Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 0ec250ca8f3a3bdc710d61f1a78b715ca4f9112c / . / docs / LangRef.html
blob: 6b5bb962c6b0f767cc9b74e59295d25e4c28dc25 [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 Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]146 <li><a href="#i_sqrt">'<tt>llvm.sqrt</tt>' Intrinsic</a></li>
147
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]148 </ol>
149 </li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]150 <li><a href="#int_count">Bit counting Intrinsics</a>
151 <ol>
152 <li><a href="#int_ctpop">'<tt>llvm.ctpop</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]153 <li><a href="#int_ctlz">'<tt>llvm.ctlz</tt>' Intrinsic </a></li>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]154 <li><a href="#int_cttz">'<tt>llvm.cttz</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]155 </ol>
156 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]157 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]158 </ol>
159 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]160</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]161
162<div class="doc_author">
163 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
164 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]165</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]166
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]167<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]168<div class="doc_section"> <a name="abstract">Abstract </a></div>
169<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]170
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]171<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]172<p>This document is a reference manual for the LLVM assembly language.
173LLVM is an SSA based representation that provides type safety,
174low-level operations, flexibility, and the capability of representing
175'all' high-level languages cleanly. It is the common code
176representation used throughout all phases of the LLVM compilation
177strategy.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]178</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]179
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]180<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]181<div class="doc_section"> <a name="introduction">Introduction</a> </div>
182<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]183
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]184<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]185
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]186<p>The LLVM code representation is designed to be used in three
187different forms: as an in-memory compiler IR, as an on-disk bytecode
188representation (suitable for fast loading by a Just-In-Time compiler),
189and as a human readable assembly language representation. This allows
190LLVM to provide a powerful intermediate representation for efficient
191compiler transformations and analysis, while providing a natural means
192to debug and visualize the transformations. The three different forms
193of LLVM are all equivalent. This document describes the human readable
194representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]195
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]196<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]197while being expressive, typed, and extensible at the same time. It
198aims to be a "universal IR" of sorts, by being at a low enough level
199that high-level ideas may be cleanly mapped to it (similar to how
200microprocessors are "universal IR's", allowing many source languages to
201be mapped to them). By providing type information, LLVM can be used as
202the target of optimizations: for example, through pointer analysis, it
203can be proven that a C automatic variable is never accessed outside of
204the current function... allowing it to be promoted to a simple SSA
205value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]206
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]207</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]208
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]209<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]210<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]211
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]212<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]213
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]214<p>It is important to note that this document describes 'well formed'
215LLVM assembly language. There is a difference between what the parser
216accepts and what is considered 'well formed'. For example, the
217following instruction is syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]218
219<pre>
220 %x = <a href="#i_add">add</a> int 1, %x
221</pre>
222
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]223<p>...because the definition of <tt>%x</tt> does not dominate all of
224its uses. The LLVM infrastructure provides a verification pass that may
225be used to verify that an LLVM module is well formed. This pass is
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]226automatically run by the parser after parsing input assembly and by
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]227the optimizer before it outputs bytecode. The violations pointed out
228by the verifier pass indicate bugs in transformation passes or input to
229the parser.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]230
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]231<!-- Describe the typesetting conventions here. --> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]232
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]233<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]234<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]235<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]236
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]237<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]238
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]239<p>LLVM uses three different forms of identifiers, for different
240purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]241
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]242<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]243 <li>Named values are represented as a string of characters with a '%' prefix.
244 For example, %foo, %DivisionByZero, %a.really.long.identifier. The actual
245 regular expression used is '<tt>%[a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
246 Identifiers which require other characters in their names can be surrounded
247 with quotes. In this way, anything except a <tt>"</tt> character can be used
248 in a name.</li>
249
250 <li>Unnamed values are represented as an unsigned numeric value with a '%'
251 prefix. For example, %12, %2, %44.</li>
252
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]253 <li>Constants, which are described in a <a href="#constants">section about
254 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]255</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]256
257<p>LLVM requires that values start with a '%' sign for two reasons: Compilers
258don't need to worry about name clashes with reserved words, and the set of
259reserved words may be expanded in the future without penalty. Additionally,
260unnamed identifiers allow a compiler to quickly come up with a temporary
261variable without having to avoid symbol table conflicts.</p>
262
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]263<p>Reserved words in LLVM are very similar to reserved words in other
264languages. There are keywords for different opcodes ('<tt><a
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]265href="#i_add">add</a></tt>', '<tt><a href="#i_cast">cast</a></tt>', '<tt><a
266href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
267href="#t_void">void</a></tt>', '<tt><a href="#t_uint">uint</a></tt>', etc...),
268and others. These reserved words cannot conflict with variable names, because
269none of them start with a '%' character.</p>
270
271<p>Here is an example of LLVM code to multiply the integer variable
272'<tt>%X</tt>' by 8:</p>
273
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]274<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]275
276<pre>
277 %result = <a href="#i_mul">mul</a> uint %X, 8
278</pre>
279
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]280<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]281
282<pre>
283 %result = <a href="#i_shl">shl</a> uint %X, ubyte 3
284</pre>
285
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]286<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]287
288<pre>
289 <a href="#i_add">add</a> uint %X, %X <i>; yields {uint}:%0</i>
290 <a href="#i_add">add</a> uint %0, %0 <i>; yields {uint}:%1</i>
291 %result = <a href="#i_add">add</a> uint %1, %1
292</pre>
293
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]294<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
295important lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]296
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]297<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]298
299 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
300 line.</li>
301
302 <li>Unnamed temporaries are created when the result of a computation is not
303 assigned to a named value.</li>
304
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]305 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]306
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]307</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]308
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]309<p>...and it also shows a convention that we follow in this document. When
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]310demonstrating instructions, we will follow an instruction with a comment that
311defines the type and name of value produced. Comments are shown in italic
312text.</p>
313
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]314</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]315
316<!-- *********************************************************************** -->
317<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
318<!-- *********************************************************************** -->
319
320<!-- ======================================================================= -->
321<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
322</div>
323
324<div class="doc_text">
325
326<p>LLVM programs are composed of "Module"s, each of which is a
327translation unit of the input programs. Each module consists of
328functions, global variables, and symbol table entries. Modules may be
329combined together with the LLVM linker, which merges function (and
330global variable) definitions, resolves forward declarations, and merges
331symbol table entries. Here is an example of the "hello world" module:</p>
332
333<pre><i>; Declare the string constant as a global constant...</i>
334<a href="#identifiers">%.LC0</a> = <a href="#linkage_internal">internal</a> <a
335 href="#globalvars">constant</a> <a href="#t_array">[13 x sbyte]</a> c"hello world\0A\00" <i>; [13 x sbyte]*</i>
336
337<i>; External declaration of the puts function</i>
338<a href="#functionstructure">declare</a> int %puts(sbyte*) <i>; int(sbyte*)* </i>
339
340<i>; Definition of main function</i>
341int %main() { <i>; int()* </i>
342 <i>; Convert [13x sbyte]* to sbyte *...</i>
343 %cast210 = <a
344 href="#i_getelementptr">getelementptr</a> [13 x sbyte]* %.LC0, long 0, long 0 <i>; sbyte*</i>
345
346 <i>; Call puts function to write out the string to stdout...</i>
347 <a
348 href="#i_call">call</a> int %puts(sbyte* %cast210) <i>; int</i>
349 <a
350 href="#i_ret">ret</a> int 0<br>}<br></pre>
351
352<p>This example is made up of a <a href="#globalvars">global variable</a>
353named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
354function, and a <a href="#functionstructure">function definition</a>
355for "<tt>main</tt>".</p>
356
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]357<p>In general, a module is made up of a list of global values,
358where both functions and global variables are global values. Global values are
359represented by a pointer to a memory location (in this case, a pointer to an
360array of char, and a pointer to a function), and have one of the following <a
361href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]362
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]363</div>
364
365<!-- ======================================================================= -->
366<div class="doc_subsection">
367 <a name="linkage">Linkage Types</a>
368</div>
369
370<div class="doc_text">
371
372<p>
373All Global Variables and Functions have one of the following types of linkage:
374</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]375
376<dl>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]377
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]378 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]379
380 <dd>Global values with internal linkage are only directly accessible by
381 objects in the current module. In particular, linking code into a module with
382 an internal global value may cause the internal to be renamed as necessary to
383 avoid collisions. Because the symbol is internal to the module, all
384 references can be updated. This corresponds to the notion of the
385 '<tt>static</tt>' keyword in C, or the idea of "anonymous namespaces" in C++.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]386 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]387
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]388 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]389
390 <dd>"<tt>linkonce</tt>" linkage is similar to <tt>internal</tt> linkage, with
391 the twist that linking together two modules defining the same
392 <tt>linkonce</tt> globals will cause one of the globals to be discarded. This
393 is typically used to implement inline functions. Unreferenced
394 <tt>linkonce</tt> globals are allowed to be discarded.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]395 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]396
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]397 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]398
399 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
400 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
401 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]402 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]403
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]404 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]405
406 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
407 pointer to array type. When two global variables with appending linkage are
408 linked together, the two global arrays are appended together. This is the
409 LLVM, typesafe, equivalent of having the system linker append together
410 "sections" with identical names when .o files are linked.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]411 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]412
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]413 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]414
415 <dd>If none of the above identifiers are used, the global is externally
416 visible, meaning that it participates in linkage and can be used to resolve
417 external symbol references.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]418 </dd>
419</dl>
420
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]421<p><a name="linkage_external">For example, since the "<tt>.LC0</tt>"
422variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
423variable and was linked with this one, one of the two would be renamed,
424preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
425external (i.e., lacking any linkage declarations), they are accessible
426outside of the current module. It is illegal for a function <i>declaration</i>
427to have any linkage type other than "externally visible".</a></p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]428
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]429</div>
430
431<!-- ======================================================================= -->
432<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]433 <a name="callingconv">Calling Conventions</a>
434</div>
435
436<div class="doc_text">
437
438<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
439and <a href="#i_invoke">invokes</a> can all have an optional calling convention
440specified for the call. The calling convention of any pair of dynamic
441caller/callee must match, or the behavior of the program is undefined. The
442following calling conventions are supported by LLVM, and more may be added in
443the future:</p>
444
445<dl>
446 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
447
448 <dd>This calling convention (the default if no other calling convention is
449 specified) matches the target C calling conventions. This calling convention
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]450 supports varargs function calls and tolerates some mismatch in the declared
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]451 prototype and implemented declaration of the function (as does normal C).
