Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 18ed029a7b75b2a39d39cd4c7f2ec68284c10102 / . / docs / LangRef.html
blob: 73e7e2db1fa549ff25765e9f871d14f4527fb079 [file] [log] [blame]
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3<html>
4<head>
5 <title>LLVM Assembly Language Reference Manual</title>
Reid Spencer3921c742004-08-26 20:44:00 +0000[diff] [blame]6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
7 <meta name="author" content="Chris Lattner">
8 <meta name="description"
9 content="LLVM Assembly Language Reference Manual.">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]10 <link rel="stylesheet" href="llvm.css" type="text/css">
11</head>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]12
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]13<body>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]14
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]15<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]16<ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]17 <li><a href="#abstract">Abstract</a></li>
18 <li><a href="#introduction">Introduction</a></li>
19 <li><a href="#identifiers">Identifiers</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]20 <li><a href="#highlevel">High Level Structure</a>
21 <ol>
22 <li><a href="#modulestructure">Module Structure</a></li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]23 <li><a href="#linkage">Linkage Types</a></li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]24 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]25 <li><a href="#globalvars">Global Variables</a></li>
26 <li><a href="#functionstructure">Function Structure</a></li>
27 </ol>
28 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]29 <li><a href="#typesystem">Type System</a>
30 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]31 <li><a href="#t_primitive">Primitive Types</a>
32 <ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]33 <li><a href="#t_classifications">Type Classifications</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]34 </ol>
35 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]36 <li><a href="#t_derived">Derived Types</a>
37 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]38 <li><a href="#t_array">Array Type</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]39 <li><a href="#t_function">Function Type</a></li>
40 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]41 <li><a href="#t_struct">Structure Type</a></li>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]42 <li><a href="#t_packed">Packed Type</a></li>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]43 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]44 </ol>
45 </li>
46 </ol>
47 </li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]48 <li><a href="#constants">Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]49 <ol>
50 <li><a href="#simpleconstants">Simple Constants</a>
51 <li><a href="#aggregateconstants">Aggregate Constants</a>
52 <li><a href="#globalconstants">Global Variable and Function Addresses</a>
53 <li><a href="#undefvalues">Undefined Values</a>
54 <li><a href="#constantexprs">Constant Expressions</a>
55 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]56 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]57 <li><a href="#instref">Instruction Reference</a>
58 <ol>
59 <li><a href="#terminators">Terminator Instructions</a>
60 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]61 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
62 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]63 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
64 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]65 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]66 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]67 </ol>
68 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]69 <li><a href="#binaryops">Binary Operations</a>
70 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]71 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
72 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
73 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
74 <li><a href="#i_div">'<tt>div</tt>' Instruction</a></li>
75 <li><a href="#i_rem">'<tt>rem</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]76 <li><a href="#i_setcc">'<tt>set<i>cc</i></tt>' Instructions</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]77 </ol>
78 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]79 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
80 <ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]81 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]82 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]83 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
84 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
85 <li><a href="#i_shr">'<tt>shr</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]86 </ol>
87 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]88 <li><a href="#memoryops">Memory Access Operations</a>
89 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]90 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
91 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
92 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
93 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
94 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
95 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
96 </ol>
97 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]98 <li><a href="#otherops">Other Operations</a>
99 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]100 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]101 <li><a href="#i_cast">'<tt>cast .. to</tt>' Instruction</a></li>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]102 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]103 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]104 <li><a href="#i_vaarg">'<tt>vaarg</tt>' Instruction</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]105 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]106 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]107 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]108 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]109 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]110 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]111 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
112 <ol>
113 <li><a href="#i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
114 <li><a href="#i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
115 <li><a href="#i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
116 </ol>
117 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]118 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
119 <ol>
120 <li><a href="#i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
121 <li><a href="#i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
122 <li><a href="#i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
123 </ol>
124 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]125 <li><a href="#int_codegen">Code Generator Intrinsics</a>
126 <ol>
127 <li><a href="#i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
128 <li><a href="#i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]129 <li><a href="#i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]130 <li><a href="#i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]131 </ol>
132 </li>
133 <li><a href="#int_os">Operating System Intrinsics</a>
134 <ol>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]135 <li><a href="#i_readport">'<tt>llvm.readport</tt>' Intrinsic</a></li>
136 <li><a href="#i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a></li>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]137 <li><a href="#i_readio">'<tt>llvm.readio</tt>' Intrinsic</a></li>
138 <li><a href="#i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a></li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]139 </ol>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]140 <li><a href="#int_libc">Standard C Library Intrinsics</a>
141 <ol>
142 <li><a href="#i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a></li>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]143 <li><a href="#i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a></li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]144 <li><a href="#i_memset">'<tt>llvm.memset</tt>' Intrinsic</a></li>
Alkis Evlogimenos96853722004-06-12 19:19:14 +0000[diff] [blame]145 <li><a href="#i_isunordered">'<tt>llvm.isunordered</tt>' Intrinsic</a></li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]146 </ol>
147 </li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]148 <li><a href="#int_count">Bit counting Intrinsics</a>
149 <ol>
150 <li><a href="#int_ctpop">'<tt>llvm.ctpop</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]151 <li><a href="#int_ctlz">'<tt>llvm.ctlz</tt>' Intrinsic </a></li>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]152 <li><a href="#int_cttz">'<tt>llvm.cttz</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]153 </ol>
154 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]155 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]156 </ol>
157 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]158</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]159
160<div class="doc_author">
161 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
162 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]163</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]164
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]165<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]166<div class="doc_section"> <a name="abstract">Abstract </a></div>
167<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]168
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]169<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]170<p>This document is a reference manual for the LLVM assembly language.
171LLVM is an SSA based representation that provides type safety,
172low-level operations, flexibility, and the capability of representing
173'all' high-level languages cleanly. It is the common code
174representation used throughout all phases of the LLVM compilation
175strategy.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]176</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]177
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]178<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]179<div class="doc_section"> <a name="introduction">Introduction</a> </div>
180<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]181
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]182<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]183
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]184<p>The LLVM code representation is designed to be used in three
185different forms: as an in-memory compiler IR, as an on-disk bytecode
186representation (suitable for fast loading by a Just-In-Time compiler),
187and as a human readable assembly language representation. This allows
188LLVM to provide a powerful intermediate representation for efficient
189compiler transformations and analysis, while providing a natural means
190to debug and visualize the transformations. The three different forms
191of LLVM are all equivalent. This document describes the human readable
192representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]193
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]194<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]195while being expressive, typed, and extensible at the same time. It
196aims to be a "universal IR" of sorts, by being at a low enough level
197that high-level ideas may be cleanly mapped to it (similar to how
198microprocessors are "universal IR's", allowing many source languages to
199be mapped to them). By providing type information, LLVM can be used as
200the target of optimizations: for example, through pointer analysis, it
201can be proven that a C automatic variable is never accessed outside of
202the current function... allowing it to be promoted to a simple SSA
203value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]204
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]205</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]206
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]207<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]208<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]209
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]210<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]211
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]212<p>It is important to note that this document describes 'well formed'
213LLVM assembly language. There is a difference between what the parser
214accepts and what is considered 'well formed'. For example, the
215following instruction is syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]216
217<pre>
218 %x = <a href="#i_add">add</a> int 1, %x
219</pre>
220
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]221<p>...because the definition of <tt>%x</tt> does not dominate all of
222its uses. The LLVM infrastructure provides a verification pass that may
223be used to verify that an LLVM module is well formed. This pass is
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]224automatically run by the parser after parsing input assembly and by
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]225the optimizer before it outputs bytecode. The violations pointed out
226by the verifier pass indicate bugs in transformation passes or input to
227the parser.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]228
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]229<!-- Describe the typesetting conventions here. --> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]230
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]231<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]232<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]233<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]234
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]235<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]236
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]237<p>LLVM uses three different forms of identifiers, for different
238purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]239
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]240<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]241 <li>Named values are represented as a string of characters with a '%' prefix.
242 For example, %foo, %DivisionByZero, %a.really.long.identifier. The actual
243 regular expression used is '<tt>%[a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
244 Identifiers which require other characters in their names can be surrounded
245 with quotes. In this way, anything except a <tt>"</tt> character can be used
246 in a name.</li>
247
248 <li>Unnamed values are represented as an unsigned numeric value with a '%'
249 prefix. For example, %12, %2, %44.</li>
250
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]251 <li>Constants, which are described in a <a href="#constants">section about
252 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]253</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]254
255<p>LLVM requires that values start with a '%' sign for two reasons: Compilers
256don't need to worry about name clashes with reserved words, and the set of
257reserved words may be expanded in the future without penalty. Additionally,
258unnamed identifiers allow a compiler to quickly come up with a temporary
259variable without having to avoid symbol table conflicts.</p>
260
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]261<p>Reserved words in LLVM are very similar to reserved words in other
262languages. There are keywords for different opcodes ('<tt><a
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]263href="#i_add">add</a></tt>', '<tt><a href="#i_cast">cast</a></tt>', '<tt><a
264href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
265href="#t_void">void</a></tt>', '<tt><a href="#t_uint">uint</a></tt>', etc...),
266and others. These reserved words cannot conflict with variable names, because
267none of them start with a '%' character.</p>
268
269<p>Here is an example of LLVM code to multiply the integer variable
270'<tt>%X</tt>' by 8:</p>
271
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]272<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]273
274<pre>
275 %result = <a href="#i_mul">mul</a> uint %X, 8
276</pre>
277
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]278<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]279
280<pre>
281 %result = <a href="#i_shl">shl</a> uint %X, ubyte 3
282</pre>
283
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]284<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]285
286<pre>
287 <a href="#i_add">add</a> uint %X, %X <i>; yields {uint}:%0</i>
288 <a href="#i_add">add</a> uint %0, %0 <i>; yields {uint}:%1</i>
289 %result = <a href="#i_add">add</a> uint %1, %1
290</pre>
291
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]292<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
293important lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]294
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]295<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]296
297 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
298 line.</li>
299
300 <li>Unnamed temporaries are created when the result of a computation is not
301 assigned to a named value.</li>
302
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]303 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]304
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]305</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]306
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]307<p>...and it also shows a convention that we follow in this document. When
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]308demonstrating instructions, we will follow an instruction with a comment that
309defines the type and name of value produced. Comments are shown in italic
310text.</p>
311
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]312</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]313
314<!-- *********************************************************************** -->
315<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
316<!-- *********************************************************************** -->
317
318<!-- ======================================================================= -->
319<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
320</div>
321
322<div class="doc_text">
323
324<p>LLVM programs are composed of "Module"s, each of which is a
325translation unit of the input programs. Each module consists of
326functions, global variables, and symbol table entries. Modules may be
327combined together with the LLVM linker, which merges function (and
328global variable) definitions, resolves forward declarations, and merges
329symbol table entries. Here is an example of the "hello world" module:</p>
330
331<pre><i>; Declare the string constant as a global constant...</i>
332<a href="#identifiers">%.LC0</a> = <a href="#linkage_internal">internal</a> <a
333 href="#globalvars">constant</a> <a href="#t_array">[13 x sbyte]</a> c"hello world\0A\00" <i>; [13 x sbyte]*</i>
334
335<i>; External declaration of the puts function</i>
336<a href="#functionstructure">declare</a> int %puts(sbyte*) <i>; int(sbyte*)* </i>
337
338<i>; Definition of main function</i>
339int %main() { <i>; int()* </i>
340 <i>; Convert [13x sbyte]* to sbyte *...</i>
341 %cast210 = <a
342 href="#i_getelementptr">getelementptr</a> [13 x sbyte]* %.LC0, long 0, long 0 <i>; sbyte*</i>
343
344 <i>; Call puts function to write out the string to stdout...</i>
345 <a
346 href="#i_call">call</a> int %puts(sbyte* %cast210) <i>; int</i>
347 <a
348 href="#i_ret">ret</a> int 0<br>}<br></pre>
349
350<p>This example is made up of a <a href="#globalvars">global variable</a>
351named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
352function, and a <a href="#functionstructure">function definition</a>
353for "<tt>main</tt>".</p>
354
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]355<p>In general, a module is made up of a list of global values,
356where both functions and global variables are global values. Global values are
357represented by a pointer to a memory location (in this case, a pointer to an
358array of char, and a pointer to a function), and have one of the following <a
359href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]360
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]361</div>
362
363<!-- ======================================================================= -->
364<div class="doc_subsection">
365 <a name="linkage">Linkage Types</a>
366</div>
367
368<div class="doc_text">
369
370<p>
371All Global Variables and Functions have one of the following types of linkage:
372</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]373
374<dl>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]375
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]376 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]377
378 <dd>Global values with internal linkage are only directly accessible by
379 objects in the current module. In particular, linking code into a module with
380 an internal global value may cause the internal to be renamed as necessary to
381 avoid collisions. Because the symbol is internal to the module, all
382 references can be updated. This corresponds to the notion of the
383 '<tt>static</tt>' keyword in C, or the idea of "anonymous namespaces" in C++.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]384 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]385
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]386 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]387
388 <dd>"<tt>linkonce</tt>" linkage is similar to <tt>internal</tt> linkage, with
389 the twist that linking together two modules defining the same
390 <tt>linkonce</tt> globals will cause one of the globals to be discarded. This
391 is typically used to implement inline functions. Unreferenced
392 <tt>linkonce</tt> globals are allowed to be discarded.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]393 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]394
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]395 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]396
397 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
398 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
399 used to implement constructs in C such as "<tt>int X;</tt>" at global scope.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]400 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]401
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]402 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]403
404 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
405 pointer to array type. When two global variables with appending linkage are
406 linked together, the two global arrays are appended together. This is the
407 LLVM, typesafe, equivalent of having the system linker append together
408 "sections" with identical names when .o files are linked.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]409 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]410
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]411 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]412
413 <dd>If none of the above identifiers are used, the global is externally
414 visible, meaning that it participates in linkage and can be used to resolve
415 external symbol references.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]416 </dd>
417</dl>
418
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]419<p><a name="linkage_external">For example, since the "<tt>.LC0</tt>"
420variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
421variable and was linked with this one, one of the two would be renamed,
422preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
423external (i.e., lacking any linkage declarations), they are accessible
424outside of the current module. It is illegal for a function <i>declaration</i>
425to have any linkage type other than "externally visible".</a></p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]426
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]427</div>
428
429<!-- ======================================================================= -->
430<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]431 <a name="callingconv">Calling Conventions</a>
432</div>
433
434<div class="doc_text">
435
436<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
437and <a href="#i_invoke">invokes</a> can all have an optional calling convention
438specified for the call. The calling convention of any pair of dynamic
439caller/callee must match, or the behavior of the program is undefined. The
440following calling conventions are supported by LLVM, and more may be added in
441the future:</p>
442
443<dl>
444 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
445
446 <dd>This calling convention (the default if no other calling convention is
447 specified) matches the target C calling conventions. This calling convention
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]448 supports varargs function calls and tolerates some mismatch in the declared
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]449 prototype and implemented declaration of the function (as does normal C).
