</tt></a>s. This is
2686 necessary to use when you need to update the list or perform a complex
2687 action that doesn't have a forwarding method.</p></li>
2689 <li><tt><a href="#BasicBlock">BasicBlock</a> &getEntryBlock()</tt>
2691 <p>Returns the entry <a href="#BasicBlock"><tt>BasicBlock</tt></a> for the
2692 function. Because the entry block for the function is always the first
2693 block, this returns the first block of the <tt>Function</tt>.</p></li>
2695 <li><tt><a href="#Type">Type</a> *getReturnType()</tt><br>
2696 <tt><a href="#FunctionType">FunctionType</a> *getFunctionType()</tt>
2698 <p>This traverses the <a href="#Type"><tt>Type</tt></a> of the
2699 <tt>Function</tt> and returns the return type of the function, or the <a
2700 href="#FunctionType"><tt>FunctionType</tt></a> of the actual
2703 <li><tt><a href="#SymbolTable">SymbolTable</a> *getSymbolTable()</tt>
2705 <p> Return a pointer to the <a href="#SymbolTable"><tt>SymbolTable</tt></a>
2706 for this <tt>Function</tt>.</p></li>
2711 <!-- ======================================================================= -->
2712 <div class="doc_subsection">
2713 <a name="GlobalVariable">The <tt>GlobalVariable</tt> class</a>
2716 <div class="doc_text">
2719 href="/doxygen/GlobalVariable_8h-source.html">llvm/GlobalVariable.h</a>"</tt>
2721 doxygen info: <a href="/doxygen/classllvm_1_1GlobalVariable.html">GlobalVariable
2723 Superclasses: <a href="#GlobalValue"><tt>GlobalValue</tt></a>,
2724 <a href="#Constant"><tt>Constant</tt></a>,
2725 <a href="#User"><tt>User</tt></a>,
2726 <a href="#Value"><tt>Value</tt></a></p>
2728 <p>Global variables are represented with the (suprise suprise)
2729 <tt>GlobalVariable</tt> class. Like functions, <tt>GlobalVariable</tt>s are also
2730 subclasses of <a href="#GlobalValue"><tt>GlobalValue</tt></a>, and as such are
2731 always referenced by their address (global values must live in memory, so their
2732 "name" refers to their constant address). See
2733 <a href="#GlobalValue"><tt>GlobalValue</tt></a> for more on this. Global
2734 variables may have an initial value (which must be a
2735 <a href="#Constant"><tt>Constant</tt></a>), and if they have an initializer,
2736 they may be marked as "constant" themselves (indicating that their contents
2737 never change at runtime).</p>
2740 <!-- _______________________________________________________________________ -->
2741 <div class="doc_subsubsection">
2742 <a name="m_GlobalVariable">Important Public Members of the
2743 <tt>GlobalVariable</tt> class</a>
2746 <div class="doc_text">
2749 <li><tt>GlobalVariable(const </tt><tt><a href="#Type">Type</a> *Ty, bool
2750 isConstant, LinkageTypes& Linkage, <a href="#Constant">Constant</a>
2751 *Initializer = 0, const std::string &Name = "", Module* Parent = 0)</tt>
2753 <p>Create a new global variable of the specified type. If
2754 <tt>isConstant</tt> is true then the global variable will be marked as
2755 unchanging for the program. The Linkage parameter specifies the type of
2756 linkage (internal, external, weak, linkonce, appending) for the variable. If
2757 the linkage is InternalLinkage, WeakLinkage, or LinkOnceLinkage, then
2758 the resultant global variable will have internal linkage. AppendingLinkage
2759 concatenates together all instances (in different translation units) of the
2760 variable into a single variable but is only applicable to arrays. See
2761 the <a href="LangRef.html#modulestructure">LLVM Language Reference</a> for
2762 further details on linkage types. Optionally an initializer, a name, and the
2763 module to put the variable into may be specified for the global variable as
2766 <li><tt>bool isConstant() const</tt>
2768 <p>Returns true if this is a global variable that is known not to
2769 be modified at runtime.</p></li>
2771 <li><tt>bool hasInitializer()</tt>
2773 <p>Returns true if this <tt>GlobalVariable</tt> has an intializer.</p></li>