452 </dd>
453
454 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
455
456 <dd>This calling convention attempts to make calls as fast as possible
457 (e.g. by passing things in registers). This calling convention allows the
458 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattner8cdc5bc2005-05-06 23:08:23 +0000[diff] [blame]459 without having to conform to an externally specified ABI. Implementations of
460 this convention should allow arbitrary tail call optimization to be supported.
461 This calling convention does not support varargs and requires the prototype of
462 all callees to exactly match the prototype of the function definition.
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]463 </dd>
464
465 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
466
467 <dd>This calling convention attempts to make code in the caller as efficient
468 as possible under the assumption that the call is not commonly executed. As
469 such, these calls often preserve all registers so that the call does not break
470 any live ranges in the caller side. This calling convention does not support
471 varargs and requires the prototype of all callees to exactly match the
472 prototype of the function definition.
473 </dd>
474
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]475 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]476
477 <dd>Any calling convention may be specified by number, allowing
478 target-specific calling conventions to be used. Target specific calling
479 conventions start at 64.
480 </dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]481</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]482
483<p>More calling conventions can be added/defined on an as-needed basis, to
484support pascal conventions or any other well-known target-independent
485convention.</p>
486
487</div>
488
489<!-- ======================================================================= -->
490<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]491 <a name="globalvars">Global Variables</a>
492</div>
493
494<div class="doc_text">
495
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]496<p>Global variables define regions of memory allocated at compilation time
497instead of run-time. Global variables may optionally be initialized. A
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]498variable may be defined as a global "constant," which indicates that the
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]499contents of the variable will <b>never</b> be modified (enabling better
500optimization, allowing the global data to be placed in the read-only section of
501an executable, etc). Note that variables that need runtime initialization
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]502cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]503
504<p>
505LLVM explicitly allows <em>declarations</em> of global variables to be marked
506constant, even if the final definition of the global is not. This capability
507can be used to enable slightly better optimization of the program, but requires
508the language definition to guarantee that optimizations based on the
509'constantness' are valid for the translation units that do not include the
510definition.
511</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]512
513<p>As SSA values, global variables define pointer values that are in
514scope (i.e. they dominate) all basic blocks in the program. Global
515variables always define a pointer to their "content" type because they
516describe a region of memory, and all memory objects in LLVM are
517accessed through pointers.</p>
518
519</div>
520
521
522<!-- ======================================================================= -->
523<div class="doc_subsection">
524 <a name="functionstructure">Functions</a>
525</div>
526
527<div class="doc_text">
528
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]529<p>LLVM function definitions consist of an optional <a href="#linkage">linkage
530type</a>, an optional <a href="#callingconv">calling convention</a>, a return
531type, a function name, a (possibly empty) argument list, an opening curly brace,
532a list of basic blocks, and a closing curly brace. LLVM function declarations
533are defined with the "<tt>declare</tt>" keyword, an optional <a
534href="#callingconv">calling convention</a>, a return type, a function name, and
535a possibly empty list of arguments.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]536
537<p>A function definition contains a list of basic blocks, forming the CFG for
538the function. Each basic block may optionally start with a label (giving the
539basic block a symbol table entry), contains a list of instructions, and ends
540with a <a href="#terminators">terminator</a> instruction (such as a branch or
541function return).</p>
542
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]543<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]544executed on entrance to the function, and it is not allowed to have predecessor
545basic blocks (i.e. there can not be any branches to the entry block of a
546function). Because the block can have no predecessors, it also cannot have any
547<a href="#i_phi">PHI nodes</a>.</p>
548
549<p>LLVM functions are identified by their name and type signature. Hence, two
550functions with the same name but different parameter lists or return values are
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]551considered different functions, and LLVM will resolve references to each
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]552appropriately.</p>
553
554</div>
555
556
557
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]558<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]559<div class="doc_section"> <a name="typesystem">Type System</a> </div>
560<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]561
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]562<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]563
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]564<p>The LLVM type system is one of the most important features of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]565intermediate representation. Being typed enables a number of
566optimizations to be performed on the IR directly, without having to do
567extra analyses on the side before the transformation. A strong type
568system makes it easier to read the generated code and enables novel
569analyses and transformations that are not feasible to perform on normal
570three address code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]571
572</div>
573
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]574<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]575<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]576<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]577<p>The primitive types are the fundamental building blocks of the LLVM
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]578system. The current set of primitive types is as follows:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]579
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]580<table class="layout">
581 <tr class="layout">
582 <td class="left">
583 <table>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]584 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]585 <tr><th>Type</th><th>Description</th></tr>
586 <tr><td><tt>void</tt></td><td>No value</td></tr>
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]587 <tr><td><tt>ubyte</tt></td><td>Unsigned 8-bit value</td></tr>
588 <tr><td><tt>ushort</tt></td><td>Unsigned 16-bit value</td></tr>
589 <tr><td><tt>uint</tt></td><td>Unsigned 32-bit value</td></tr>
590 <tr><td><tt>ulong</tt></td><td>Unsigned 64-bit value</td></tr>
591 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]592 <tr><td><tt>label</tt></td><td>Branch destination</td></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]593 </tbody>
594 </table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]595 </td>
596 <td class="right">
597 <table>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]598 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]599 <tr><th>Type</th><th>Description</th></tr>
600 <tr><td><tt>bool</tt></td><td>True or False value</td></tr>
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]601 <tr><td><tt>sbyte</tt></td><td>Signed 8-bit value</td></tr>
602 <tr><td><tt>short</tt></td><td>Signed 16-bit value</td></tr>
603 <tr><td><tt>int</tt></td><td>Signed 32-bit value</td></tr>
604 <tr><td><tt>long</tt></td><td>Signed 64-bit value</td></tr>
605 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]606 </tbody>
607 </table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]608 </td>
609 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]610</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]611</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]612
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]613<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]614<div class="doc_subsubsection"> <a name="t_classifications">Type
615Classifications</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]616<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]617<p>These different primitive types fall into a few useful
618classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]619
620<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]621 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]622 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]623 <tr>
624 <td><a name="t_signed">signed</a></td>
625 <td><tt>sbyte, short, int, long, float, double</tt></td>
626 </tr>
627 <tr>
628 <td><a name="t_unsigned">unsigned</a></td>
629 <td><tt>ubyte, ushort, uint, ulong</tt></td>
630 </tr>
631 <tr>
632 <td><a name="t_integer">integer</a></td>
633 <td><tt>ubyte, sbyte, ushort, short, uint, int, ulong, long</tt></td>
634 </tr>
635 <tr>
636 <td><a name="t_integral">integral</a></td>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]637 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long</tt>
638 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]639 </tr>
640 <tr>
641 <td><a name="t_floating">floating point</a></td>
642 <td><tt>float, double</tt></td>
643 </tr>
644 <tr>
645 <td><a name="t_firstclass">first class</a></td>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]646 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long,<br>
647 float, double, <a href="#t_pointer">pointer</a>,
648 <a href="#t_packed">packed</a></tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]649 </tr>
650 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]651</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]652
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]653<p>The <a href="#t_firstclass">first class</a> types are perhaps the
654most important. Values of these types are the only ones which can be
655produced by instructions, passed as arguments, or used as operands to
656instructions. This means that all structures and arrays must be
657manipulated either by pointer or by component.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]658</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]659
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]660<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]661<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]662
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]663<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]664
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]665<p>The real power in LLVM comes from the derived types in the system.