450 </dd>
451
452 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
453
454 <dd>This calling convention attempts to make calls as fast as possible
455 (e.g. by passing things in registers). This calling convention allows the
456 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattner8cdc5bc2005-05-06 23:08:23 +0000[diff] [blame]457 without having to conform to an externally specified ABI. Implementations of
458 this convention should allow arbitrary tail call optimization to be supported.
459 This calling convention does not support varargs and requires the prototype of
460 all callees to exactly match the prototype of the function definition.
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]461 </dd>
462
463 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
464
465 <dd>This calling convention attempts to make code in the caller as efficient
466 as possible under the assumption that the call is not commonly executed. As
467 such, these calls often preserve all registers so that the call does not break
468 any live ranges in the caller side. This calling convention does not support
469 varargs and requires the prototype of all callees to exactly match the
470 prototype of the function definition.
471 </dd>
472
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]473 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]474
475 <dd>Any calling convention may be specified by number, allowing
476 target-specific calling conventions to be used. Target specific calling
477 conventions start at 64.
478 </dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]479</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]480
481<p>More calling conventions can be added/defined on an as-needed basis, to
482support pascal conventions or any other well-known target-independent
483convention.</p>
484
485</div>
486
487<!-- ======================================================================= -->
488<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]489 <a name="globalvars">Global Variables</a>
490</div>
491
492<div class="doc_text">
493
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]494<p>Global variables define regions of memory allocated at compilation time
495instead of run-time. Global variables may optionally be initialized. A
496variable may be defined as a global "constant", which indicates that the
497contents of the variable will <b>never</b> be modified (enabling better
498optimization, allowing the global data to be placed in the read-only section of
499an executable, etc). Note that variables that need runtime initialization
500cannot be marked "constant", as there is a store to the variable.</p>
501
502<p>
503LLVM explicitly allows <em>declarations</em> of global variables to be marked
504constant, even if the final definition of the global is not. This capability
505can be used to enable slightly better optimization of the program, but requires
506the language definition to guarantee that optimizations based on the
507'constantness' are valid for the translation units that do not include the
508definition.
509</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]510
511<p>As SSA values, global variables define pointer values that are in
512scope (i.e. they dominate) all basic blocks in the program. Global
513variables always define a pointer to their "content" type because they
514describe a region of memory, and all memory objects in LLVM are
515accessed through pointers.</p>
516
517</div>
518
519
520<!-- ======================================================================= -->
521<div class="doc_subsection">
522 <a name="functionstructure">Functions</a>
523</div>
524
525<div class="doc_text">
526
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]527<p>LLVM function definitions consist of an optional <a href="#linkage">linkage
528type</a>, an optional <a href="#callingconv">calling convention</a>, a return
529type, a function name, a (possibly empty) argument list, an opening curly brace,
530a list of basic blocks, and a closing curly brace. LLVM function declarations
531are defined with the "<tt>declare</tt>" keyword, an optional <a
532href="#callingconv">calling convention</a>, a return type, a function name, and
533a possibly empty list of arguments.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]534
535<p>A function definition contains a list of basic blocks, forming the CFG for
536the function. Each basic block may optionally start with a label (giving the
537basic block a symbol table entry), contains a list of instructions, and ends
538with a <a href="#terminators">terminator</a> instruction (such as a branch or
539function return).</p>
540
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]541<p>The first basic block in a program is special in two ways: it is immediately
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]542executed on entrance to the function, and it is not allowed to have predecessor
543basic blocks (i.e. there can not be any branches to the entry block of a
544function). Because the block can have no predecessors, it also cannot have any
545<a href="#i_phi">PHI nodes</a>.</p>
546
547<p>LLVM functions are identified by their name and type signature. Hence, two
548functions with the same name but different parameter lists or return values are
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]549considered different functions, and LLVM will resolve references to each
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]550appropriately.</p>
551
552</div>
553
554
555
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]556<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]557<div class="doc_section"> <a name="typesystem">Type System</a> </div>
558<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]559
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]560<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]561
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]562<p>The LLVM type system is one of the most important features of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]563intermediate representation. Being typed enables a number of
564optimizations to be performed on the IR directly, without having to do
565extra analyses on the side before the transformation. A strong type
566system makes it easier to read the generated code and enables novel
567analyses and transformations that are not feasible to perform on normal
568three address code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]569
570</div>
571
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]572<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]573<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]574<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]575<p>The primitive types are the fundamental building blocks of the LLVM
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]576system. The current set of primitive types is as follows:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]577
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]578<table class="layout">
579 <tr class="layout">
580 <td class="left">
581 <table>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]582 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]583 <tr><th>Type</th><th>Description</th></tr>
584 <tr><td><tt>void</tt></td><td>No value</td></tr>
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]585 <tr><td><tt>ubyte</tt></td><td>Unsigned 8-bit value</td></tr>
586 <tr><td><tt>ushort</tt></td><td>Unsigned 16-bit value</td></tr>
587 <tr><td><tt>uint</tt></td><td>Unsigned 32-bit value</td></tr>
588 <tr><td><tt>ulong</tt></td><td>Unsigned 64-bit value</td></tr>
589 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]590 <tr><td><tt>label</tt></td><td>Branch destination</td></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]591 </tbody>
592 </table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]593 </td>
594 <td class="right">
595 <table>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]596 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]597 <tr><th>Type</th><th>Description</th></tr>
598 <tr><td><tt>bool</tt></td><td>True or False value</td></tr>
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]599 <tr><td><tt>sbyte</tt></td><td>Signed 8-bit value</td></tr>
600 <tr><td><tt>short</tt></td><td>Signed 16-bit value</td></tr>
601 <tr><td><tt>int</tt></td><td>Signed 32-bit value</td></tr>
602 <tr><td><tt>long</tt></td><td>Signed 64-bit value</td></tr>
603 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]604 </tbody>
605 </table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]606 </td>
607 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]608</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]609</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]610
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]611<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]612<div class="doc_subsubsection"> <a name="t_classifications">Type
613Classifications</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]614<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]615<p>These different primitive types fall into a few useful
616classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]617
618<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]619 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]620 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]621 <tr>
622 <td><a name="t_signed">signed</a></td>
623 <td><tt>sbyte, short, int, long, float, double</tt></td>
624 </tr>
625 <tr>
626 <td><a name="t_unsigned">unsigned</a></td>
627 <td><tt>ubyte, ushort, uint, ulong</tt></td>
628 </tr>
629 <tr>
630 <td><a name="t_integer">integer</a></td>
631 <td><tt>ubyte, sbyte, ushort, short, uint, int, ulong, long</tt></td>
632 </tr>
633 <tr>
634 <td><a name="t_integral">integral</a></td>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]635 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long</tt>
636 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]637 </tr>
638 <tr>
639 <td><a name="t_floating">floating point</a></td>
640 <td><tt>float, double</tt></td>
641 </tr>
642 <tr>
643 <td><a name="t_firstclass">first class</a></td>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]644 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long,<br>
645 float, double, <a href="#t_pointer">pointer</a>,
646 <a href="#t_packed">packed</a></tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]647 </tr>
648 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]649</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]650
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]651<p>The <a href="#t_firstclass">first class</a> types are perhaps the
652most important. Values of these types are the only ones which can be
653produced by instructions, passed as arguments, or used as operands to
654instructions. This means that all structures and arrays must be
655manipulated either by pointer or by component.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]656</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]657
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]658<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]659<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]660
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]661<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]662
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]663<p>The real power in LLVM comes from the derived types in the system.
664This is what allows a programmer to represent arrays, functions,
665pointers, and other useful types. Note that these derived types may be
666recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]667
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]668</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]669
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]670<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]671<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]672
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]673<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]674
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]675<h5>Overview:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]676
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]677<p>The array type is a very simple derived type that arranges elements
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]678sequentially in memory. The array type requires a size (number of
679elements) and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]680
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]681<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]682
683<pre>
684 [&lt;# elements&gt; x &lt;elementtype&gt;]
685</pre>
686
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]687<p>The number of elements is a constant integer value; elementtype may
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]688be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]689
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]690<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]691<table class="layout">
692 <tr class="layout">
693 <td class="left">
694 <tt>[40 x int ]</tt><br/>
695 <tt>[41 x int ]</tt><br/>
696 <tt>[40 x uint]</tt><br/>
697 </td>
698 <td class="left">
699 Array of 40 integer values.<br/>
700 Array of 41 integer values.<br/>
701 Array of 40 unsigned integer values.<br/>
702 </td>
703 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]704</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]705<p>Here are some examples of multidimensional arrays:</p>
706<table class="layout">
707 <tr class="layout">
708 <td class="left">
709 <tt>[3 x [4 x int]]</tt><br/>
710 <tt>[12 x [10 x float]]</tt><br/>
711 <tt>[2 x [3 x [4 x uint]]]</tt><br/>
712 </td>
713 <td class="left">
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]714 3x4 array of integer values.<br/>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]715 12x10 array of single precision floating point values.<br/>
716 2x3x4 array of unsigned integer values.<br/>
717 </td>
718 </tr>
719</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]720
721<p>Note that 'variable sized arrays' can be implemented in LLVM With a zero
722length array. Normally accesses past the end of an array are undefined in
723LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
724As a special case, however, zero length arrays are recognized to be variable
725length. This allows implementation of 'pascal style arrays' with the LLVM
726type "{ int, [0 x float]}", for example.</p>
727
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]728</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]729
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]730<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]731<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]732<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]733<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]734<p>The function type can be thought of as a function signature. It
735consists of a return type and a list of formal parameter types.
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]736Function types are usually used to build virtual function tables
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]737(which are structures of pointers to functions), for indirect function
738calls, and when defining a function.</p>
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]739<p>
740The return type of a function type cannot be an aggregate type.
741</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]742<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]743<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]744<p>Where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
745specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]746which indicates that the function takes a variable number of arguments.