2775 <li><tt><a href="#Constant">Constant</a> *getInitializer()</tt>
2777 <p>Returns the intial value for a <tt>GlobalVariable</tt>. It is not legal
2778 to call this method if there is no initializer.</p></li>
2783 <!-- ======================================================================= -->
2784 <div class="doc_subsection">
2785 <a name="Module">The <tt>Module</tt> class</a>
2788 <div class="doc_text">
2791 href="/doxygen/Module_8h-source.html">llvm/Module.h</a>"</tt><br> doxygen info:
2792 <a href="/doxygen/classllvm_1_1Module.html">Module Class</a></p>
2794 <p>The <tt>Module</tt> class represents the top level structure present in LLVM
2795 programs. An LLVM module is effectively either a translation unit of the
2796 original program or a combination of several translation units merged by the
2797 linker. The <tt>Module</tt> class keeps track of a list of <a
2798 href="#Function"><tt>Function</tt></a>s, a list of <a
2799 href="#GlobalVariable"><tt>GlobalVariable</tt></a>s, and a <a
2800 href="#SymbolTable"><tt>SymbolTable</tt></a>. Additionally, it contains a few
2801 helpful member functions that try to make common operations easy.</p>
2805 <!-- _______________________________________________________________________ -->
2806 <div class="doc_subsubsection">
2807 <a name="m_Module">Important Public Members of the <tt>Module</tt> class</a>
2810 <div class="doc_text">
2813 <li><tt>Module::Module(std::string name = "")</tt></li>
2816 <p>Constructing a <a href="#Module">Module</a> is easy. You can optionally
2817 provide a name for it (probably based on the name of the translation unit).</p>
2820 <li><tt>Module::iterator</tt> - Typedef for function list iterator<br>
2821 <tt>Module::const_iterator</tt> - Typedef for const_iterator.<br>
2823 <tt>begin()</tt>, <tt>end()</tt>
2824 <tt>size()</tt>, <tt>empty()</tt>
2826 <p>These are forwarding methods that make it easy to access the contents of
2827 a <tt>Module</tt> object's <a href="#Function"><tt>Function</tt></a>
2830 <li><tt>Module::FunctionListType &getFunctionList()</tt>
2832 <p> Returns the list of <a href="#Function"><tt>Function</tt></a>s. This is
2833 necessary to use when you need to update the list or perform a complex
2834 action that doesn't have a forwarding method.</p>
2836 <p><!-- Global Variable --></p></li>
2842 <li><tt>Module::global_iterator</tt> - Typedef for global variable list iterator<br>
2844 <tt>Module::const_global_iterator</tt> - Typedef for const_iterator.<br>
2846 <tt>global_begin()</tt>, <tt>global_end()</tt>
2847 <tt>global_size()</tt>, <tt>global_empty()</tt>
2849 <p> These are forwarding methods that make it easy to access the contents of
2850 a <tt>Module</tt> object's <a
2851 href="#GlobalVariable"><tt>GlobalVariable</tt></a> list.</p></li>
2853 <li><tt>Module::GlobalListType &getGlobalList()</tt>
2855 <p>Returns the list of <a
2856 href="#GlobalVariable"><tt>GlobalVariable</tt></a>s. This is necessary to
2857 use when you need to update the list or perform a complex action that
2858 doesn't have a forwarding method.</p>
2860 <p><!-- Symbol table stuff --> </p></li>
2866 <li><tt><a href="#SymbolTable">SymbolTable</a> *getSymbolTable()</tt>
2868 <p>Return a reference to the <a href="#SymbolTable"><tt>SymbolTable</tt></a>
2869 for this <tt>Module</tt>.</p>
2871 <p><!-- Convenience methods --></p></li>
2877 <li><tt><a href="#Function">Function</a> *getFunction(const std::string
2878 &Name, const <a href="#FunctionType">FunctionType</a> *Ty)</tt>
2880 <p>Look up the specified function in the <tt>Module</tt> <a
2881 href="#SymbolTable"><tt>SymbolTable</tt></a>. If it does not exist, return
2882 <tt>null</tt>.</p></li>
2884 <li><tt><a href="#Function">Function</a> *getOrInsertFunction(const
2885 std::string &Name, const <a href="#FunctionType">FunctionType</a> *T)</tt>
2887 <p>Look up the specified function in the <tt>Module</tt> <a