666This is what allows a programmer to represent arrays, functions,
667pointers, and other useful types. Note that these derived types may be
668recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]669
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]670</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]671
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]672<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]673<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]674
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]675<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]676
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]677<h5>Overview:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]678
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]679<p>The array type is a very simple derived type that arranges elements
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]680sequentially in memory. The array type requires a size (number of
681elements) and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]682
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]683<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]684
685<pre>
686 [&lt;# elements&gt; x &lt;elementtype&gt;]
687</pre>
688
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]689<p>The number of elements is a constant integer value; elementtype may
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]690be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]691
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]692<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]693<table class="layout">
694 <tr class="layout">
695 <td class="left">
696 <tt>[40 x int ]</tt><br/>
697 <tt>[41 x int ]</tt><br/>
698 <tt>[40 x uint]</tt><br/>
699 </td>
700 <td class="left">
701 Array of 40 integer values.<br/>
702 Array of 41 integer values.<br/>
703 Array of 40 unsigned integer values.<br/>
704 </td>
705 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]706</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]707<p>Here are some examples of multidimensional arrays:</p>
708<table class="layout">
709 <tr class="layout">
710 <td class="left">
711 <tt>[3 x [4 x int]]</tt><br/>
712 <tt>[12 x [10 x float]]</tt><br/>
713 <tt>[2 x [3 x [4 x uint]]]</tt><br/>
714 </td>
715 <td class="left">
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]716 3x4 array of integer values.<br/>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]717 12x10 array of single precision floating point values.<br/>
718 2x3x4 array of unsigned integer values.<br/>
719 </td>
720 </tr>
721</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]722
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]723<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
724length array. Normally, accesses past the end of an array are undefined in
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]725LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
726As a special case, however, zero length arrays are recognized to be variable
727length. This allows implementation of 'pascal style arrays' with the LLVM
728type "{ int, [0 x float]}", for example.</p>
729
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]730</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]731
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]732<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]733<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]734<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]735<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]736<p>The function type can be thought of as a function signature. It
737consists of a return type and a list of formal parameter types.
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]738Function types are usually used to build virtual function tables
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]739(which are structures of pointers to functions), for indirect function
740calls, and when defining a function.</p>
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]741<p>
742The return type of a function type cannot be an aggregate type.
743</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]744<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]745<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]746<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]747specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]748which indicates that the function takes a variable number of arguments.
749Variable argument functions can access their arguments with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]750 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]751<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]752<table class="layout">
753 <tr class="layout">
754 <td class="left">
755 <tt>int (int)</tt> <br/>
756 <tt>float (int, int *) *</tt><br/>
757 <tt>int (sbyte *, ...)</tt><br/>
758 </td>
759 <td class="left">
760 function taking an <tt>int</tt>, returning an <tt>int</tt><br/>
761 <a href="#t_pointer">Pointer</a> to a function that takes an
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]762 <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]763 returning <tt>float</tt>.<br/>
764 A vararg function that takes at least one <a href="#t_pointer">pointer</a>
765 to <tt>sbyte</tt> (signed char in C), which returns an integer. This is
766 the signature for <tt>printf</tt> in LLVM.<br/>
767 </td>
768 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]769</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]770
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]771</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]772<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]773<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]774<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]775<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]776<p>The structure type is used to represent a collection of data members
777together in memory. The packing of the field types is defined to match
778the ABI of the underlying processor. The elements of a structure may
779be any type that has a size.</p>
780<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
781and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
782field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
783instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]784<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]785<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]786<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]787<table class="layout">
788 <tr class="layout">
789 <td class="left">
790 <tt>{ int, int, int }</tt><br/>
791 <tt>{ float, int (int) * }</tt><br/>
792 </td>
793 <td class="left">
794 a triple of three <tt>int</tt> values<br/>
795 A pair, where the first element is a <tt>float</tt> and the second element
796 is a <a href="#t_pointer">pointer</a> to a <a href="#t_function">function</a>
797 that takes an <tt>int</tt>, returning an <tt>int</tt>.<br/>
798 </td>
799 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]800</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]801</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]802
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]803<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]804<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]805<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]806<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]807<p>As in many languages, the pointer type represents a pointer or
808reference to another object, which must live in memory.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]809<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]810<pre> &lt;type&gt; *<br></pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]811<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]812<table class="layout">
813 <tr class="layout">
814 <td class="left">
815 <tt>[4x int]*</tt><br/>
816 <tt>int (int *) *</tt><br/>
817 </td>
818 <td class="left">
819 A <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a> of
820 four <tt>int</tt> values<br/>
821 A <a href="#t_pointer">pointer</a> to a <a
Chris Lattnera977c482005-02-19 02:22:14 +0000[diff] [blame]822 href="#t_function">function</a> that takes an <tt>int*</tt>, returning an
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]823 <tt>int</tt>.<br/>
824 </td>
825 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]826</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]827</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]828
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]829<!-- _______________________________________________________________________ -->
830<div class="doc_subsubsection"> <a name="t_packed">Packed Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]831<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]832
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]833<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]834
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]835<p>A packed type is a simple derived type that represents a vector
836of elements. Packed types are used when multiple primitive data
837are operated in parallel using a single instruction (SIMD).
838A packed type requires a size (number of
839elements) and an underlying primitive data type. Packed types are
840considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]841
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]842<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]843
844<pre>
845 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
846</pre>
847
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]848<p>The number of elements is a constant integer value; elementtype may
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]849be any integral or floating point type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]850
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]851<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]852
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]853<table class="layout">
854 <tr class="layout">
855 <td class="left">
856 <tt>&lt;4 x int&gt;</tt><br/>
857 <tt>&lt;8 x float&gt;</tt><br/>
858 <tt>&lt;2 x uint&gt;</tt><br/>
859 </td>
860 <td class="left">
861 Packed vector of 4 integer values.<br/>
862 Packed vector of 8 floating-point values.<br/>
863 Packed vector of 2 unsigned integer values.<br/>
864 </td>
865 </tr>
866</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]867</div>
868
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]869<!-- _______________________________________________________________________ -->
870<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
871<div class="doc_text">
872
873<h5>Overview:</h5>
874
875<p>Opaque types are used to represent unknown types in the system. This
876corresponds (for example) to the C notion of a foward declared structure type.
877In LLVM, opaque types can eventually be resolved to any type (not just a
878structure type).</p>
879
880<h5>Syntax:</h5>
881
882<pre>
883 opaque
884</pre>
885
886<h5>Examples:</h5>
887
888<table class="layout">
889 <tr class="layout">
890 <td class="left">
891 <tt>opaque</tt>
892 </td>
893 <td class="left">
894 An opaque type.<br/>
895 </td>
896 </tr>
897</table>
898</div>
899
900
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]901<!-- *********************************************************************** -->
902<div class="doc_section"> <a name="constants">Constants</a> </div>
903<!-- *********************************************************************** -->
904
905<div class="doc_text">
906
907<p>LLVM has several different basic types of constants. This section describes
908them all and their syntax.</p>
909
910</div>
911
912<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]913<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]914
915<div class="doc_text">
916
917<dl>
918 <dt><b>Boolean constants</b></dt>
919
920 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
921 constants of the <tt><a href="#t_primitive">bool</a></tt> type.
922 </dd>
923
924 <dt><b>Integer constants</b></dt>
925
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]926 <dd>Standard integers (such as '4') are constants of the <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]927 href="#t_integer">integer</a> type. Negative numbers may be used with signed
928 integer types.
929 </dd>
930
931 <dt><b>Floating point constants</b></dt>
932
933 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
934 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]935 notation (see below). Floating point constants must have a <a
936 href="#t_floating">floating point</a> type. </dd>
937
938 <dt><b>Null pointer constants</b></dt>
939
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]940 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]941 and must be of <a href="#t_pointer">pointer type</a>.</dd>
942
943</dl>
944
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]945<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]946of floating point constants. For example, the form '<tt>double
9470x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
9484.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]949(and the only time that they are generated by the disassembler) is when a
950floating point constant must be emitted but it cannot be represented as a
951decimal floating point number. For example, NaN's, infinities, and other
952special values are represented in their IEEE hexadecimal format so that
953assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]954
955</div>
956
957<!-- ======================================================================= -->
958<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
959</div>
960
961<div class="doc_text">
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]962<p>Aggregate constants arise from aggregation of simple constants
963and smaller aggregate constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]964
965<dl>
966 <dt><b>Structure constants</b></dt>
967
968 <dd>Structure constants are represented with notation similar to structure
969 type definitions (a comma separated list of elements, surrounded by braces
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]970 (<tt>{}</tt>)). For example: "<tt>{ int 4, float 17.0, int* %G }</tt>",
971 where "<tt>%G</tt>" is declared as "<tt>%G = external global int</tt>". Structure constants
972 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]973 types of elements must match those specified by the type.
974 </dd>
975
976 <dt><b>Array constants</b></dt>
977
978 <dd>Array constants are represented with notation similar to array type
979 definitions (a comma separated list of elements, surrounded by square brackets
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]980 (<tt>[]</tt>)). For example: "<tt>[ int 42, int 11, int 74 ]</tt>". Array
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]981 constants must have <a href="#t_array">array type</a>, and the number and
982 types of elements must match those specified by the type.
983 </dd>
984
985 <dt><b>Packed constants</b></dt>
986
987 <dd>Packed constants are represented with notation similar to packed type
988 definitions (a comma separated list of elements, surrounded by
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]989 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]990 int 11, int 74, int 100 &gt;</tt>". Packed constants must have <a
991 href="#t_packed">packed type</a>, and the number and types of elements must
992 match those specified by the type.
993 </dd>
994
995 <dt><b>Zero initialization</b></dt>
996
997 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
998 value to zero of <em>any</em> type, including scalar and aggregate types.
999 This is often used to avoid having to print large zero initializers (e.g. for
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]1000 large arrays) and is always exactly equivalent to using explicit zero
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1001 initializers.