747Variable argument functions can access their arguments with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]748 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]749<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]750<table class="layout">
751 <tr class="layout">
752 <td class="left">
753 <tt>int (int)</tt> <br/>
754 <tt>float (int, int *) *</tt><br/>
755 <tt>int (sbyte *, ...)</tt><br/>
756 </td>
757 <td class="left">
758 function taking an <tt>int</tt>, returning an <tt>int</tt><br/>
759 <a href="#t_pointer">Pointer</a> to a function that takes an
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]760 <tt>int</tt> and a <a href="#t_pointer">pointer</a> to <tt>int</tt>,
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]761 returning <tt>float</tt>.<br/>
762 A vararg function that takes at least one <a href="#t_pointer">pointer</a>
763 to <tt>sbyte</tt> (signed char in C), which returns an integer. This is
764 the signature for <tt>printf</tt> in LLVM.<br/>
765 </td>
766 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]767</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]768
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]769</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]770<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]771<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]772<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]773<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]774<p>The structure type is used to represent a collection of data members
775together in memory. The packing of the field types is defined to match
776the ABI of the underlying processor. The elements of a structure may
777be any type that has a size.</p>
778<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
779and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
780field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
781instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]782<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]783<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]784<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]785<table class="layout">
786 <tr class="layout">
787 <td class="left">
788 <tt>{ int, int, int }</tt><br/>
789 <tt>{ float, int (int) * }</tt><br/>
790 </td>
791 <td class="left">
792 a triple of three <tt>int</tt> values<br/>
793 A pair, where the first element is a <tt>float</tt> and the second element
794 is a <a href="#t_pointer">pointer</a> to a <a href="#t_function">function</a>
795 that takes an <tt>int</tt>, returning an <tt>int</tt>.<br/>
796 </td>
797 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]798</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]799</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]800
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]801<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]802<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]803<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]804<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]805<p>As in many languages, the pointer type represents a pointer or
806reference to another object, which must live in memory.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]807<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]808<pre> &lt;type&gt; *<br></pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]809<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]810<table class="layout">
811 <tr class="layout">
812 <td class="left">
813 <tt>[4x int]*</tt><br/>
814 <tt>int (int *) *</tt><br/>
815 </td>
816 <td class="left">
817 A <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a> of
818 four <tt>int</tt> values<br/>
819 A <a href="#t_pointer">pointer</a> to a <a
Chris Lattnera977c482005-02-19 02:22:14 +0000[diff] [blame]820 href="#t_function">function</a> that takes an <tt>int*</tt>, returning an
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]821 <tt>int</tt>.<br/>
822 </td>
823 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]824</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]825</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]826
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]827<!-- _______________________________________________________________________ -->
828<div class="doc_subsubsection"> <a name="t_packed">Packed Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]829<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]830
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]831<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]832
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]833<p>A packed type is a simple derived type that represents a vector
834of elements. Packed types are used when multiple primitive data
835are operated in parallel using a single instruction (SIMD).
836A packed type requires a size (number of
837elements) and an underlying primitive data type. Packed types are
838considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]839
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]840<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]841
842<pre>
843 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
844</pre>
845
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]846<p>The number of elements is a constant integer value; elementtype may
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]847be any integral or floating point type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]848
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]849<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]850
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]851<table class="layout">
852 <tr class="layout">
853 <td class="left">
854 <tt>&lt;4 x int&gt;</tt><br/>
855 <tt>&lt;8 x float&gt;</tt><br/>
856 <tt>&lt;2 x uint&gt;</tt><br/>
857 </td>
858 <td class="left">
859 Packed vector of 4 integer values.<br/>
860 Packed vector of 8 floating-point values.<br/>
861 Packed vector of 2 unsigned integer values.<br/>
862 </td>
863 </tr>
864</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]865</div>
866
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]867<!-- _______________________________________________________________________ -->
868<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
869<div class="doc_text">
870
871<h5>Overview:</h5>
872
873<p>Opaque types are used to represent unknown types in the system. This
874corresponds (for example) to the C notion of a foward declared structure type.
875In LLVM, opaque types can eventually be resolved to any type (not just a
876structure type).</p>
877
878<h5>Syntax:</h5>
879
880<pre>
881 opaque
882</pre>
883
884<h5>Examples:</h5>
885
886<table class="layout">
887 <tr class="layout">
888 <td class="left">
889 <tt>opaque</tt>
890 </td>
891 <td class="left">
892 An opaque type.<br/>
893 </td>
894 </tr>
895</table>
896</div>
897
898
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]899<!-- *********************************************************************** -->
900<div class="doc_section"> <a name="constants">Constants</a> </div>
901<!-- *********************************************************************** -->
902
903<div class="doc_text">
904
905<p>LLVM has several different basic types of constants. This section describes
906them all and their syntax.</p>
907
908</div>
909
910<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]911<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]912
913<div class="doc_text">
914
915<dl>
916 <dt><b>Boolean constants</b></dt>
917
918 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
919 constants of the <tt><a href="#t_primitive">bool</a></tt> type.
920 </dd>
921
922 <dt><b>Integer constants</b></dt>
923
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]924 <dd>Standard integers (such as '4') are constants of the <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]925 href="#t_integer">integer</a> type. Negative numbers may be used with signed
926 integer types.
927 </dd>
928
929 <dt><b>Floating point constants</b></dt>
930
931 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
932 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]933 notation (see below). Floating point constants must have a <a
934 href="#t_floating">floating point</a> type. </dd>
935
936 <dt><b>Null pointer constants</b></dt>
937
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]938 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]939 and must be of <a href="#t_pointer">pointer type</a>.</dd>
940
941</dl>
942
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]943<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]944of floating point constants. For example, the form '<tt>double
9450x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
9464.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]947(and the only time that they are generated by the disassembler) is when a
948floating point constant must be emitted but it cannot be represented as a
949decimal floating point number. For example, NaN's, infinities, and other
950special values are represented in their IEEE hexadecimal format so that
951assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]952
953</div>
954
955<!-- ======================================================================= -->
956<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
957</div>
958
959<div class="doc_text">
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]960<p>Aggregate constants arise from aggregation of simple constants
961and smaller aggregate constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]962
963<dl>
964 <dt><b>Structure constants</b></dt>
965
966 <dd>Structure constants are represented with notation similar to structure
967 type definitions (a comma separated list of elements, surrounded by braces
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]968 (<tt>{}</tt>)). For example: "<tt>{ int 4, float 17.0, int* %G }</tt>",
969 where "<tt>%G</tt>" is declared as "<tt>%G = external global int</tt>". Structure constants
970 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]971 types of elements must match those specified by the type.
972 </dd>
973
974 <dt><b>Array constants</b></dt>
975
976 <dd>Array constants are represented with notation similar to array type
977 definitions (a comma separated list of elements, surrounded by square brackets
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]978 (<tt>[]</tt>)). For example: "<tt>[ int 42, int 11, int 74 ]</tt>". Array
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]979 constants must have <a href="#t_array">array type</a>, and the number and
980 types of elements must match those specified by the type.
981 </dd>
982
983 <dt><b>Packed constants</b></dt>
984
985 <dd>Packed constants are represented with notation similar to packed type
986 definitions (a comma separated list of elements, surrounded by
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]987 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; int 42,
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]988 int 11, int 74, int 100 &gt;</tt>". Packed constants must have <a
989 href="#t_packed">packed type</a>, and the number and types of elements must
990 match those specified by the type.
991 </dd>
992
993 <dt><b>Zero initialization</b></dt>
994
995 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
996 value to zero of <em>any</em> type, including scalar and aggregate types.
997 This is often used to avoid having to print large zero initializers (e.g. for
998 large arrays), and is always exactly equivalent to using explicit zero
999 initializers.
1000 </dd>
1001</dl>
1002
1003</div>
1004
1005<!-- ======================================================================= -->
1006<div class="doc_subsection">
1007 <a name="globalconstants">Global Variable and Function Addresses</a>
1008</div>
1009
1010<div class="doc_text">
1011
1012<p>The addresses of <a href="#globalvars">global variables</a> and <a
1013href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1014constants. These constants are explicitly referenced when the <a
1015href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1016href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1017file:</p>
1018
1019<pre>
1020 %X = global int 17
1021 %Y = global int 42
1022 %Z = global [2 x int*] [ int* %X, int* %Y ]
1023</pre>
1024
1025</div>
1026
1027<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1028<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1029<div class="doc_text">
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1030 <p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1031 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1032 a constant is permitted.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1033
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1034 <p>Undefined values indicate to the compiler that the program is well defined
1035 no matter what value is used, giving the compiler more freedom to optimize.
1036 </p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1037</div>
1038
1039<!-- ======================================================================= -->
1040<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1041</div>
1042
1043<div class="doc_text">
1044
1045<p>Constant expressions are used to allow expressions involving other constants
1046to be used as constants. Constant expressions may be of any <a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1047href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1048that does not have side effects (e.g. load and call are not supported). The
1049following is the syntax for constant expressions:</p>
1050
1051<dl>
1052 <dt><b><tt>cast ( CST to TYPE )</tt></b></dt>
1053
1054 <dd>Cast a constant to another type.</dd>
1055
1056 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1057
1058 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1059 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1060 instruction, the index list may have zero or more indexes, which are required
1061 to make sense for the type of "CSTPTR".</dd>
1062
1063 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1064
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1065 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1066 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1067 binary</a> operations. The constraints on operands are the same as those for
1068 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1069 values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1070</dl>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1071</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]1072
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1073<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1074<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1075<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1076
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1077<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1078
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1079<p>The LLVM instruction set consists of several different
1080classifications of instructions: <a href="#terminators">terminator
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1081instructions</a>, <a href="#binaryops">binary instructions</a>,
1082<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1083 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1084instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1085
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1086</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1087
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1088<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1089<div class="doc_subsection"> <a name="terminators">Terminator
1090Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1091
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1092<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1093
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1094<p>As mentioned <a href="#functionstructure">previously</a>, every
1095basic block in a program ends with a "Terminator" instruction, which
1096indicates which block should be executed after the current block is
1097finished. These terminator instructions typically yield a '<tt>void</tt>'
1098value: they produce control flow, not values (the one exception being
1099the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1100<p>There are six different terminator instructions: the '<a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1101 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1102instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1103the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1104 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1105 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1106
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1107</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1108
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1109<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1110<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1111Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1112<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1113<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1114<pre> ret &lt;type&gt; &lt;value&gt; <i>; Return a value from a non-void function</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1115 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1116</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1117<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1118<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1119value) from a function back to the caller.</p>
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]1120<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1121returns a value and then causes control flow, and one that just causes
1122control flow to occur.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1123<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1124<p>The '<tt>ret</tt>' instruction may return any '<a
1125 href="#t_firstclass">first class</a>' type. Notice that a function is
1126not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1127instruction inside of the function that returns a value that does not
1128match the return type of the function.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1129<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1130<p>When the '<tt>ret</tt>' instruction is executed, control flow
1131returns back to the calling function's context. If the caller is a "<a
John Criswellfa081872004-06-25 15:16:57 +0000[diff] [blame]1132 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1133the instruction after the call. If the caller was an "<a
1134 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1135at the beginning of the "normal" destination block. If the instruction
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1136returns a value, that value shall set the call or invoke instruction's
1137return value.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1138<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1139<pre> ret int 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1140 ret void <i>; Return from a void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1141</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1142</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1143<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1144<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1145<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1146<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1147<pre> br bool &lt;cond&gt;, label &lt;iftrue&gt;, label &lt;iffalse&gt;<br> br label &lt;dest&gt; <i>; Unconditional branch</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1148</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1149<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1150<p>The '<tt>br</tt>' instruction is used to cause control flow to
1151transfer to a different basic block in the current function. There are
1152two forms of this instruction, corresponding to a conditional branch
1153and an unconditional branch.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1154<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1155<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
1156single '<tt>bool</tt>' value and two '<tt>label</tt>' values. The
1157unconditional form of the '<tt>br</tt>' instruction takes a single '<tt>label</tt>'
1158value as a target.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1159<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1160<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>bool</tt>'
1161argument is evaluated. If the value is <tt>true</tt>, control flows
1162to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1163control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1164<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1165<pre>Test:<br> %cond = <a href="#i_setcc">seteq</a> int %a, %b<br> br bool %cond, label %IfEqual, label %IfUnequal<br>IfEqual:<br> <a
1166 href="#i_ret">ret</a> int 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> int 0<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1167</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1168<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1169<div class="doc_subsubsection">
1170 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1171</div>
1172
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1173<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1174<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1175
1176<pre>
1177 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1178</pre>
1179
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1180<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1181
1182<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1183several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1184instruction, allowing a branch to occur to one of many possible
1185destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1186
1187
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1188<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1189
1190<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1191comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1192an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1193table is not allowed to contain duplicate constant entries.</p>
1194
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1195<h5>Semantics:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1196
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1197<p>The <tt>switch</tt> instruction specifies a table of values and
1198destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1199table is searched for the given value. If the value is found, control flow is
1200transfered to the corresponding destination; otherwise, control flow is
1201transfered to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1202
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1203<h5>Implementation:</h5>
1204
1205<p>Depending on properties of the target machine and the particular
1206<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1207ways. For example, it could be generated as a series of chained conditional
1208branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1209
1210<h5>Example:</h5>
1211
1212<pre>
1213 <i>; Emulate a conditional br instruction</i>
1214 %Val = <a href="#i_cast">cast</a> bool %value to int
1215 switch int %Val, label %truedest [int 0, label %falsedest ]
1216
1217 <i>; Emulate an unconditional br instruction</i>
1218 switch uint 0, label %dest [ ]
1219
1220 <i>; Implement a jump table:</i>
1221 switch uint %val, label %otherwise [ uint 0, label %onzero
1222 uint 1, label %onone
1223 uint 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1224</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1225</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1226
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1227<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1228<div class="doc_subsubsection">
1229 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1230</div>
1231
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1232<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1233
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1234<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1235
1236<pre>
1237 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)
1238 to label &lt;normal label&gt; except label &lt;exception label&gt;
1239</pre>
1240
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1241<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1242
1243<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1244function, with the possibility of control flow transfer to either the
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1245'<tt>normal</tt>' label or the
1246'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1247"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1248"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1249href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1250continued at the dynamically nearest "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1251
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1252<h5>Arguments:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1253
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1254<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1255
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1256<ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1257 <li>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1258 The optional "cconv" marker indicates which <a href="callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1259 convention</a> the call should use. If none is specified, the call defaults
1260 to using C calling conventions.