2888 href="#SymbolTable"><tt>SymbolTable</tt></a>. If it does not exist, add an
2889 external declaration for the function and return it.</p></li>
2891 <li><tt>std::string getTypeName(const <a href="#Type">Type</a> *Ty)</tt>
2893 <p>If there is at least one entry in the <a
2894 href="#SymbolTable"><tt>SymbolTable</tt></a> for the specified <a
2895 href="#Type"><tt>Type</tt></a>, return it. Otherwise return the empty
2898 <li><tt>bool addTypeName(const std::string &Name, const <a
2899 href="#Type">Type</a> *Ty)</tt>
2901 <p>Insert an entry in the <a href="#SymbolTable"><tt>SymbolTable</tt></a>
2902 mapping <tt>Name</tt> to <tt>Ty</tt>. If there is already an entry for this
2903 name, true is returned and the <a
2904 href="#SymbolTable"><tt>SymbolTable</tt></a> is not modified.</p></li>
2909 <!-- ======================================================================= -->
2910 <div class="doc_subsection">
2911 <a name="Constant">The <tt>Constant</tt> class and subclasses</a>
2914 <div class="doc_text">
2916 <p>Constant represents a base class for different types of constants. It
2917 is subclassed by ConstantInt, ConstantArray, etc. for representing
2918 the various types of Constants.</p>
2922 <!-- _______________________________________________________________________ -->
2923 <div class="doc_subsubsection">
2924 <a name="m_Constant">Important Public Methods</a>
2926 <div class="doc_text">
2929 <!-- _______________________________________________________________________ -->
2930 <div class="doc_subsubsection">Important Subclasses of Constant </div>
2931 <div class="doc_text">
2933 <li>ConstantInt : This subclass of Constant represents an integer constant of
2934 any width, including boolean (1 bit integer).
2936 <li><tt>int64_t getSExtValue() const</tt>: Returns the underlying value of
2937 this constant as a sign extended signed integer value.</li>
2938 <li><tt>uint64_t getZExtValue() const</tt>: Returns the underlying value
2939 of this constant as a zero extended unsigned integer value.</li>
2940 <li><tt>static ConstantInt* get(const Type *Ty, uint64_t Val)</tt>:
2941 Returns the ConstantInt object that represents the value provided by
2942 <tt>Val</tt> for integer type <tt>Ty</tt>.</li>
2945 <li>ConstantFP : This class represents a floating point constant.
2947 <li><tt>double getValue() const</tt>: Returns the underlying value of
2948 this constant. </li>
2951 <li>ConstantArray : This represents a constant array.
2953 <li><tt>const std::vector<Use> &getValues() const</tt>: Returns
2954 a vector of component constants that makeup this array. </li>
2957 <li>ConstantStruct : This represents a constant struct.
2959 <li><tt>const std::vector<Use> &getValues() const</tt>: Returns
2960 a vector of component constants that makeup this array. </li>
2963 <li>GlobalValue : This represents either a global variable or a function. In
2964 either case, the value is a constant fixed address (after linking).
2968 <!-- ======================================================================= -->
2969 <div class="doc_subsection">
2970 <a name="Argument">The <tt>Argument</tt> class</a>
2973 <div class="doc_text">
2975 <p>This subclass of Value defines the interface for incoming formal
2976 arguments to a function. A Function maintains a list of its formal
2977 arguments. An argument has a pointer to the parent Function.</p>
2981 <!-- *********************************************************************** -->
2984 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
2985 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
2986 <a href="http://validator.w3.org/check/referer"><img
2987 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
2989 <a href="mailto:dhurjati@cs.uiuc.edu">Dinakar Dhurjati</a> and
2990 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
2991 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
2992 Last modified: $Date$