1002 </dd>
1003</dl>
1004
1005</div>
1006
1007<!-- ======================================================================= -->
1008<div class="doc_subsection">
1009 <a name="globalconstants">Global Variable and Function Addresses</a>
1010</div>
1011
1012<div class="doc_text">
1013
1014<p>The addresses of <a href="#globalvars">global variables</a> and <a
1015href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1016constants. These constants are explicitly referenced when the <a
1017href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1018href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1019file:</p>
1020
1021<pre>
1022 %X = global int 17
1023 %Y = global int 42
1024 %Z = global [2 x int*] [ int* %X, int* %Y ]
1025</pre>
1026
1027</div>
1028
1029<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1030<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1031<div class="doc_text">
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1032 <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]1033 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1034 a constant is permitted.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1035
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1036 <p>Undefined values indicate to the compiler that the program is well defined
1037 no matter what value is used, giving the compiler more freedom to optimize.
1038 </p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1039</div>
1040
1041<!-- ======================================================================= -->
1042<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1043</div>
1044
1045<div class="doc_text">
1046
1047<p>Constant expressions are used to allow expressions involving other constants
1048to be used as constants. Constant expressions may be of any <a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1049href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1050that does not have side effects (e.g. load and call are not supported). The
1051following is the syntax for constant expressions:</p>
1052
1053<dl>
1054 <dt><b><tt>cast ( CST to TYPE )</tt></b></dt>
1055
1056 <dd>Cast a constant to another type.</dd>
1057
1058 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1059
1060 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1061 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1062 instruction, the index list may have zero or more indexes, which are required
1063 to make sense for the type of "CSTPTR".</dd>
1064
1065 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1066
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1067 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1068 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1069 binary</a> operations. The constraints on operands are the same as those for
1070 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1071 values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1072</dl>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1073</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]1074
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1075<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1076<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1077<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1078
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1079<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1080
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1081<p>The LLVM instruction set consists of several different
1082classifications of instructions: <a href="#terminators">terminator
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1083instructions</a>, <a href="#binaryops">binary instructions</a>,
1084<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1085 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1086instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1087
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1088</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1089
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1090<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1091<div class="doc_subsection"> <a name="terminators">Terminator
1092Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1093
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1094<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1095
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1096<p>As mentioned <a href="#functionstructure">previously</a>, every
1097basic block in a program ends with a "Terminator" instruction, which
1098indicates which block should be executed after the current block is
1099finished. These terminator instructions typically yield a '<tt>void</tt>'
1100value: they produce control flow, not values (the one exception being
1101the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1102<p>There are six different terminator instructions: the '<a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1103 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1104instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1105the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1106 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1107 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1108
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1109</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1110
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1111<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1112<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1113Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1114<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1115<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1116<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]1117 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1118</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1119<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1120<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1121value) from a function back to the caller.</p>
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]1122<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1123returns a value and then causes control flow, and one that just causes
1124control flow to occur.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1125<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1126<p>The '<tt>ret</tt>' instruction may return any '<a
1127 href="#t_firstclass">first class</a>' type. Notice that a function is
1128not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1129instruction inside of the function that returns a value that does not
1130match the return type of the function.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1131<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1132<p>When the '<tt>ret</tt>' instruction is executed, control flow
1133returns back to the calling function's context. If the caller is a "<a
John Criswellfa081872004-06-25 15:16:57 +0000[diff] [blame]1134 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1135the instruction after the call. If the caller was an "<a
1136 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1137at the beginning of the "normal" destination block. If the instruction
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1138returns a value, that value shall set the call or invoke instruction's
1139return value.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1140<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1141<pre> ret int 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1142 ret void <i>; Return from a void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1143</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1144</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1145<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1146<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1147<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1148<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1149<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]1150</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1151<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1152<p>The '<tt>br</tt>' instruction is used to cause control flow to
1153transfer to a different basic block in the current function. There are
1154two forms of this instruction, corresponding to a conditional branch
1155and an unconditional branch.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1156<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1157<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
1158single '<tt>bool</tt>' value and two '<tt>label</tt>' values. The
1159unconditional form of the '<tt>br</tt>' instruction takes a single '<tt>label</tt>'
1160value as a target.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1161<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1162<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>bool</tt>'
1163argument is evaluated. If the value is <tt>true</tt>, control flows
1164to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1165control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1166<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1167<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
1168 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]1169</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1170<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1171<div class="doc_subsubsection">
1172 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1173</div>
1174
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1175<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1176<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1177
1178<pre>
1179 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1180</pre>
1181
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1182<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1183
1184<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1185several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1186instruction, allowing a branch to occur to one of many possible
1187destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1188
1189
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1190<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1191
1192<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1193comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1194an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1195table is not allowed to contain duplicate constant entries.</p>
1196
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1197<h5>Semantics:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1198
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1199<p>The <tt>switch</tt> instruction specifies a table of values and
1200destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1201table is searched for the given value. If the value is found, control flow is
1202transfered to the corresponding destination; otherwise, control flow is
1203transfered to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1204
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1205<h5>Implementation:</h5>
1206
1207<p>Depending on properties of the target machine and the particular
1208<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1209ways. For example, it could be generated as a series of chained conditional
1210branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1211
1212<h5>Example:</h5>
1213
1214<pre>
1215 <i>; Emulate a conditional br instruction</i>
1216 %Val = <a href="#i_cast">cast</a> bool %value to int
1217 switch int %Val, label %truedest [int 0, label %falsedest ]
1218
1219 <i>; Emulate an unconditional br instruction</i>
1220 switch uint 0, label %dest [ ]
1221
1222 <i>; Implement a jump table:</i>
1223 switch uint %val, label %otherwise [ uint 0, label %onzero
1224 uint 1, label %onone
1225 uint 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1226</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1227</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1228
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1229<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1230<div class="doc_subsubsection">
1231 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1232</div>
1233
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1234<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1235
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1236<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1237
1238<pre>
1239 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)
1240 to label &lt;normal label&gt; except label &lt;exception label&gt;
1241</pre>
1242
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1243<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1244
1245<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1246function, with the possibility of control flow transfer to either the
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1247'<tt>normal</tt>' label or the
1248'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1249"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1250"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1251href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1252continued at the dynamically nearest "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1253
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1254<h5>Arguments:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1255
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1256<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1257
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1258<ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1259 <li>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1260 The optional "cconv" marker indicates which <a href="callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1261 convention</a> the call should use. If none is specified, the call defaults
1262 to using C calling conventions.
1263 </li>
1264 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1265 function value being invoked. In most cases, this is a direct function
1266 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1267 an arbitrary pointer to function value.
1268 </li>
1269
1270 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1271 function to be invoked. </li>
1272
1273 <li>'<tt>function args</tt>': argument list whose types match the function
1274 signature argument types. If the function signature indicates the function
1275 accepts a variable number of arguments, the extra arguments can be
1276 specified. </li>
1277
1278 <li>'<tt>normal label</tt>': the label reached when the called function
1279 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1280
1281 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1282 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1283
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1284</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1285
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1286<h5>Semantics:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1287
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1288<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1289href="#i_call">call</a></tt>' instruction in most regards. The primary
1290difference is that it establishes an association with a label, which is used by
1291the runtime library to unwind the stack.</p>
1292
1293<p>This instruction is used in languages with destructors to ensure that proper
1294cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1295exception. Additionally, this is important for implementation of
1296'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1297
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1298<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1299<pre>
1300 %retval = invoke int %Test(int 15) to label %Continue
1301 except label %TestCleanup <i>; {int}:retval set</i>
1302 %retval = invoke <a href="#callingconv">coldcc</a> int %Test(int 15) to label %Continue
1303 except label %TestCleanup <i>; {int}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1304</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1305</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1306
1307
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1308<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1309
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1310<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1311Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1312
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1313<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1314
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1315<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1316<pre>
1317 unwind
1318</pre>
1319
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1320<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1321
1322<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1323at the first callee in the dynamic call stack which used an <a
1324href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1325primarily used to implement exception handling.</p>
1326
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1327<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1328
1329<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
1330immediately halt. The dynamic call stack is then searched for the first <a
1331href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
1332execution continues at the "exceptional" destination block specified by the
1333<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
1334dynamic call chain, undefined behavior results.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1335</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1336
1337<!-- _______________________________________________________________________ -->
1338
1339<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
1340Instruction</a> </div>
1341
1342<div class="doc_text">
1343
1344<h5>Syntax:</h5>
1345<pre>
1346 unreachable
1347</pre>
1348
1349<h5>Overview:</h5>
1350
1351<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
1352instruction is used to inform the optimizer that a particular portion of the
1353code is not reachable. This can be used to indicate that the code after a
1354no-return function cannot be reached, and other facts.</p>
1355
1356<h5>Semantics:</h5>
1357
1358<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
1359</div>
1360
1361
1362
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1363<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1364<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1365<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1366<p>Binary operators are used to do most of the computation in a
1367program. They require two operands, execute an operation on them, and
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1368produce a single value. The operands might represent
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1369multiple data, as is the case with the <a href="#t_packed">packed</a> data type.
1370The result value of a binary operator is not
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1371necessarily the same type as its operands.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1372<p>There are several different binary operators:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1373</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1374<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1375<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
1376Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1377<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1378<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1379<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]1380</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1381<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1382<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1383<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1384<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1385 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
1386 This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1387Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1388<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1389<p>The value produced is the integer or floating point sum of the two
1390operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1391<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1392<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]1393</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1394</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1395<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1396<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
1397Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1398<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1399<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1400<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]1401</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1402<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1403<p>The '<tt>sub</tt>' instruction returns the difference of its two
1404operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1405<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
1406instruction present in most other intermediate representations.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1407<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1408<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1409 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1410values.