1261 </li>
1262 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1263 function value being invoked. In most cases, this is a direct function
1264 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1265 an arbitrary pointer to function value.
1266 </li>
1267
1268 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1269 function to be invoked. </li>
1270
1271 <li>'<tt>function args</tt>': argument list whose types match the function
1272 signature argument types. If the function signature indicates the function
1273 accepts a variable number of arguments, the extra arguments can be
1274 specified. </li>
1275
1276 <li>'<tt>normal label</tt>': the label reached when the called function
1277 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1278
1279 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1280 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1281
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1282</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1283
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1284<h5>Semantics:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1285
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1286<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1287href="#i_call">call</a></tt>' instruction in most regards. The primary
1288difference is that it establishes an association with a label, which is used by
1289the runtime library to unwind the stack.</p>
1290
1291<p>This instruction is used in languages with destructors to ensure that proper
1292cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1293exception. Additionally, this is important for implementation of
1294'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1295
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1296<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1297<pre>
1298 %retval = invoke int %Test(int 15) to label %Continue
1299 except label %TestCleanup <i>; {int}:retval set</i>
1300 %retval = invoke <a href="#callingconv">coldcc</a> int %Test(int 15) to label %Continue
1301 except label %TestCleanup <i>; {int}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1302</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1303</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1304
1305
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1306<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1307
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1308<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1309Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1310
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1311<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1312
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1313<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1314<pre>
1315 unwind
1316</pre>
1317
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1318<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1319
1320<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1321at the first callee in the dynamic call stack which used an <a
1322href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1323primarily used to implement exception handling.</p>
1324
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1325<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1326
1327<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
1328immediately halt. The dynamic call stack is then searched for the first <a
1329href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
1330execution continues at the "exceptional" destination block specified by the
1331<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
1332dynamic call chain, undefined behavior results.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1333</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1334
1335<!-- _______________________________________________________________________ -->
1336
1337<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
1338Instruction</a> </div>
1339
1340<div class="doc_text">
1341
1342<h5>Syntax:</h5>
1343<pre>
1344 unreachable
1345</pre>
1346
1347<h5>Overview:</h5>
1348
1349<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
1350instruction is used to inform the optimizer that a particular portion of the
1351code is not reachable. This can be used to indicate that the code after a
1352no-return function cannot be reached, and other facts.</p>
1353
1354<h5>Semantics:</h5>
1355
1356<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
1357</div>
1358
1359
1360
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1361<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1362<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1363<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1364<p>Binary operators are used to do most of the computation in a
1365program. They require two operands, execute an operation on them, and
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1366produce a single value. The operands might represent
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1367multiple data, as is the case with the <a href="#t_packed">packed</a> data type.
1368The result value of a binary operator is not
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1369necessarily the same type as its operands.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1370<p>There are several different binary operators:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1371</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1372<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1373<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
1374Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1375<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1376<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1377<pre> &lt;result&gt; = add &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1378</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1379<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1380<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1381<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1382<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1383 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
1384 This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1385Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1386<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1387<p>The value produced is the integer or floating point sum of the two
1388operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1389<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1390<pre> &lt;result&gt; = add int 4, %var <i>; yields {int}:result = 4 + %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1391</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1392</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1393<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1394<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
1395Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1396<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1397<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1398<pre> &lt;result&gt; = sub &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1399</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1400<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1401<p>The '<tt>sub</tt>' instruction returns the difference of its two
1402operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1403<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
1404instruction present in most other intermediate representations.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1405<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1406<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1407 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1408values.
1409This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1410Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1411<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1412<p>The value produced is the integer or floating point difference of
1413the two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1414<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1415<pre> &lt;result&gt; = sub int 4, %var <i>; yields {int}:result = 4 - %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1416 &lt;result&gt; = sub int 0, %val <i>; yields {int}:result = -%var</i>
1417</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1418</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1419<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1420<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
1421Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1422<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1423<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1424<pre> &lt;result&gt; = mul &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1425</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1426<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1427<p>The '<tt>mul</tt>' instruction returns the product of its two
1428operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1429<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1430<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1431 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1432values.
1433This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1434Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1435<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1436<p>The value produced is the integer or floating point product of the
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1437two operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1438<p>There is no signed vs unsigned multiplication. The appropriate
1439action is taken based on the type of the operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1440<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1441<pre> &lt;result&gt; = mul int 4, %var <i>; yields {int}:result = 4 * %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1442</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1443</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1444<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1445<div class="doc_subsubsection"> <a name="i_div">'<tt>div</tt>'
1446Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1447<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1448<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1449<pre> &lt;result&gt; = div &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1450</pre>
1451<h5>Overview:</h5>
1452<p>The '<tt>div</tt>' instruction returns the quotient of its two
1453operands.</p>
1454<h5>Arguments:</h5>
1455<p>The two arguments to the '<tt>div</tt>' instruction must be either <a
1456 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1457values.
1458This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1459Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1460<h5>Semantics:</h5>
1461<p>The value produced is the integer or floating point quotient of the
1462two operands.</p>
1463<h5>Example:</h5>
1464<pre> &lt;result&gt; = div int 4, %var <i>; yields {int}:result = 4 / %var</i>
1465</pre>
1466</div>
1467<!-- _______________________________________________________________________ -->
1468<div class="doc_subsubsection"> <a name="i_rem">'<tt>rem</tt>'
1469Instruction</a> </div>
1470<div class="doc_text">
1471<h5>Syntax:</h5>
1472<pre> &lt;result&gt; = rem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1473</pre>
1474<h5>Overview:</h5>
1475<p>The '<tt>rem</tt>' instruction returns the remainder from the
1476division of its two operands.</p>
1477<h5>Arguments:</h5>
1478<p>The two arguments to the '<tt>rem</tt>' instruction must be either <a
1479 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1480values.
1481This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1482Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1483<h5>Semantics:</h5>
1484<p>This returns the <i>remainder</i> of a division (where the result
1485has the same sign as the divisor), not the <i>modulus</i> (where the
1486result has the same sign as the dividend) of a value. For more
1487information about the difference, see: <a
1488 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
1489Math Forum</a>.</p>
1490<h5>Example:</h5>
1491<pre> &lt;result&gt; = rem int 4, %var <i>; yields {int}:result = 4 % %var</i>
1492</pre>
1493</div>
1494<!-- _______________________________________________________________________ -->
1495<div class="doc_subsubsection"> <a name="i_setcc">'<tt>set<i>cc</i></tt>'
1496Instructions</a> </div>
1497<div class="doc_text">
1498<h5>Syntax:</h5>
1499<pre> &lt;result&gt; = seteq &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1500 &lt;result&gt; = setne &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1501 &lt;result&gt; = setlt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1502 &lt;result&gt; = setgt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1503 &lt;result&gt; = setle &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1504 &lt;result&gt; = setge &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1505</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1506<h5>Overview:</h5>
1507<p>The '<tt>set<i>cc</i></tt>' family of instructions returns a boolean
1508value based on a comparison of their two operands.</p>
1509<h5>Arguments:</h5>
1510<p>The two arguments to the '<tt>set<i>cc</i></tt>' instructions must
1511be of <a href="#t_firstclass">first class</a> type (it is not possible
1512to compare '<tt>label</tt>'s, '<tt>array</tt>'s, '<tt>structure</tt>'
1513or '<tt>void</tt>' values, etc...). Both arguments must have identical
1514types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1515<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1516<p>The '<tt>seteq</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1517value if both operands are equal.<br>
1518The '<tt>setne</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1519value if both operands are unequal.<br>
1520The '<tt>setlt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1521value if the first operand is less than the second operand.<br>
1522The '<tt>setgt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1523value if the first operand is greater than the second operand.<br>
1524The '<tt>setle</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1525value if the first operand is less than or equal to the second operand.<br>
1526The '<tt>setge</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1527value if the first operand is greater than or equal to the second
1528operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1529<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1530<pre> &lt;result&gt; = seteq int 4, 5 <i>; yields {bool}:result = false</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1531 &lt;result&gt; = setne float 4, 5 <i>; yields {bool}:result = true</i>
1532 &lt;result&gt; = setlt uint 4, 5 <i>; yields {bool}:result = true</i>
1533 &lt;result&gt; = setgt sbyte 4, 5 <i>; yields {bool}:result = false</i>
1534 &lt;result&gt; = setle sbyte 4, 5 <i>; yields {bool}:result = true</i>
1535 &lt;result&gt; = setge sbyte 4, 5 <i>; yields {bool}:result = false</i>
1536</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1537</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1538<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1539<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
1540Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1541<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1542<p>Bitwise binary operators are used to do various forms of
1543bit-twiddling in a program. They are generally very efficient
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1544instructions and can commonly be strength reduced from other
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1545instructions. They require two operands, execute an operation on them,
1546and produce a single value. The resulting value of the bitwise binary
1547operators is always the same type as its first operand.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1548</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1549<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1550<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
1551Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1552<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1553<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1554<pre> &lt;result&gt; = and &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1555</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1556<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1557<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
1558its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1559<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1560<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1561 href="#t_integral">integral</a> values. Both arguments must have
1562identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1563<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1564<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1565<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1566<div style="align: center">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1567<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1568 <tbody>
1569 <tr>
1570 <td>In0</td>
1571 <td>In1</td>
1572 <td>Out</td>
1573 </tr>
1574 <tr>
1575 <td>0</td>
1576 <td>0</td>
1577 <td>0</td>
1578 </tr>
1579 <tr>
1580 <td>0</td>
1581 <td>1</td>
1582 <td>0</td>
1583 </tr>
1584 <tr>
1585 <td>1</td>
1586 <td>0</td>
1587 <td>0</td>
1588 </tr>
1589 <tr>
1590 <td>1</td>
1591 <td>1</td>
1592 <td>1</td>
1593 </tr>
1594 </tbody>
1595</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1596</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1597<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1598<pre> &lt;result&gt; = and int 4, %var <i>; yields {int}:result = 4 &amp; %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1599 &lt;result&gt; = and int 15, 40 <i>; yields {int}:result = 8</i>
1600 &lt;result&gt; = and int 4, 8 <i>; yields {int}:result = 0</i>
1601</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1602</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1603<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1604<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1605<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1606<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1607<pre> &lt;result&gt; = or &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1608</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1609<h5>Overview:</h5>
1610<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
1611or of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1612<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1613<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1614 href="#t_integral">integral</a> values. Both arguments must have
1615identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1616<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1617<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1618<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1619<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1620<table border="1" cellspacing="0" cellpadding="4">
1621 <tbody>
1622 <tr>
1623 <td>In0</td>
1624 <td>In1</td>
1625 <td>Out</td>
1626 </tr>
1627 <tr>
1628 <td>0</td>
1629 <td>0</td>
1630 <td>0</td>
1631 </tr>
1632 <tr>
1633 <td>0</td>
1634 <td>1</td>
1635 <td>1</td>
1636 </tr>
1637 <tr>
1638 <td>1</td>
1639 <td>0</td>
1640 <td>1</td>
1641 </tr>
1642 <tr>
1643 <td>1</td>
1644 <td>1</td>
1645 <td>1</td>
1646 </tr>
1647 </tbody>
1648</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1649</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1650<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1651<pre> &lt;result&gt; = or int 4, %var <i>; yields {int}:result = 4 | %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1652 &lt;result&gt; = or int 15, 40 <i>; yields {int}:result = 47</i>
1653 &lt;result&gt; = or int 4, 8 <i>; yields {int}:result = 12</i>
1654</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1655</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1656<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1657<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
1658Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1659<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1660<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1661<pre> &lt;result&gt; = xor &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1662</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1663<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1664<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
1665or of its two operands. The <tt>xor</tt> is used to implement the
1666"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1667<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1668<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1669 href="#t_integral">integral</a> values. Both arguments must have
1670identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1671<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1672<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1673<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1674<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1675<table border="1" cellspacing="0" cellpadding="4">
1676 <tbody>
1677 <tr>
1678 <td>In0</td>
1679 <td>In1</td>
1680 <td>Out</td>
1681 </tr>
1682 <tr>
1683 <td>0</td>
1684 <td>0</td>
1685 <td>0</td>
1686 </tr>
1687 <tr>
1688 <td>0</td>
1689 <td>1</td>
1690 <td>1</td>
1691 </tr>
1692 <tr>
1693 <td>1</td>
1694 <td>0</td>
1695 <td>1</td>
1696 </tr>
1697 <tr>
1698 <td>1</td>
1699 <td>1</td>
1700 <td>0</td>
1701 </tr>
1702 </tbody>
1703</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1704</div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1705<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1706<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1707<pre> &lt;result&gt; = xor int 4, %var <i>; yields {int}:result = 4 ^ %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1708 &lt;result&gt; = xor int 15, 40 <i>; yields {int}:result = 39</i>
1709 &lt;result&gt; = xor int 4, 8 <i>; yields {int}:result = 12</i>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1710 &lt;result&gt; = xor int %V, -1 <i>; yields {int}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1711</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1712</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1713<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1714<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
1715Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1716<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1717<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1718<pre> &lt;result&gt; = shl &lt;ty&gt; &lt;var1&gt;, ubyte &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1719</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1720<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1721<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
1722the left a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1723<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1724<p>The first argument to the '<tt>shl</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1725 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1726type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1727<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1728<p>The value produced is <tt>var1</tt> * 2<sup><tt>var2</tt></sup>.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1729<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1730<pre> &lt;result&gt; = shl int 4, ubyte %var <i>; yields {int}:result = 4 &lt;&lt; %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1731 &lt;result&gt; = shl int 4, ubyte 2 <i>; yields {int}:result = 16</i>
1732 &lt;result&gt; = shl int 1, ubyte 10 <i>; yields {int}:result = 1024</i>
1733</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1734</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1735<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1736<div class="doc_subsubsection"> <a name="i_shr">'<tt>shr</tt>'
1737Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1738<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1739<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1740<pre> &lt;result&gt; = shr &lt;ty&gt; &lt;var1&gt;, ubyte &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1741</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1742<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1743<p>The '<tt>shr</tt>' instruction returns the first operand shifted to
1744the right a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1745<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1746<p>The first argument to the '<tt>shr</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1747 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1748type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1749<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1750<p>If the first argument is a <a href="#t_signed">signed</a> type, the
1751most significant bit is duplicated in the newly free'd bit positions.