1411This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1412Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1413<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1414<p>The value produced is the integer or floating point difference of
1415the two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1416<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1417<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]1418 &lt;result&gt; = sub int 0, %val <i>; yields {int}:result = -%var</i>
1419</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1420</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1421<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1422<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
1423Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1424<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1425<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1426<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]1427</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1428<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1429<p>The '<tt>mul</tt>' instruction returns the product of its two
1430operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1431<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1432<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1433 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1434values.
1435This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1436Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1437<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1438<p>The value produced is the integer or floating point product of the
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1439two operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1440<p>There is no signed vs unsigned multiplication. The appropriate
1441action is taken based on the type of the operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1442<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1443<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]1444</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1445</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1446<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1447<div class="doc_subsubsection"> <a name="i_div">'<tt>div</tt>'
1448Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1449<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1450<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1451<pre> &lt;result&gt; = div &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1452</pre>
1453<h5>Overview:</h5>
1454<p>The '<tt>div</tt>' instruction returns the quotient of its two
1455operands.</p>
1456<h5>Arguments:</h5>
1457<p>The two arguments to the '<tt>div</tt>' instruction must be either <a
1458 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1459values.
1460This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1461Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1462<h5>Semantics:</h5>
1463<p>The value produced is the integer or floating point quotient of the
1464two operands.</p>
1465<h5>Example:</h5>
1466<pre> &lt;result&gt; = div int 4, %var <i>; yields {int}:result = 4 / %var</i>
1467</pre>
1468</div>
1469<!-- _______________________________________________________________________ -->
1470<div class="doc_subsubsection"> <a name="i_rem">'<tt>rem</tt>'
1471Instruction</a> </div>
1472<div class="doc_text">
1473<h5>Syntax:</h5>
1474<pre> &lt;result&gt; = rem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1475</pre>
1476<h5>Overview:</h5>
1477<p>The '<tt>rem</tt>' instruction returns the remainder from the
1478division of its two operands.</p>
1479<h5>Arguments:</h5>
1480<p>The two arguments to the '<tt>rem</tt>' instruction must be either <a
1481 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1482values.
1483This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1484Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1485<h5>Semantics:</h5>
1486<p>This returns the <i>remainder</i> of a division (where the result
1487has the same sign as the divisor), not the <i>modulus</i> (where the
1488result has the same sign as the dividend) of a value. For more
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]1489information about the difference, see <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1490 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
1491Math Forum</a>.</p>
1492<h5>Example:</h5>
1493<pre> &lt;result&gt; = rem int 4, %var <i>; yields {int}:result = 4 % %var</i>
1494</pre>
1495</div>
1496<!-- _______________________________________________________________________ -->
1497<div class="doc_subsubsection"> <a name="i_setcc">'<tt>set<i>cc</i></tt>'
1498Instructions</a> </div>
1499<div class="doc_text">
1500<h5>Syntax:</h5>
1501<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]1502 &lt;result&gt; = setne &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1503 &lt;result&gt; = setlt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1504 &lt;result&gt; = setgt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1505 &lt;result&gt; = setle &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1506 &lt;result&gt; = setge &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1507</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1508<h5>Overview:</h5>
1509<p>The '<tt>set<i>cc</i></tt>' family of instructions returns a boolean
1510value based on a comparison of their two operands.</p>
1511<h5>Arguments:</h5>
1512<p>The two arguments to the '<tt>set<i>cc</i></tt>' instructions must
1513be of <a href="#t_firstclass">first class</a> type (it is not possible
1514to compare '<tt>label</tt>'s, '<tt>array</tt>'s, '<tt>structure</tt>'
1515or '<tt>void</tt>' values, etc...). Both arguments must have identical
1516types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1517<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1518<p>The '<tt>seteq</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1519value if both operands are equal.<br>
1520The '<tt>setne</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1521value if both operands are unequal.<br>
1522The '<tt>setlt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1523value if the first operand is less than the second operand.<br>
1524The '<tt>setgt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1525value if the first operand is greater than the second operand.<br>
1526The '<tt>setle</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1527value if the first operand is less than or equal to the second operand.<br>
1528The '<tt>setge</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1529value if the first operand is greater than or equal to the second
1530operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1531<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1532<pre> &lt;result&gt; = seteq int 4, 5 <i>; yields {bool}:result = false</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1533 &lt;result&gt; = setne float 4, 5 <i>; yields {bool}:result = true</i>
1534 &lt;result&gt; = setlt uint 4, 5 <i>; yields {bool}:result = true</i>
1535 &lt;result&gt; = setgt sbyte 4, 5 <i>; yields {bool}:result = false</i>
1536 &lt;result&gt; = setle sbyte 4, 5 <i>; yields {bool}:result = true</i>
1537 &lt;result&gt; = setge sbyte 4, 5 <i>; yields {bool}:result = false</i>
1538</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1539</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1540<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1541<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
1542Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1543<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1544<p>Bitwise binary operators are used to do various forms of
1545bit-twiddling in a program. They are generally very efficient
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1546instructions and can commonly be strength reduced from other
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1547instructions. They require two operands, execute an operation on them,
1548and produce a single value. The resulting value of the bitwise binary
1549operators is always the same type as its first operand.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1550</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1551<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1552<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
1553Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1554<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1555<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1556<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]1557</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1558<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1559<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
1560its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1561<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1562<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1563 href="#t_integral">integral</a> values. Both arguments must have
1564identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1565<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1566<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1567<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1568<div style="align: center">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1569<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1570 <tbody>
1571 <tr>
1572 <td>In0</td>
1573 <td>In1</td>
1574 <td>Out</td>
1575 </tr>
1576 <tr>
1577 <td>0</td>
1578 <td>0</td>
1579 <td>0</td>
1580 </tr>
1581 <tr>
1582 <td>0</td>
1583 <td>1</td>
1584 <td>0</td>
1585 </tr>
1586 <tr>
1587 <td>1</td>
1588 <td>0</td>
1589 <td>0</td>
1590 </tr>
1591 <tr>
1592 <td>1</td>
1593 <td>1</td>
1594 <td>1</td>
1595 </tr>
1596 </tbody>
1597</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1598</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1599<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1600<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]1601 &lt;result&gt; = and int 15, 40 <i>; yields {int}:result = 8</i>
1602 &lt;result&gt; = and int 4, 8 <i>; yields {int}:result = 0</i>
1603</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1604</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1605<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1606<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1607<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1608<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1609<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]1610</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1611<h5>Overview:</h5>
1612<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
1613or of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1614<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1615<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1616 href="#t_integral">integral</a> values. Both arguments must have
1617identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1618<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1619<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1620<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1621<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1622<table border="1" cellspacing="0" cellpadding="4">
1623 <tbody>
1624 <tr>
1625 <td>In0</td>
1626 <td>In1</td>
1627 <td>Out</td>
1628 </tr>
1629 <tr>
1630 <td>0</td>
1631 <td>0</td>
1632 <td>0</td>
1633 </tr>
1634 <tr>
1635 <td>0</td>
1636 <td>1</td>
1637 <td>1</td>
1638 </tr>
1639 <tr>
1640 <td>1</td>
1641 <td>0</td>
1642 <td>1</td>
1643 </tr>
1644 <tr>
1645 <td>1</td>
1646 <td>1</td>
1647 <td>1</td>
1648 </tr>
1649 </tbody>
1650</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1651</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1652<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1653<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]1654 &lt;result&gt; = or int 15, 40 <i>; yields {int}:result = 47</i>
1655 &lt;result&gt; = or int 4, 8 <i>; yields {int}:result = 12</i>
1656</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1657</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1658<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1659<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
1660Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1661<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1662<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1663<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]1664</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1665<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1666<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
1667or of its two operands. The <tt>xor</tt> is used to implement the
1668"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1669<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1670<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1671 href="#t_integral">integral</a> values. Both arguments must have
1672identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1673<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1674<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1675<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1676<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1677<table border="1" cellspacing="0" cellpadding="4">
1678 <tbody>
1679 <tr>
1680 <td>In0</td>
1681 <td>In1</td>
1682 <td>Out</td>
1683 </tr>
1684 <tr>
1685 <td>0</td>
1686 <td>0</td>
1687 <td>0</td>
1688 </tr>
1689 <tr>
1690 <td>0</td>
1691 <td>1</td>
1692 <td>1</td>
1693 </tr>
1694 <tr>
1695 <td>1</td>
1696 <td>0</td>
1697 <td>1</td>
1698 </tr>
1699 <tr>
1700 <td>1</td>
1701 <td>1</td>
1702 <td>0</td>
1703 </tr>
1704 </tbody>
1705</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1706</div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1707<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1708<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1709<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]1710 &lt;result&gt; = xor int 15, 40 <i>; yields {int}:result = 39</i>
1711 &lt;result&gt; = xor int 4, 8 <i>; yields {int}:result = 12</i>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1712 &lt;result&gt; = xor int %V, -1 <i>; yields {int}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1713</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1714</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1715<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1716<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
1717Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1718<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1719<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1720<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]1721</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1722<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1723<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
1724the left a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1725<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1726<p>The first argument to the '<tt>shl</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1727 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1728type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1729<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1730<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]1731<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1732<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]1733 &lt;result&gt; = shl int 4, ubyte 2 <i>; yields {int}:result = 16</i>
1734 &lt;result&gt; = shl int 1, ubyte 10 <i>; yields {int}:result = 1024</i>
1735</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1736</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1737<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1738<div class="doc_subsubsection"> <a name="i_shr">'<tt>shr</tt>'
1739Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1740<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1741<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1742<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]1743</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1744<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1745<p>The '<tt>shr</tt>' instruction returns the first operand shifted to
1746the right a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1747<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1748<p>The first argument to the '<tt>shr</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1749 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1750type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1751<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1752<p>If the first argument is a <a href="#t_signed">signed</a> type, the
1753most significant bit is duplicated in the newly free'd bit positions.
1754If the first argument is unsigned, zero bits shall fill the empty
1755positions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1756<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1757<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]1758 &lt;result&gt; = shr uint 4, ubyte 1 <i>; yields {uint}:result = 2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1759 &lt;result&gt; = shr int 4, ubyte 2 <i>; yields {int}:result = 1</i>
Chris Lattner8c6bb902003-06-18 21:30:51 +0000[diff] [blame]1760 &lt;result&gt; = shr sbyte 4, ubyte 3 <i>; yields {sbyte}:result = 0</i>
1761 &lt;result&gt; = shr sbyte -2, ubyte 1 <i>; yields {sbyte}:result = -1</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1762</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1763</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1764<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1765<div class="doc_subsection"> <a name="memoryops">Memory Access
1766Operations</a></div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1767<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1768<p>A key design point of an SSA-based representation is how it
1769represents memory. In LLVM, no memory locations are in SSA form, which
1770makes things very simple. This section describes how to read, write,
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1771allocate, and free memory in LLVM.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1772</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1773<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1774<div class="doc_subsubsection"> <a name="i_malloc">'<tt>malloc</tt>'
1775Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1776<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1777<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1778<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]1779 &lt;result&gt; = malloc &lt;type&gt; <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1780</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1781<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1782<p>The '<tt>malloc</tt>' instruction allocates memory from the system
1783heap and returns a pointer to it.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1784<h5>Arguments:</h5>
John Criswell6e4ca612004-02-24 16:13:56 +0000[diff] [blame]1785<p>The '<tt>malloc</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
1786bytes of memory from the operating system and returns a pointer of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1787appropriate type to the program. The second form of the instruction is
1788a shorter version of the first instruction that defaults to allocating
1789one element.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1790<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1791<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1792<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
1793a pointer is returned.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1794<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1795<pre> %array = malloc [4 x ubyte ] <i>; yields {[%4 x ubyte]*}:array</i>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1796
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1797 %size = <a
1798 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]1799 %array1 = malloc ubyte, uint 4 <i>; yields {ubyte*}:array1</i>
1800 %array2 = malloc [12 x ubyte], uint %size <i>; yields {[12 x ubyte]*}:array2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1801</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1802</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1803<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1804<div class="doc_subsubsection"> <a name="i_free">'<tt>free</tt>'
1805Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1806<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1807<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1808<pre> free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1809</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1810<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1811<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1812memory heap to be reallocated in the future.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1813<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1814<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1815<p>'<tt>value</tt>' shall be a pointer value that points to a value
1816that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
1817instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1818<h5>Semantics:</h5>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1819<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1820after this instruction executes.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1821<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1822<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]1823 free [4 x ubyte]* %array
1824</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1825</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1826<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1827<div class="doc_subsubsection"> <a name="i_alloca">'<tt>alloca</tt>'
1828Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1829<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1830<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1831<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]1832 &lt;result&gt; = alloca &lt;type&gt; <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1833</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1834<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1835<p>The '<tt>alloca</tt>' instruction allocates memory on the current
1836stack frame of the procedure that is live until the current function
1837returns to its caller.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1838<h5>Arguments:</h5>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1839<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1840bytes of memory on the runtime stack, returning a pointer of the
1841appropriate type to the program. The second form of the instruction is
1842a shorter version of the first that defaults to allocating one element.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1843<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1844<h5>Semantics:</h5>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1845<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1846memory is automatically released when the function returns. The '<tt>alloca</tt>'
1847instruction is commonly used to represent automatic variables that must
1848have an address available. When the function returns (either with the <tt><a
John Criswelldae2e932005-05-12 16:55:34 +0000[diff] [blame]1849 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]1850instructions), the memory is reclaimed.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1851<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1852<pre> %ptr = alloca int <i>; yields {int*}:ptr</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1853 %ptr = alloca int, uint 4 <i>; yields {int*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1854</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1855</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1856<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1857<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
1858Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1859<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1860<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1861<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]1862<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1863<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1864<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1865<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]1866address from which to load. The pointer must point to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1867 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]1868marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1869the number or order of execution of this <tt>load</tt> with other
1870volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
1871instructions. </p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1872<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1873<p>The location of memory pointed to is loaded.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1874<h5>Examples:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1875<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1876 <a
1877 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1878 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1879</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1880</div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1881<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1882<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
1883Instruction</a> </div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1884<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1885<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]1886 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]1887</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1888<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1889<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1890<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1891<p>There are two arguments to the '<tt>store</tt>' instruction: a value
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame^]1892to store and an address in which to store it. The type of the '<tt>&lt;pointer&gt;</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1893operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1894operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1895optimizer is not allowed to modify the number or order of execution of
1896this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
1897 href="#i_store">store</a></tt> instructions.</p>
1898<h5>Semantics:</h5>
1899<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
1900at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1901<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1902<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1903 <a
1904 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1905 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1906</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1907<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1908<div class="doc_subsubsection">
1909 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
1910</div>
1911
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1912<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1913<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1914<pre>
1915 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
1916</pre>
1917
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1918<h5>Overview:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1919
1920<p>
1921The '<tt>getelementptr</tt>' instruction is used to get the address of a
1922subelement of an aggregate data structure.</p>
1923
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1924<h5>Arguments:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1925
1926<p>This instruction takes a list of integer constants that indicate what
1927elements of the aggregate object to index to. The actual types of the arguments
1928provided depend on the type of the first pointer argument. The
1929'<tt>getelementptr</tt>' instruction is used to index down through the type
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]1930levels of a structure or to a specific index in an array. When indexing into a
1931structure, only <tt>uint</tt>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1932integer constants are allowed. When indexing into an array or pointer,
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1933<tt>int</tt> and <tt>long</tt> indexes are allowed of any sign.</p>
1934
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1935<p>For example, let's consider a C code fragment and how it gets
1936compiled to LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1937
1938<pre>
1939 struct RT {
1940 char A;
1941 int B[10][20];
1942 char C;
1943 };
1944 struct ST {
1945 int X;
1946 double Y;
1947 struct RT Z;
1948 };
1949
1950 int *foo(struct ST *s) {
1951 return &amp;s[1].Z.B[5][13];
1952 }
1953</pre>
1954
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1955<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1956
1957<pre>
1958 %RT = type { sbyte, [10 x [20 x int]], sbyte }
1959 %ST = type { int, double, %RT }
1960
Brian Gaeke7283e7c2004-07-02 21:08:14 +0000[diff] [blame]1961 implementation
1962
1963 int* %foo(%ST* %s) {
1964 entry:
1965 %reg = getelementptr %ST* %s, int 1, uint 2, uint 1, int 5, int 13
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1966 ret int* %reg
1967 }
1968</pre>
1969
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1970<h5>Semantics:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1971
1972<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1973on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1974and <a href="#t_array">array</a> types require <tt>uint</tt>, <tt>int</tt>,
1975<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]1976types require <tt>uint</tt> <b>constants</b>.</p>
1977
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1978<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]1979type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ int, double, %RT
1980}</tt>' type, a structure. The second index indexes into the third element of
1981the structure, yielding a '<tt>%RT</tt>' = '<tt>{ sbyte, [10 x [20 x int]],
1982sbyte }</tt>' type, another structure. The third index indexes into the second
1983element of the structure, yielding a '<tt>[10 x [20 x int]]</tt>' type, an
1984array. The two dimensions of the array are subscripted into, yielding an
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]1985'<tt>int</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1986to this element, thus computing a value of '<tt>int*</tt>' type.</p>
1987
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1988<p>Note that it is perfectly legal to index partially through a
1989structure, returning a pointer to an inner element. Because of this,
1990the LLVM code for the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1991
1992<pre>
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]1993 int* %foo(%ST* %s) {
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1994 %t1 = getelementptr %ST* %s, int 1 <i>; yields %ST*:%t1</i>
1995 %t2 = getelementptr %ST* %t1, int 0, uint 2 <i>; yields %RT*:%t2</i>
1996 %t3 = getelementptr %RT* %t2, int 0, uint 1 <i>; yields [10 x [20 x int]]*:%t3</i>
1997 %t4 = getelementptr [10 x [20 x int]]* %t3, int 0, int 5 <i>; yields [20 x int]*:%t4</i>
1998 %t5 = getelementptr [20 x int]* %t4, int 0, int 13 <i>; yields int*:%t5</i>
1999 ret int* %t5
2000 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2001</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]2002
2003<p>Note that it is undefined to access an array out of bounds: array and
2004pointer indexes must always be within the defined bounds of the array type.