1752If the first argument is unsigned, zero bits shall fill the empty
1753positions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1754<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1755<pre> &lt;result&gt; = shr int 4, ubyte %var <i>; yields {int}:result = 4 &gt;&gt; %var</i>
Chris Lattner8c6bb902003-06-18 21:30:51 +0000[diff] [blame]1756 &lt;result&gt; = shr uint 4, ubyte 1 <i>; yields {uint}:result = 2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1757 &lt;result&gt; = shr int 4, ubyte 2 <i>; yields {int}:result = 1</i>
Chris Lattner8c6bb902003-06-18 21:30:51 +0000[diff] [blame]1758 &lt;result&gt; = shr sbyte 4, ubyte 3 <i>; yields {sbyte}:result = 0</i>
1759 &lt;result&gt; = shr sbyte -2, ubyte 1 <i>; yields {sbyte}:result = -1</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1760</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1761</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1762<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1763<div class="doc_subsection"> <a name="memoryops">Memory Access
1764Operations</a></div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1765<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1766<p>A key design point of an SSA-based representation is how it
1767represents memory. In LLVM, no memory locations are in SSA form, which
1768makes things very simple. This section describes how to read, write,
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1769allocate, and free memory in LLVM.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1770</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1771<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1772<div class="doc_subsubsection"> <a name="i_malloc">'<tt>malloc</tt>'
1773Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1774<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1775<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1776<pre> &lt;result&gt; = malloc &lt;type&gt;, uint &lt;NumElements&gt; <i>; yields {type*}:result</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1777 &lt;result&gt; = malloc &lt;type&gt; <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1778</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1779<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1780<p>The '<tt>malloc</tt>' instruction allocates memory from the system
1781heap and returns a pointer to it.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1782<h5>Arguments:</h5>
John Criswell6e4ca612004-02-24 16:13:56 +0000[diff] [blame]1783<p>The '<tt>malloc</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
1784bytes of memory from the operating system and returns a pointer of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1785appropriate type to the program. The second form of the instruction is
1786a shorter version of the first instruction that defaults to allocating
1787one element.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1788<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1789<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1790<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
1791a pointer is returned.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1792<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1793<pre> %array = malloc [4 x ubyte ] <i>; yields {[%4 x ubyte]*}:array</i>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1794
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1795 %size = <a
1796 href="#i_add">add</a> uint 2, 2 <i>; yields {uint}:size = uint 4</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1797 %array1 = malloc ubyte, uint 4 <i>; yields {ubyte*}:array1</i>
1798 %array2 = malloc [12 x ubyte], uint %size <i>; yields {[12 x ubyte]*}:array2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1799</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1800</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1801<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1802<div class="doc_subsubsection"> <a name="i_free">'<tt>free</tt>'
1803Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1804<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1805<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1806<pre> free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1807</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1808<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1809<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1810memory heap to be reallocated in the future.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1811<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1812<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1813<p>'<tt>value</tt>' shall be a pointer value that points to a value
1814that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
1815instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1816<h5>Semantics:</h5>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1817<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1818after this instruction executes.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1819<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1820<pre> %array = <a href="#i_malloc">malloc</a> [4 x ubyte] <i>; yields {[4 x ubyte]*}:array</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1821 free [4 x ubyte]* %array
1822</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1823</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1824<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1825<div class="doc_subsubsection"> <a name="i_alloca">'<tt>alloca</tt>'
1826Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1827<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1828<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1829<pre> &lt;result&gt; = alloca &lt;type&gt;, uint &lt;NumElements&gt; <i>; yields {type*}:result</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1830 &lt;result&gt; = alloca &lt;type&gt; <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1831</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1832<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1833<p>The '<tt>alloca</tt>' instruction allocates memory on the current
1834stack frame of the procedure that is live until the current function
1835returns to its caller.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1836<h5>Arguments:</h5>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1837<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1838bytes of memory on the runtime stack, returning a pointer of the
1839appropriate type to the program. The second form of the instruction is
1840a shorter version of the first that defaults to allocating one element.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1841<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1842<h5>Semantics:</h5>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1843<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1844memory is automatically released when the function returns. The '<tt>alloca</tt>'
1845instruction is commonly used to represent automatic variables that must
1846have an address available. When the function returns (either with the <tt><a
John Criswelldae2e932005-05-12 16:55:34 +0000[diff] [blame]1847 href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1848instructions), the memory is reclaimed.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1849<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1850<pre> %ptr = alloca int <i>; yields {int*}:ptr</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1851 %ptr = alloca int, uint 4 <i>; yields {int*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1852</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1853</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1854<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1855<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
1856Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1857<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1858<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1859<pre> &lt;result&gt; = load &lt;ty&gt;* &lt;pointer&gt;<br> &lt;result&gt; = volatile load &lt;ty&gt;* &lt;pointer&gt;<br></pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1860<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1861<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1862<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1863<p>The argument to the '<tt>load</tt>' instruction specifies the memory
1864address to load from. The pointer must point to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1865 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1866marked as <tt>volatile</tt> then the optimizer is not allowed to modify
1867the number or order of execution of this <tt>load</tt> with other
1868volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
1869instructions. </p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1870<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1871<p>The location of memory pointed to is loaded.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1872<h5>Examples:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1873<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1874 <a
1875 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1876 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1877</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1878</div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1879<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1880<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
1881Instruction</a> </div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1882<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1883<pre> store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt; <i>; yields {void}</i>
Chris Lattnerf0651072003-09-08 18:27:49 +0000[diff] [blame]1884 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt; <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1885</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1886<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1887<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1888<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1889<p>There are two arguments to the '<tt>store</tt>' instruction: a value
1890to store and an address to store it into. The type of the '<tt>&lt;pointer&gt;</tt>'
1891operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1892operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1893optimizer is not allowed to modify the number or order of execution of
1894this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
1895 href="#i_store">store</a></tt> instructions.</p>
1896<h5>Semantics:</h5>
1897<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
1898at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1899<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1900<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1901 <a
1902 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1903 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1904</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1905<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1906<div class="doc_subsubsection">
1907 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
1908</div>
1909
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1910<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1911<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1912<pre>
1913 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
1914</pre>
1915
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1916<h5>Overview:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1917
1918<p>
1919The '<tt>getelementptr</tt>' instruction is used to get the address of a
1920subelement of an aggregate data structure.</p>
1921
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1922<h5>Arguments:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1923
1924<p>This instruction takes a list of integer constants that indicate what
1925elements of the aggregate object to index to. The actual types of the arguments
1926provided depend on the type of the first pointer argument. The
1927'<tt>getelementptr</tt>' instruction is used to index down through the type
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]1928levels of a structure or to a specific index in an array. When indexing into a
1929structure, only <tt>uint</tt>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1930integer constants are allowed. When indexing into an array or pointer,
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1931<tt>int</tt> and <tt>long</tt> indexes are allowed of any sign.</p>
1932
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1933<p>For example, let's consider a C code fragment and how it gets
1934compiled to LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1935
1936<pre>
1937 struct RT {
1938 char A;
1939 int B[10][20];
1940 char C;
1941 };
1942 struct ST {
1943 int X;
1944 double Y;
1945 struct RT Z;
1946 };
1947
1948 int *foo(struct ST *s) {
1949 return &amp;s[1].Z.B[5][13];
1950 }
1951</pre>
1952
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1953<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1954
1955<pre>
1956 %RT = type { sbyte, [10 x [20 x int]], sbyte }
1957 %ST = type { int, double, %RT }
1958
Brian Gaeke7283e7c2004-07-02 21:08:14 +0000[diff] [blame]1959 implementation
1960
1961 int* %foo(%ST* %s) {
1962 entry:
1963 %reg = getelementptr %ST* %s, int 1, uint 2, uint 1, int 5, int 13
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1964 ret int* %reg
1965 }
1966</pre>
1967
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1968<h5>Semantics:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1969
1970<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1971on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1972and <a href="#t_array">array</a> types require <tt>uint</tt>, <tt>int</tt>,
1973<tt>ulong</tt>, or <tt>long</tt> values, and <a href="#t_struct">structure</a>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1974types require <tt>uint</tt> <b>constants</b>.</p>
1975
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1976<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1977type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ int, double, %RT
1978}</tt>' type, a structure. The second index indexes into the third element of
1979the structure, yielding a '<tt>%RT</tt>' = '<tt>{ sbyte, [10 x [20 x int]],
1980sbyte }</tt>' type, another structure. The third index indexes into the second
1981element of the structure, yielding a '<tt>[10 x [20 x int]]</tt>' type, an
1982array. The two dimensions of the array are subscripted into, yielding an
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]1983'<tt>int</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1984to this element, thus computing a value of '<tt>int*</tt>' type.</p>
1985
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1986<p>Note that it is perfectly legal to index partially through a
1987structure, returning a pointer to an inner element. Because of this,
1988the LLVM code for the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1989
1990<pre>
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]1991 int* %foo(%ST* %s) {
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1992 %t1 = getelementptr %ST* %s, int 1 <i>; yields %ST*:%t1</i>
1993 %t2 = getelementptr %ST* %t1, int 0, uint 2 <i>; yields %RT*:%t2</i>
1994 %t3 = getelementptr %RT* %t2, int 0, uint 1 <i>; yields [10 x [20 x int]]*:%t3</i>
1995 %t4 = getelementptr [10 x [20 x int]]* %t3, int 0, int 5 <i>; yields [20 x int]*:%t4</i>
1996 %t5 = getelementptr [20 x int]* %t4, int 0, int 13 <i>; yields int*:%t5</i>
1997 ret int* %t5
1998 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1999</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]2000
2001<p>Note that it is undefined to access an array out of bounds: array and
2002pointer indexes must always be within the defined bounds of the array type.