2005The one exception for this rules is zero length arrays. These arrays are
2006defined to be accessible as variable length arrays, which requires access
2007beyond the zero'th element.</p>
2008
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2009<h5>Example:</h5>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]2010
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2011<pre>
2012 <i>; yields [12 x ubyte]*:aptr</i>
2013 %aptr = getelementptr {int, [12 x ubyte]}* %sptr, long 0, uint 1
2014</pre>
2015
2016</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2017<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2018<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2019<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]2020<p>The instructions in this category are the "miscellaneous"
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2021instructions, which defy better classification.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2022</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2023<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2024<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
2025Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2026<div class="doc_text">
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2027<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2028<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]2029<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2030<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
2031the SSA graph representing the function.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2032<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2033<p>The type of the incoming values are specified with the first type
2034field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
2035as arguments, with one pair for each predecessor basic block of the
2036current block. Only values of <a href="#t_firstclass">first class</a>
2037type may be used as the value arguments to the PHI node. Only labels
2038may be used as the label arguments.</p>
2039<p>There must be no non-phi instructions between the start of a basic
2040block and the PHI instructions: i.e. PHI instructions must be first in
2041a basic block.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2042<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2043<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the
2044value specified by the parameter, depending on which basic block we
2045came from in the last <a href="#terminators">terminator</a> instruction.</p>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2046<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2047<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]2048</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2049
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2050<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2051<div class="doc_subsubsection">
2052 <a name="i_cast">'<tt>cast .. to</tt>' Instruction</a>
2053</div>
2054
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2055<div class="doc_text">
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2056
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2057<h5>Syntax:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2058
2059<pre>
2060 &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]2061</pre>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2062
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2063<h5>Overview:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2064
2065<p>
2066The '<tt>cast</tt>' instruction is used as the primitive means to convert
2067integers to floating point, change data type sizes, and break type safety (by
2068casting pointers).
2069</p>
2070
2071
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2072<h5>Arguments:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2073
2074<p>
2075The '<tt>cast</tt>' instruction takes a value to cast, which must be a first
2076class value, and a type to cast it to, which must also be a <a
2077href="#t_firstclass">first class</a> type.
2078</p>
2079
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2080<h5>Semantics:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2081
2082<p>
2083This instruction follows the C rules for explicit casts when determining how the
2084data being cast must change to fit in its new container.
2085</p>
2086
2087<p>
2088When casting to bool, any value that would be considered true in the context of
2089a C '<tt>if</tt>' condition is converted to the boolean '<tt>true</tt>' values,
2090all else are '<tt>false</tt>'.
2091</p>
2092
2093<p>
2094When extending an integral value from a type of one signness to another (for
2095example '<tt>sbyte</tt>' to '<tt>ulong</tt>'), the value is sign-extended if the
2096<b>source</b> value is signed, and zero-extended if the source value is
2097unsigned. <tt>bool</tt> values are always zero extended into either zero or
2098one.
2099</p>
2100
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2101<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2102
2103<pre>
2104 %X = cast int 257 to ubyte <i>; yields ubyte:1</i>
Chris Lattner7bae3952002-06-25 18:03:17 +0000[diff] [blame]2105 %Y = cast int 123 to bool <i>; yields bool:true</i>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2106</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2107</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2108
2109<!-- _______________________________________________________________________ -->
2110<div class="doc_subsubsection">
2111 <a name="i_select">'<tt>select</tt>' Instruction</a>
2112</div>
2113
2114<div class="doc_text">
2115
2116<h5>Syntax:</h5>
2117
2118<pre>
2119 &lt;result&gt; = select bool &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
2120</pre>
2121
2122<h5>Overview:</h5>
2123
2124<p>
2125The '<tt>select</tt>' instruction is used to choose one value based on a
2126condition, without branching.
2127</p>
2128
2129
2130<h5>Arguments:</h5>
2131
2132<p>
2133The '<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.
2134</p>
2135
2136<h5>Semantics:</h5>
2137
2138<p>
2139If the boolean condition evaluates to true, the instruction returns the first
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2140value argument; otherwise, it returns the second value argument.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2141</p>
2142
2143<h5>Example:</h5>
2144
2145<pre>
2146 %X = select bool true, ubyte 17, ubyte 42 <i>; yields ubyte:17</i>
2147</pre>
2148</div>
2149
2150
2151
2152
2153
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2154<!-- _______________________________________________________________________ -->
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2155<div class="doc_subsubsection">
2156 <a name="i_call">'<tt>call</tt>' Instruction</a>
2157</div>
2158
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2159<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2160
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2161<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2162<pre>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2163 &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]2164</pre>
2165
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2166<h5>Overview:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2167
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2168<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2169
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2170<h5>Arguments:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2171
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2172<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2173
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2174<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2175 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2176 <p>The optional "tail" marker indicates whether the callee function accesses
2177 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2178 function call is eligible for tail call optimization. Note that calls may
2179 be marked "tail" even if they do not occur before a <a
2180 href="#i_ret"><tt>ret</tt></a> instruction.
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2181 </li>
2182 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2183 <p>The optional "cconv" marker indicates which <a href="callingconv">calling
2184 convention</a> the call should use. If none is specified, the call defaults
2185 to using C calling conventions.
2186 </li>
2187 <li>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2188 <p>'<tt>ty</tt>': shall be the signature of the pointer to function value
2189 being invoked. The argument types must match the types implied by this
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2190 signature. This type can be omitted if the function is not varargs and
2191 if the function type does not return a pointer to a function.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2192 </li>
2193 <li>
2194 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
2195 be invoked. In most cases, this is a direct function invocation, but
2196 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2197 to function value.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2198 </li>
2199 <li>
2200 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencera7e302a2005-05-01 22:22:57 +0000[diff] [blame]2201 function signature argument types. All arguments must be of
2202 <a href="#t_firstclass">first class</a> type. If the function signature
2203 indicates the function accepts a variable number of arguments, the extra
2204 arguments can be specified.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2205 </li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2206</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2207
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2208<h5>Semantics:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2209
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2210<p>The '<tt>call</tt>' instruction is used to cause control flow to
2211transfer to a specified function, with its incoming arguments bound to
2212the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
2213instruction in the called function, control flow continues with the
2214instruction after the function call, and the return value of the
2215function is bound to the result argument. This is a simpler case of
2216the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2217
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2218<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2219
2220<pre>
2221 %retval = call int %test(int %argc)
2222 call int(sbyte*, ...) *%printf(sbyte* %msg, int 12, sbyte 42);
2223 %X = tail call int %foo()
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2224 %Y = tail call <a href="#callingconv">fastcc</a> int %foo()
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2225</pre>
2226
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2227</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2228
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2229<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2230<div class="doc_subsubsection">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2231 <a name="i_vaarg">'<tt>vaarg</tt>' Instruction</a>
2232</div>
2233
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2234<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2235
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2236<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2237
2238<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2239 &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]2240</pre>
2241
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2242<h5>Overview:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2243
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2244<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2245the "variable argument" area of a function call. It is used to implement the
2246<tt>va_arg</tt> macro in C.</p>
2247
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2248<h5>Arguments:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2249
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2250<p>This instruction takes a <tt>va_list*</tt> value and the type of
2251the argument. It returns a value of the specified argument type and
2252increments the <tt>va_list</tt> to poin to the next argument. Again, the
2253actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2254
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2255<h5>Semantics:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2256
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2257<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
2258type from the specified <tt>va_list</tt> and causes the
2259<tt>va_list</tt> to point to the next argument. For more information,
2260see the variable argument handling <a href="#int_varargs">Intrinsic
2261Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2262
2263<p>It is legal for this instruction to be called in a function which does not
2264take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2265function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2266
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2267<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2268href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2269argument.</p>
2270
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2271<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2272
2273<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
2274
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2275</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2276
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2277<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2278<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
2279<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2280
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2281<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2282
2283<p>LLVM supports the notion of an "intrinsic function". These functions have
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2284well known names and semantics and are required to follow certain
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2285restrictions. Overall, these instructions represent an extension mechanism for
2286the LLVM language that does not require changing all of the transformations in
2287LLVM to add to the language (or the bytecode reader/writer, the parser,
2288etc...).</p>
2289
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2290<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
2291prefix is reserved in LLVM for intrinsic names; thus, functions may not be named
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2292this. Intrinsic functions must always be external functions: you cannot define
2293the body of intrinsic functions. Intrinsic functions may only be used in call
2294or invoke instructions: it is illegal to take the address of an intrinsic
2295function. Additionally, because intrinsic functions are part of the LLVM
2296language, it is required that they all be documented here if any are added.</p>
2297
2298
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2299<p>To learn how to add an intrinsic function, please see the <a
Chris Lattner590cff32005-05-11 03:35:57 +0000[diff] [blame]2300href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2301</p>
2302
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2303</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2304
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2305<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2306<div class="doc_subsection">
2307 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
2308</div>
2309
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2310<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2311
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2312<p>Variable argument support is defined in LLVM with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2313 href="#i_vanext"><tt>vanext</tt></a> instruction and these three
2314intrinsic functions. These functions are related to the similarly
2315named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2316
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2317<p>All of these functions operate on arguments that use a
2318target-specific value type "<tt>va_list</tt>". The LLVM assembly
2319language reference manual does not define what this type is, so all
2320transformations should be prepared to handle intrinsics with any type
2321used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2322
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2323<p>This example shows how the <a href="#i_vanext"><tt>vanext</tt></a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2324instruction and the variable argument handling intrinsic functions are
2325used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2326
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2327<pre>
2328int %test(int %X, ...) {
2329 ; Initialize variable argument processing
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2330 %ap = alloca sbyte*
2331 call void %<a href="#i_va_start">llvm.va_start</a>(sbyte** %ap)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2332
2333 ; Read a single integer argument
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2334 %tmp = va_arg sbyte** %ap, int
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2335
2336 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2337 %aq = alloca sbyte*
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2338 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]2339 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte** %aq)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2340
2341 ; Stop processing of arguments.