2003The one exception for this rules is zero length arrays. These arrays are
2004defined to be accessible as variable length arrays, which requires access
2005beyond the zero'th element.</p>
2006
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2007<h5>Example:</h5>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]2008
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2009<pre>
2010 <i>; yields [12 x ubyte]*:aptr</i>
2011 %aptr = getelementptr {int, [12 x ubyte]}* %sptr, long 0, uint 1
2012</pre>
2013
2014</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2015<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2016<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2017<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]2018<p>The instructions in this category are the "miscellaneous"
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2019instructions, which defy better classification.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2020</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2021<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2022<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
2023Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2024<div class="doc_text">
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2025<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2026<pre> &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...<br></pre>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2027<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2028<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
2029the SSA graph representing the function.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2030<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2031<p>The type of the incoming values are specified with the first type
2032field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
2033as arguments, with one pair for each predecessor basic block of the
2034current block. Only values of <a href="#t_firstclass">first class</a>
2035type may be used as the value arguments to the PHI node. Only labels
2036may be used as the label arguments.</p>
2037<p>There must be no non-phi instructions between the start of a basic
2038block and the PHI instructions: i.e. PHI instructions must be first in
2039a basic block.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2040<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2041<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the
2042value specified by the parameter, depending on which basic block we
2043came from in the last <a href="#terminators">terminator</a> instruction.</p>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2044<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2045<pre>Loop: ; Infinite loop that counts from 0 on up...<br> %indvar = phi uint [ 0, %LoopHeader ], [ %nextindvar, %Loop ]<br> %nextindvar = add uint %indvar, 1<br> br label %Loop<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2046</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2047
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2048<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2049<div class="doc_subsubsection">
2050 <a name="i_cast">'<tt>cast .. to</tt>' Instruction</a>
2051</div>
2052
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2053<div class="doc_text">
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2054
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2055<h5>Syntax:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2056
2057<pre>
2058 &lt;result&gt; = cast &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2059</pre>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2060
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2061<h5>Overview:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2062
2063<p>
2064The '<tt>cast</tt>' instruction is used as the primitive means to convert
2065integers to floating point, change data type sizes, and break type safety (by
2066casting pointers).
2067</p>
2068
2069
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2070<h5>Arguments:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2071
2072<p>
2073The '<tt>cast</tt>' instruction takes a value to cast, which must be a first
2074class value, and a type to cast it to, which must also be a <a
2075href="#t_firstclass">first class</a> type.
2076</p>
2077
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2078<h5>Semantics:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2079
2080<p>
2081This instruction follows the C rules for explicit casts when determining how the
2082data being cast must change to fit in its new container.
2083</p>
2084
2085<p>
2086When casting to bool, any value that would be considered true in the context of
2087a C '<tt>if</tt>' condition is converted to the boolean '<tt>true</tt>' values,
2088all else are '<tt>false</tt>'.
2089</p>
2090
2091<p>
2092When extending an integral value from a type of one signness to another (for
2093example '<tt>sbyte</tt>' to '<tt>ulong</tt>'), the value is sign-extended if the
2094<b>source</b> value is signed, and zero-extended if the source value is
2095unsigned. <tt>bool</tt> values are always zero extended into either zero or
2096one.
2097</p>
2098
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2099<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2100
2101<pre>
2102 %X = cast int 257 to ubyte <i>; yields ubyte:1</i>
Chris Lattner7bae3952002-06-25 18:03:17 +0000[diff] [blame]2103 %Y = cast int 123 to bool <i>; yields bool:true</i>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2104</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2105</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2106
2107<!-- _______________________________________________________________________ -->
2108<div class="doc_subsubsection">
2109 <a name="i_select">'<tt>select</tt>' Instruction</a>
2110</div>
2111
2112<div class="doc_text">
2113
2114<h5>Syntax:</h5>
2115
2116<pre>
2117 &lt;result&gt; = select bool &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
2118</pre>
2119
2120<h5>Overview:</h5>
2121
2122<p>
2123The '<tt>select</tt>' instruction is used to choose one value based on a
2124condition, without branching.
2125</p>
2126
2127
2128<h5>Arguments:</h5>
2129
2130<p>
2131The '<tt>select</tt>' instruction requires a boolean value indicating the condition, and two values of the same <a href="#t_firstclass">first class</a> type.
2132</p>
2133
2134<h5>Semantics:</h5>
2135
2136<p>
2137If the boolean condition evaluates to true, the instruction returns the first
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2138value argument; otherwise, it returns the second value argument.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2139</p>
2140
2141<h5>Example:</h5>
2142
2143<pre>
2144 %X = select bool true, ubyte 17, ubyte 42 <i>; yields ubyte:17</i>
2145</pre>
2146</div>
2147
2148
2149
2150
2151
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2152<!-- _______________________________________________________________________ -->
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2153<div class="doc_subsubsection">
2154 <a name="i_call">'<tt>call</tt>' Instruction</a>
2155</div>
2156
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2157<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2158
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2159<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2160<pre>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2161 &lt;result&gt; = [tail] call [<a href="#callingconv">cconv</a>] &lt;ty&gt;* &lt;fnptrval&gt;(&lt;param list&gt;)
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2162</pre>
2163
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2164<h5>Overview:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2165
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2166<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2167
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2168<h5>Arguments:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2169
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2170<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2171
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2172<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2173 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2174 <p>The optional "tail" marker indicates whether the callee function accesses
2175 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2176 function call is eligible for tail call optimization. Note that calls may
2177 be marked "tail" even if they do not occur before a <a
2178 href="#i_ret"><tt>ret</tt></a> instruction.
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2179 </li>
2180 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2181 <p>The optional "cconv" marker indicates which <a href="callingconv">calling
2182 convention</a> the call should use. If none is specified, the call defaults
2183 to using C calling conventions.
2184 </li>
2185 <li>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2186 <p>'<tt>ty</tt>': shall be the signature of the pointer to function value
2187 being invoked. The argument types must match the types implied by this
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2188 signature. This type can be omitted if the function is not varargs and
2189 if the function type does not return a pointer to a function.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2190 </li>
2191 <li>
2192 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
2193 be invoked. In most cases, this is a direct function invocation, but
2194 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2195 to function value.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2196 </li>
2197 <li>
2198 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencera7e302a2005-05-01 22:22:57 +0000[diff] [blame]2199 function signature argument types. All arguments must be of
2200 <a href="#t_firstclass">first class</a> type. If the function signature
2201 indicates the function accepts a variable number of arguments, the extra
2202 arguments can be specified.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2203 </li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2204</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2205
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2206<h5>Semantics:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2207
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2208<p>The '<tt>call</tt>' instruction is used to cause control flow to
2209transfer to a specified function, with its incoming arguments bound to
2210the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
2211instruction in the called function, control flow continues with the
2212instruction after the function call, and the return value of the
2213function is bound to the result argument. This is a simpler case of
2214the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2215
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2216<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2217
2218<pre>
2219 %retval = call int %test(int %argc)
2220 call int(sbyte*, ...) *%printf(sbyte* %msg, int 12, sbyte 42);
2221 %X = tail call int %foo()
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2222 %Y = tail call <a href="#callingconv">fastcc</a> int %foo()
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2223</pre>
2224
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2225</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2226
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2227<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2228<div class="doc_subsubsection">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2229 <a name="i_vaarg">'<tt>vaarg</tt>' Instruction</a>
2230</div>
2231
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2232<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2233
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2234<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2235
2236<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2237 &lt;resultval&gt; = va_arg &lt;va_list*&gt; &lt;arglist&gt;, &lt;argty&gt;
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2238</pre>
2239
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2240<h5>Overview:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2241
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2242<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2243the "variable argument" area of a function call. It is used to implement the
2244<tt>va_arg</tt> macro in C.</p>
2245
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2246<h5>Arguments:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2247
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2248<p>This instruction takes a <tt>va_list*</tt> value and the type of
2249the argument. It returns a value of the specified argument type and
2250increments the <tt>va_list</tt> to poin to the next argument. Again, the
2251actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2252
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2253<h5>Semantics:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2254
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2255<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
2256type from the specified <tt>va_list</tt> and causes the
2257<tt>va_list</tt> to point to the next argument. For more information,
2258see the variable argument handling <a href="#int_varargs">Intrinsic
2259Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2260
2261<p>It is legal for this instruction to be called in a function which does not
2262take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2263function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2264
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2265<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2266href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2267argument.</p>
2268
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2269<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2270
2271<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
2272
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2273</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2274
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2275<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2276<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
2277<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2278
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2279<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2280
2281<p>LLVM supports the notion of an "intrinsic function". These functions have
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2282well known names and semantics and are required to follow certain
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2283restrictions. Overall, these instructions represent an extension mechanism for
2284the LLVM language that does not require changing all of the transformations in
2285LLVM to add to the language (or the bytecode reader/writer, the parser,
2286etc...).</p>
2287
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2288<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
2289prefix is reserved in LLVM for intrinsic names; thus, functions may not be named
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2290this. Intrinsic functions must always be external functions: you cannot define
2291the body of intrinsic functions. Intrinsic functions may only be used in call
2292or invoke instructions: it is illegal to take the address of an intrinsic
2293function. Additionally, because intrinsic functions are part of the LLVM
2294language, it is required that they all be documented here if any are added.</p>
2295
2296
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2297<p>To learn how to add an intrinsic function, please see the <a
Chris Lattner590cff32005-05-11 03:35:57 +0000[diff] [blame]2298href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2299</p>
2300
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2301</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2302
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2303<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2304<div class="doc_subsection">
2305 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
2306</div>
2307
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2308<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2309
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2310<p>Variable argument support is defined in LLVM with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2311 href="#i_vanext"><tt>vanext</tt></a> instruction and these three
2312intrinsic functions. These functions are related to the similarly
2313named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2314
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2315<p>All of these functions operate on arguments that use a
2316target-specific value type "<tt>va_list</tt>". The LLVM assembly
2317language reference manual does not define what this type is, so all
2318transformations should be prepared to handle intrinsics with any type
2319used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2320
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2321<p>This example shows how the <a href="#i_vanext"><tt>vanext</tt></a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2322instruction and the variable argument handling intrinsic functions are
2323used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2324
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2325<pre>
2326int %test(int %X, ...) {
2327 ; Initialize variable argument processing
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2328 %ap = alloca sbyte*
2329 call void %<a href="#i_va_start">llvm.va_start</a>(sbyte** %ap)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2330
2331 ; Read a single integer argument
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2332 %tmp = va_arg sbyte** %ap, int
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2333
2334 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2335 %aq = alloca sbyte*
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2336 call void %<a href="#i_va_copy">llvm.va_copy</a>(sbyte** %aq, sbyte** %ap)
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2337 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte** %aq)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2338
2339 ; Stop processing of arguments.
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2340 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte** %ap)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2341 ret int %tmp
2342}
2343</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2344</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2345
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2346<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2347<div class="doc_subsubsection">
2348 <a name="i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
2349</div>
2350
2351
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2352<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2353<h5>Syntax:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2354<pre> declare void %llvm.va_start(&lt;va_list&gt;* &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2355<h5>Overview:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2356<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
2357<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
2358href="#i_va_arg">va_arg</a></tt>.</p>
2359
2360<h5>Arguments:</h5>
2361
2362<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
2363
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2364<h5>Semantics:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2365
2366<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
2367macro available in C. In a target-dependent way, it initializes the
2368<tt>va_list</tt> element the argument points to, so that the next call to
2369<tt>va_arg</tt> will produce the first variable argument passed to the function.
2370Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
2371last argument of the function, the compiler can figure that out.</p>
2372
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2373</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2374
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2375<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2376<div class="doc_subsubsection">
2377 <a name="i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
2378</div>
2379
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2380<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2381<h5>Syntax:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2382<pre> declare void %llvm.va_end(&lt;va_list*&gt; &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2383<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2384<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>&lt;arglist&gt;</tt>
2385which has been initialized previously with <tt><a href="#i_va_start">llvm.va_start</a></tt>
2386or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2387<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2388<p>The argument is a <tt>va_list</tt> to destroy.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2389<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2390<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2391macro available in C. In a target-dependent way, it destroys the <tt>va_list</tt>.
2392Calls to <a href="#i_va_start"><tt>llvm.va_start</tt></a> and <a
2393 href="#i_va_copy"><tt>llvm.va_copy</tt></a> must be matched exactly
2394with calls to <tt>llvm.va_end</tt>.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2395</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2396
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2397<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2398<div class="doc_subsubsection">
2399 <a name="i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
2400</div>
2401
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2402<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2403
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2404<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2405
2406<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2407 declare void %llvm.va_copy(&lt;va_list&gt;* &lt;destarglist&gt;,
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2408 &lt;va_list&gt;* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2409</pre>
2410
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2411<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2412
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2413<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position from
2414the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2415
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2416<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2417
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2418<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2419The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2420
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2421
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2422<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2423
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2424<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt> macro
2425available in C. In a target-dependent way, it copies the source
2426<tt>va_list</tt> element into the destination list. This intrinsic is necessary
2427because the <tt><a href="i_va_begin">llvm.va_begin</a></tt> intrinsic may be
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2428arbitrarily complex and require memory allocation, for example.</p>
2429
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2430</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2431
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2432<!-- ======================================================================= -->
2433<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2434 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
2435</div>
2436
2437<div class="doc_text">
2438
2439<p>
2440LLVM support for <a href="GarbageCollection.html">Accurate Garbage
2441Collection</a> requires the implementation and generation of these intrinsics.
2442These intrinsics allow identification of <a href="#i_gcroot">GC roots on the
2443stack</a>, as well as garbage collector implementations that require <a
2444href="#i_gcread">read</a> and <a href="#i_gcwrite">write</a> barriers.
2445Front-ends for type-safe garbage collected languages should generate these
2446intrinsics to make use of the LLVM garbage collectors. For more details, see <a
2447href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
2448</p>
2449</div>
2450
2451<!-- _______________________________________________________________________ -->
2452<div class="doc_subsubsection">
2453 <a name="i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
2454</div>
2455
2456<div class="doc_text">
2457
2458<h5>Syntax:</h5>
2459
2460<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2461 declare void %llvm.gcroot(&lt;ty&gt;** %ptrloc, &lt;ty2&gt;* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2462</pre>
2463
2464<h5>Overview:</h5>
2465
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2466<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2467the code generator, and allows some metadata to be associated with it.</p>
2468
2469<h5>Arguments:</h5>
2470
2471<p>The first argument specifies the address of a stack object that contains the
2472root pointer. The second pointer (which must be either a constant or a global
2473value address) contains the meta-data to be associated with the root.</p>
2474
2475<h5>Semantics:</h5>
2476
2477<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
2478location. At compile-time, the code generator generates information to allow
2479the runtime to find the pointer at GC safe points.
2480</p>
2481
2482</div>
2483
2484
2485<!-- _______________________________________________________________________ -->
2486<div class="doc_subsubsection">
2487 <a name="i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
2488</div>
2489
2490<div class="doc_text">
2491
2492<h5>Syntax:</h5>
2493
2494<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2495 declare sbyte* %llvm.gcread(sbyte** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2496</pre>
2497
2498<h5>Overview:</h5>
2499
2500<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
2501locations, allowing garbage collector implementations that require read
2502barriers.</p>
2503
2504<h5>Arguments:</h5>
2505
2506<p>The argument is the address to read from, which should be an address
2507allocated from the garbage collector.</p>
2508
2509<h5>Semantics:</h5>
2510
2511<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
2512instruction, but may be replaced with substantially more complex code by the
2513garbage collector runtime, as needed.</p>
2514
2515</div>
2516
2517
2518<!-- _______________________________________________________________________ -->
2519<div class="doc_subsubsection">
2520 <a name="i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
2521</div>
2522
2523<div class="doc_text">
2524
2525<h5>Syntax:</h5>
2526
2527<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2528 declare void %llvm.gcwrite(sbyte* %P1, sbyte** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2529</pre>
2530
2531<h5>Overview:</h5>
2532
2533<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
2534locations, allowing garbage collector implementations that require write
2535barriers (such as generational or reference counting collectors).</p>
2536
2537<h5>Arguments:</h5>
2538
2539<p>The first argument is the reference to store, and the second is the heap
2540location to store to.</p>
2541
2542<h5>Semantics:</h5>
2543
2544<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
2545instruction, but may be replaced with substantially more complex code by the
2546garbage collector runtime, as needed.</p>
2547
2548</div>
2549
2550
2551
2552<!-- ======================================================================= -->
2553<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2554 <a name="int_codegen">Code Generator Intrinsics</a>
2555</div>
2556
2557<div class="doc_text">
2558<p>
2559These intrinsics are provided by LLVM to expose special features that may only
2560be implemented with code generator support.
2561</p>
2562
2563</div>
2564
2565<!-- _______________________________________________________________________ -->
2566<div class="doc_subsubsection">
2567 <a name="i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
2568</div>
2569
2570<div class="doc_text">
2571
2572<h5>Syntax:</h5>
2573<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2574 declare void* %llvm.returnaddress(uint &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2575</pre>
2576
2577<h5>Overview:</h5>
2578
2579<p>
2580The '<tt>llvm.returnaddress</tt>' intrinsic returns a target-specific value
2581indicating the return address of the current function or one of its callers.
2582</p>
2583
2584<h5>Arguments:</h5>
2585
2586<p>
2587The argument to this intrinsic indicates which function to return the address
2588for. Zero indicates the calling function, one indicates its caller, etc. The
2589argument is <b>required</b> to be a constant integer value.
2590</p>
2591
2592<h5>Semantics:</h5>
2593
2594<p>
2595The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
2596the return address of the specified call frame, or zero if it cannot be
2597identified. The value returned by this intrinsic is likely to be incorrect or 0
2598for arguments other than zero, so it should only be used for debugging purposes.
2599</p>
2600
2601<p>
2602Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]2603aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2604source-language caller.
2605</p>
2606</div>
2607
2608
2609<!-- _______________________________________________________________________ -->
2610<div class="doc_subsubsection">
2611 <a name="i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
2612</div>
2613
2614<div class="doc_text">
2615
2616<h5>Syntax:</h5>
2617<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2618 declare void* %llvm.frameaddress(uint &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2619</pre>
2620
2621<h5>Overview:</h5>
2622
2623<p>
2624The '<tt>llvm.frameaddress</tt>' intrinsic returns the target-specific frame
2625pointer value for the specified stack frame.
2626</p>
2627
2628<h5>Arguments:</h5>
2629
2630<p>
2631The argument to this intrinsic indicates which function to return the frame
2632pointer for. Zero indicates the calling function, one indicates its caller,
2633etc. The argument is <b>required</b> to be a constant integer value.
2634</p>
2635
2636<h5>Semantics:</h5>
2637
2638<p>
2639The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
2640the frame address of the specified call frame, or zero if it cannot be
2641identified. The value returned by this intrinsic is likely to be incorrect or 0
2642for arguments other than zero, so it should only be used for debugging purposes.
2643</p>
2644
2645<p>
2646Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]2647aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2648source-language caller.
2649</p>
2650</div>
2651
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2652<!-- _______________________________________________________________________ -->
2653<div class="doc_subsubsection">
2654 <a name="i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
2655</div>
2656
2657<div class="doc_text">
2658
2659<h5>Syntax:</h5>
2660<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2661 declare void %llvm.prefetch(sbyte * &lt;address&gt;,
2662 uint &lt;rw&gt;, uint &lt;locality&gt;)
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2663</pre>
2664
2665<h5>Overview:</h5>
2666
2667
2668<p>
2669The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2670a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
2671no
2672effect on the behavior of the program but can change its performance
Chris Lattner2a615362005-02-28 19:47:14 +0000[diff] [blame]2673characteristics.
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2674</p>
2675
2676<h5>Arguments:</h5>
2677
2678<p>
2679<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
2680determining if the fetch should be for a read (0) or write (1), and
2681<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattneraeffb4a2005-03-07 20:31:38 +0000[diff] [blame]2682locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2683<tt>locality</tt> arguments must be constant integers.
2684</p>
2685
2686<h5>Semantics:</h5>
2687
2688<p>
2689This intrinsic does not modify the behavior of the program. In particular,
2690prefetches cannot trap and do not produce a value. On targets that support this
2691intrinsic, the prefetch can provide hints to the processor cache for better
2692performance.
2693</p>
2694
2695</div>
2696
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2697<!-- _______________________________________________________________________ -->
2698<div class="doc_subsubsection">
2699 <a name="i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
2700</div>
2701
2702<div class="doc_text">
2703
2704<h5>Syntax:</h5>
2705<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2706 declare void %llvm.pcmarker( uint &lt;id&gt; )
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2707</pre>
2708
2709<h5>Overview:</h5>
2710
2711
2712<p>
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2713The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
2714(PC) in a region of
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2715code to simulators and other tools. The method is target specific, but it is
2716expected that the marker will use exported symbols to transmit the PC of the marker.
2717The marker makes no guaranties that it will remain with any specific instruction
2718after optimizations. It is possible that the presense of a marker will inhibit
2719optimizations. The intended use is to be inserted after optmizations to allow
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2720correlations of simulation runs.
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2721</p>
2722
2723<h5>Arguments:</h5>
2724
2725<p>
2726<tt>id</tt> is a numerical id identifying the marker.
2727</p>
2728
2729<h5>Semantics:</h5>
2730
2731<p>
2732This intrinsic does not modify the behavior of the program. Backends that do not
2733support this intrinisic may ignore it.
2734</p>
2735
2736</div>
2737
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2738
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2739<!-- ======================================================================= -->
2740<div class="doc_subsection">
2741 <a name="int_os">Operating System Intrinsics</a>
2742</div>
2743
2744<div class="doc_text">
2745<p>
2746These intrinsics are provided by LLVM to support the implementation of
2747operating system level code.
2748</p>
2749
2750</div>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2751
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2752<!-- _______________________________________________________________________ -->
2753<div class="doc_subsubsection">
2754 <a name="i_readport">'<tt>llvm.readport</tt>' Intrinsic</a>
2755</div>
2756
2757<div class="doc_text">
2758
2759<h5>Syntax:</h5>
2760<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2761 declare &lt;integer type&gt; %llvm.readport (&lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2762</pre>
2763
2764<h5>Overview:</h5>
2765
2766<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2767The '<tt>llvm.readport</tt>' intrinsic reads data from the specified hardware
2768I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2769</p>
2770
2771<h5>Arguments:</h5>
2772
2773<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2774The argument to this intrinsic indicates the hardware I/O address from which
2775to read the data. The address is in the hardware I/O address namespace (as
2776opposed to being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2777</p>
2778
2779<h5>Semantics:</h5>
2780
2781<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2782The '<tt>llvm.readport</tt>' intrinsic reads data from the hardware I/O port
2783specified by <i>address</i> and returns the value. The address and return
2784value must be integers, but the size is dependent upon the platform upon which
2785the program is code generated. For example, on x86, the address must be an
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2786unsigned 16-bit value, and the return value must be 8, 16, or 32 bits.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2787</p>
2788
2789</div>
2790
2791<!-- _______________________________________________________________________ -->
2792<div class="doc_subsubsection">
2793 <a name="i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a>
2794</div>
2795
2796<div class="doc_text">
2797
2798<h5>Syntax:</h5>
2799<pre>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2800 call void (&lt;integer type&gt;, &lt;integer type&gt;)*
2801 %llvm.writeport (&lt;integer type&gt; &lt;value&gt;,
2802 &lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2803</pre>
2804
2805<h5>Overview:</h5>
2806
2807<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2808The '<tt>llvm.writeport</tt>' intrinsic writes data to the specified hardware
2809I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2810</p>
2811
2812<h5>Arguments:</h5>
2813
2814<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2815The first argument is the value to write to the I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2816</p>
2817
2818<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2819The second argument indicates the hardware I/O address to which data should be
2820written. The address is in the hardware I/O address namespace (as opposed to
2821being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2822</p>
2823
2824<h5>Semantics:</h5>
2825
2826<p>
2827The '<tt>llvm.writeport</tt>' intrinsic writes <i>value</i> to the I/O port
2828specified by <i>address</i>. The address and value must be integers, but the
2829size is dependent upon the platform upon which the program is code generated.
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2830For example, on x86, the address must be an unsigned 16-bit value, and the
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2831value written must be 8, 16, or 32 bits in length.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2832</p>
2833
2834</div>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2835
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2836<!-- _______________________________________________________________________ -->
2837<div class="doc_subsubsection">
2838 <a name="i_readio">'<tt>llvm.readio</tt>' Intrinsic</a>
2839</div>
2840
2841<div class="doc_text">
2842
2843<h5>Syntax:</h5>
2844<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2845 declare &lt;result&gt; %llvm.readio (&lt;ty&gt; * &lt;pointer&gt;)
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2846</pre>
2847
2848<h5>Overview:</h5>
2849
2850<p>
2851The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
2852address.
2853</p>
2854
2855<h5>Arguments:</h5>
2856
2857<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2858The argument to this intrinsic is a pointer indicating the memory address from
2859which to read the data. The data must be a
2860<a href="#t_firstclass">first class</a> type.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2861</p>
2862
2863<h5>Semantics:</h5>
2864
2865<p>
2866The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2867location specified by <i>pointer</i> and returns the value. The argument must
2868be a pointer, and the return value must be a
2869<a href="#t_firstclass">first class</a> type. However, certain architectures
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2870may not support I/O on all first class types. For example, 32-bit processors
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2871may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2872</p>
2873
2874<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2875This intrinsic enforces an in-order memory model for llvm.readio and
2876llvm.writeio calls on machines that use dynamic scheduling. Dynamically
2877scheduled processors may execute loads and stores out of order, re-ordering at
2878run time accesses to memory mapped I/O registers. Using these intrinsics
2879ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2880</p>
2881
2882</div>
2883
2884<!-- _______________________________________________________________________ -->
2885<div class="doc_subsubsection">
2886 <a name="i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a>
2887</div>
2888
2889<div class="doc_text">
2890
2891<h5>Syntax:</h5>
2892<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2893 declare void %llvm.writeio (&lt;ty1&gt; &lt;value&gt;, &lt;ty2&gt; * &lt;pointer&gt;)
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2894</pre>
2895
2896<h5>Overview:</h5>
2897
2898<p>
2899The '<tt>llvm.writeio</tt>' intrinsic writes data to the specified memory
2900mapped I/O address.
2901</p>
2902
2903<h5>Arguments:</h5>
2904
2905<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2906The first argument is the value to write to the memory mapped I/O location.
2907The second argument is a pointer indicating the memory address to which the
2908data should be written.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2909</p>
2910
2911<h5>Semantics:</h5>
2912
2913<p>
2914The '<tt>llvm.writeio</tt>' intrinsic writes <i>value</i> to the memory mapped
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2915I/O address specified by <i>pointer</i>. The value must be a
2916<a href="#t_firstclass">first class</a> type. However, certain architectures
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2917may not support I/O on all first class types. For example, 32-bit processors
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2918may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2919</p>
2920
2921<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2922This intrinsic enforces an in-order memory model for llvm.readio and
2923llvm.writeio calls on machines that use dynamic scheduling. Dynamically
2924scheduled processors may execute loads and stores out of order, re-ordering at
2925run time accesses to memory mapped I/O registers. Using these intrinsics
2926ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2927</p>
2928
2929</div>
2930
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2931<!-- ======================================================================= -->
2932<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2933 <a name="int_libc">Standard C Library Intrinsics</a>
2934</div>
2935
2936<div class="doc_text">
2937<p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2938LLVM provides intrinsics for a few important standard C library functions.
2939These intrinsics allow source-language front-ends to pass information about the
2940alignment of the pointer arguments to the code generator, providing opportunity
2941for more efficient code generation.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2942</p>
2943
2944</div>
2945
2946<!-- _______________________________________________________________________ -->
2947<div class="doc_subsubsection">
2948 <a name="i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
2949</div>
2950
2951<div class="doc_text">
2952
2953<h5>Syntax:</h5>
2954<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]2955 declare void %llvm.memcpy(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
2956 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2957</pre>
2958
2959<h5>Overview:</h5>
2960
2961<p>
2962The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
2963location to the destination location.
2964</p>
2965
2966<p>
2967Note that, unlike the standard libc function, the <tt>llvm.memcpy</tt> intrinsic
2968does not return a value, and takes an extra alignment argument.
2969</p>
2970
2971<h5>Arguments:</h5>
2972
2973<p>
2974The first argument is a pointer to the destination, the second is a pointer to
2975the source. The third argument is an (arbitrarily sized) integer argument
2976specifying the number of bytes to copy, and the fourth argument is the alignment
2977of the source and destination locations.
2978</p>
2979
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]2980<p>
2981If the call to this intrinisic has an alignment value that is not 0 or 1, then
2982the caller guarantees that the size of the copy is a multiple of the alignment
2983and that both the source and destination pointers are aligned to that boundary.
2984</p>
2985
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2986<h5>Semantics:</h5>
2987
2988<p>
2989The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
2990location to the destination location, which are not allowed to overlap. It
2991copies "len" bytes of memory over. If the argument is known to be aligned to
2992some boundary, this can be specified as the fourth argument, otherwise it should
2993be set to 0 or 1.
2994</p>
2995</div>
2996
2997
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]2998<!-- _______________________________________________________________________ -->
2999<div class="doc_subsubsection">
3000 <a name="i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
3001</div>
3002
3003<div class="doc_text">
3004
3005<h5>Syntax:</h5>
3006<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3007 declare void %llvm.memmove(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
3008 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3009</pre>
3010
3011<h5>Overview:</h5>
3012
3013<p>
3014The '<tt>llvm.memmove</tt>' intrinsic moves a block of memory from the source
3015location to the destination location. It is similar to the '<tt>llvm.memcpy</tt>'
3016intrinsic but allows the two memory locations to overlap.
3017</p>
3018
3019<p>
3020Note that, unlike the standard libc function, the <tt>llvm.memmove</tt> intrinsic
3021does not return a value, and takes an extra alignment argument.
3022</p>
3023
3024<h5>Arguments:</h5>
3025
3026<p>
3027The first argument is a pointer to the destination, the second is a pointer to
3028the source. The third argument is an (arbitrarily sized) integer argument
3029specifying the number of bytes to copy, and the fourth argument is the alignment
3030of the source and destination locations.
3031</p>
3032
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]3033<p>
3034If the call to this intrinisic has an alignment value that is not 0 or 1, then
3035the caller guarantees that the size of the copy is a multiple of the alignment
3036and that both the source and destination pointers are aligned to that boundary.
3037</p>
3038
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3039<h5>Semantics:</h5>
3040
3041<p>
3042The '<tt>llvm.memmove</tt>' intrinsic copies a block of memory from the source
3043location to the destination location, which may overlap. It
3044copies "len" bytes of memory over. If the argument is known to be aligned to
3045some boundary, this can be specified as the fourth argument, otherwise it should
3046be set to 0 or 1.
3047</p>
3048</div>
3049
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3050
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3051<!-- _______________________________________________________________________ -->
3052<div class="doc_subsubsection">
3053 <a name="i_memset">'<tt>llvm.memset</tt>' Intrinsic</a>
3054</div>
3055
3056<div class="doc_text">
3057
3058<h5>Syntax:</h5>
3059<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3060 declare void %llvm.memset(sbyte* &lt;dest&gt;, ubyte &lt;val&gt;,
3061 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3062</pre>
3063
3064<h5>Overview:</h5>
3065
3066<p>
3067The '<tt>llvm.memset</tt>' intrinsic fills a block of memory with a particular
3068byte value.
3069</p>
3070
3071<p>
3072Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
3073does not return a value, and takes an extra alignment argument.
3074</p>
3075
3076<h5>Arguments:</h5>
3077
3078<p>
3079The first argument is a pointer to the destination to fill, the second is the
3080byte value to fill it with, the third argument is an (arbitrarily sized) integer
3081argument specifying the number of bytes to fill, and the fourth argument is the
3082known alignment of destination location.
3083</p>
3084
3085<p>
3086If the call to this intrinisic has an alignment value that is not 0 or 1, then
3087the caller guarantees that the size of the copy is a multiple of the alignment
3088and that the destination pointer is aligned to that boundary.
3089</p>
3090
3091<h5>Semantics:</h5>
3092
3093<p>
3094The '<tt>llvm.memset</tt>' intrinsic fills "len" bytes of memory starting at the
3095destination location. If the argument is known to be aligned to some boundary,
3096this can be specified as the fourth argument, otherwise it should be set to 0 or
30971.
3098</p>
3099</div>
3100
3101
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]3102<!-- _______________________________________________________________________ -->
3103<div class="doc_subsubsection">
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]3104 <a name="i_isunordered">'<tt>llvm.isunordered</tt>' Intrinsic</a>
3105</div>
3106
3107<div class="doc_text">
3108
3109<h5>Syntax:</h5>
3110<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]3111 declare bool %llvm.isunordered(&lt;float or double&gt; Val1, &lt;float or double&gt; Val2)
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]3112</pre>
3113
3114<h5>Overview:</h5>
3115
3116<p>
3117The '<tt>llvm.isunordered</tt>' intrinsic returns true if either or both of the
3118specified floating point values is a NAN.
3119</p>
3120
3121<h5>Arguments:</h5>
3122
3123<p>
3124The arguments are floating point numbers of the same type.
3125</p>
3126
3127<h5>Semantics:</h5>
3128
3129<p>
3130If either or both of the arguments is a SNAN or QNAN, it returns true, otherwise
3131false.
3132</p>
3133</div>
3134
3135
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3136<!-- ======================================================================= -->
3137<div class="doc_subsection">
3138 <a name="int_count">Bit Counting Intrinsics</a>
3139</div>
3140
3141<div class="doc_text">
3142<p>
3143LLVM provides intrinsics for a few important bit counting operations.
3144These allow efficient code generation for some algorithms.
3145</p>
3146
3147</div>
3148
3149<!-- _______________________________________________________________________ -->
3150<div class="doc_subsubsection">
3151 <a name="int_ctpop">'<tt>llvm.ctpop</tt>' Intrinsic</a>
3152</div>
3153
3154<div class="doc_text">
3155
3156<h5>Syntax:</h5>
3157<pre>
3158 declare int %llvm.ctpop(int &lt;src&gt;)
3159
3160</pre>
3161
3162<h5>Overview:</h5>
3163
3164<p>
3165The '<tt>llvm.ctpop</tt>' intrinsic counts the number of ones in a variable.
3166</p>
3167
3168<h5>Arguments:</h5>
3169
3170<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3171The only argument is the value to be counted. The argument may be of any
3172integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3173</p>
3174
3175<h5>Semantics:</h5>
3176
3177<p>
3178The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
3179</p>
3180</div>
3181
3182<!-- _______________________________________________________________________ -->
3183<div class="doc_subsubsection">
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3184 <a name="int_ctlz">'<tt>llvm.ctlz</tt>' Intrinsic</a>
3185</div>
3186
3187<div class="doc_text">
3188
3189<h5>Syntax:</h5>
3190<pre>
3191 declare int %llvm.ctlz(int &lt;src&gt;)
3192
3193</pre>
3194
3195<h5>Overview:</h5>
3196
3197<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3198The '<tt>llvm.ctlz</tt>' intrinsic counts the number of leading zeros in a
3199variable.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3200</p>
3201
3202<h5>Arguments:</h5>
3203
3204<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3205The only argument is the value to be counted. The argument may be of any
3206integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3207</p>
3208
3209<h5>Semantics:</h5>
3210
3211<p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3212The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
3213in a variable. If the src == 0 then the result is the size in bits of the type
3214of src. For example, <tt>llvm.cttz(int 2) = 30</tt>.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3215</p>
3216</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]3217
3218
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3219
3220<!-- _______________________________________________________________________ -->
3221<div class="doc_subsubsection">
3222 <a name="int_cttz">'<tt>llvm.cttz</tt>' Intrinsic</a>
3223</div>
3224
3225<div class="doc_text">
3226
3227<h5>Syntax:</h5>
3228<pre>
3229 declare int %llvm.cttz(int &lt;src&gt;)
3230
3231</pre>
3232
3233<h5>Overview:</h5>
3234
3235<p>
3236The '<tt>llvm.cttz</tt>' intrinsic counts the number of trailing zeros.
3237</p>
3238
3239<h5>Arguments:</h5>
3240
3241<p>
3242The only argument is the value to be counted. The argument may be of any
3243integer type. The return type must match the argument type.
3244</p>
3245
3246<h5>Semantics:</h5>
3247
3248<p>
3249The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
3250in a variable. If the src == 0 then the result is the size in bits of the type
3251of src. For example, <tt>llvm.cttz(2) = 1</tt>.
3252</p>
3253</div>
3254
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3255<!-- ======================================================================= -->
3256<div class="doc_subsection">
3257 <a name="int_debugger">Debugger Intrinsics</a>
3258</div>
3259
3260<div class="doc_text">
3261<p>
3262The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
3263are described in the <a
3264href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
3265Debugging</a> document.
3266</p>
3267</div>
3268
3269
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3270<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3271<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]3272<address>
3273 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
3274 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
3275 <a href="http://validator.w3.org/check/referer"><img
3276 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
3277
3278 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
3279 <a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br>
3280 Last modified: $Date$
3281</address>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3282</body>
3283</html>