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2342 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte** %ap)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2343 ret int %tmp
2344}
2345</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2346</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2347
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2348<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2349<div class="doc_subsubsection">
2350 <a name="i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
2351</div>
2352
2353
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2354<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2355<h5>Syntax:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2356<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]2357<h5>Overview:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2358<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
2359<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
2360href="#i_va_arg">va_arg</a></tt>.</p>
2361
2362<h5>Arguments:</h5>
2363
2364<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
2365
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2366<h5>Semantics:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2367
2368<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
2369macro available in C. In a target-dependent way, it initializes the
2370<tt>va_list</tt> element the argument points to, so that the next call to
2371<tt>va_arg</tt> will produce the first variable argument passed to the function.
2372Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
2373last argument of the function, the compiler can figure that out.</p>
2374
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2375</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2376
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2377<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2378<div class="doc_subsubsection">
2379 <a name="i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
2380</div>
2381
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2382<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2383<h5>Syntax:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2384<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]2385<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2386<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>&lt;arglist&gt;</tt>
2387which has been initialized previously with <tt><a href="#i_va_start">llvm.va_start</a></tt>
2388or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2389<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2390<p>The argument is a <tt>va_list</tt> to destroy.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2391<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2392<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]2393macro available in C. In a target-dependent way, it destroys the <tt>va_list</tt>.
2394Calls to <a href="#i_va_start"><tt>llvm.va_start</tt></a> and <a
2395 href="#i_va_copy"><tt>llvm.va_copy</tt></a> must be matched exactly
2396with calls to <tt>llvm.va_end</tt>.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2397</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2398
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2399<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2400<div class="doc_subsubsection">
2401 <a name="i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
2402</div>
2403
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2404<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2405
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2406<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2407
2408<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2409 declare void %llvm.va_copy(&lt;va_list&gt;* &lt;destarglist&gt;,
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2410 &lt;va_list&gt;* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2411</pre>
2412
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2413<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2414
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2415<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position from
2416the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2417
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2418<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2419
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2420<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]2421The 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]2422
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2423
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2424<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2425
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2426<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt> macro
2427available in C. In a target-dependent way, it copies the source
2428<tt>va_list</tt> element into the destination list. This intrinsic is necessary
2429because 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]2430arbitrarily complex and require memory allocation, for example.</p>
2431
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2432</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2433
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2434<!-- ======================================================================= -->
2435<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2436 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
2437</div>
2438
2439<div class="doc_text">
2440
2441<p>
2442LLVM support for <a href="GarbageCollection.html">Accurate Garbage
2443Collection</a> requires the implementation and generation of these intrinsics.
2444These intrinsics allow identification of <a href="#i_gcroot">GC roots on the
2445stack</a>, as well as garbage collector implementations that require <a
2446href="#i_gcread">read</a> and <a href="#i_gcwrite">write</a> barriers.
2447Front-ends for type-safe garbage collected languages should generate these
2448intrinsics to make use of the LLVM garbage collectors. For more details, see <a
2449href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
2450</p>
2451</div>
2452
2453<!-- _______________________________________________________________________ -->
2454<div class="doc_subsubsection">
2455 <a name="i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
2456</div>
2457
2458<div class="doc_text">
2459
2460<h5>Syntax:</h5>
2461
2462<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2463 declare void %llvm.gcroot(&lt;ty&gt;** %ptrloc, &lt;ty2&gt;* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2464</pre>
2465
2466<h5>Overview:</h5>
2467
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2468<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]2469the code generator, and allows some metadata to be associated with it.</p>
2470
2471<h5>Arguments:</h5>
2472
2473<p>The first argument specifies the address of a stack object that contains the
2474root pointer. The second pointer (which must be either a constant or a global
2475value address) contains the meta-data to be associated with the root.</p>
2476
2477<h5>Semantics:</h5>
2478
2479<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
2480location. At compile-time, the code generator generates information to allow
2481the runtime to find the pointer at GC safe points.
2482</p>
2483
2484</div>
2485
2486
2487<!-- _______________________________________________________________________ -->
2488<div class="doc_subsubsection">
2489 <a name="i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
2490</div>
2491
2492<div class="doc_text">
2493
2494<h5>Syntax:</h5>
2495
2496<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2497 declare sbyte* %llvm.gcread(sbyte** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2498</pre>
2499
2500<h5>Overview:</h5>
2501
2502<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
2503locations, allowing garbage collector implementations that require read
2504barriers.</p>
2505
2506<h5>Arguments:</h5>
2507
2508<p>The argument is the address to read from, which should be an address
2509allocated from the garbage collector.</p>
2510
2511<h5>Semantics:</h5>
2512
2513<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
2514instruction, but may be replaced with substantially more complex code by the
2515garbage collector runtime, as needed.</p>
2516
2517</div>
2518
2519
2520<!-- _______________________________________________________________________ -->
2521<div class="doc_subsubsection">
2522 <a name="i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
2523</div>
2524
2525<div class="doc_text">
2526
2527<h5>Syntax:</h5>
2528
2529<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2530 declare void %llvm.gcwrite(sbyte* %P1, sbyte** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2531</pre>
2532
2533<h5>Overview:</h5>
2534
2535<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
2536locations, allowing garbage collector implementations that require write
2537barriers (such as generational or reference counting collectors).</p>
2538
2539<h5>Arguments:</h5>
2540
2541<p>The first argument is the reference to store, and the second is the heap
2542location to store to.</p>
2543
2544<h5>Semantics:</h5>
2545
2546<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
2547instruction, but may be replaced with substantially more complex code by the
2548garbage collector runtime, as needed.</p>
2549
2550</div>
2551
2552
2553
2554<!-- ======================================================================= -->
2555<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2556 <a name="int_codegen">Code Generator Intrinsics</a>
2557</div>
2558
2559<div class="doc_text">
2560<p>
2561These intrinsics are provided by LLVM to expose special features that may only
2562be implemented with code generator support.
2563</p>
2564
2565</div>
2566
2567<!-- _______________________________________________________________________ -->
2568<div class="doc_subsubsection">
2569 <a name="i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
2570</div>
2571
2572<div class="doc_text">
2573
2574<h5>Syntax:</h5>
2575<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2576 declare void* %llvm.returnaddress(uint &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2577</pre>
2578
2579<h5>Overview:</h5>
2580
2581<p>
2582The '<tt>llvm.returnaddress</tt>' intrinsic returns a target-specific value
2583indicating the return address of the current function or one of its callers.
2584</p>
2585
2586<h5>Arguments:</h5>
2587
2588<p>
2589The argument to this intrinsic indicates which function to return the address
2590for. Zero indicates the calling function, one indicates its caller, etc. The
2591argument is <b>required</b> to be a constant integer value.
2592</p>
2593
2594<h5>Semantics:</h5>
2595
2596<p>
2597The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
2598the return address of the specified call frame, or zero if it cannot be
2599identified. The value returned by this intrinsic is likely to be incorrect or 0
2600for arguments other than zero, so it should only be used for debugging purposes.
2601</p>
2602
2603<p>
2604Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]2605aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2606source-language caller.
2607</p>
2608</div>
2609
2610
2611<!-- _______________________________________________________________________ -->
2612<div class="doc_subsubsection">
2613 <a name="i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
2614</div>
2615
2616<div class="doc_text">
2617
2618<h5>Syntax:</h5>
2619<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2620 declare void* %llvm.frameaddress(uint &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2621</pre>
2622
2623<h5>Overview:</h5>
2624
2625<p>
2626The '<tt>llvm.frameaddress</tt>' intrinsic returns the target-specific frame
2627pointer value for the specified stack frame.
2628</p>
2629
2630<h5>Arguments:</h5>
2631
2632<p>
2633The argument to this intrinsic indicates which function to return the frame
2634pointer for. Zero indicates the calling function, one indicates its caller,
2635etc. The argument is <b>required</b> to be a constant integer value.
2636</p>
2637
2638<h5>Semantics:</h5>
2639
2640<p>
2641The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
2642the frame address of the specified call frame, or zero if it cannot be
2643identified. The value returned by this intrinsic is likely to be incorrect or 0
2644for arguments other than zero, so it should only be used for debugging purposes.
2645</p>
2646
2647<p>
2648Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]2649aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2650source-language caller.
2651</p>
2652</div>
2653
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2654<!-- _______________________________________________________________________ -->
2655<div class="doc_subsubsection">
2656 <a name="i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
2657</div>
2658
2659<div class="doc_text">
2660
2661<h5>Syntax:</h5>
2662<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2663 declare void %llvm.prefetch(sbyte * &lt;address&gt;,
2664 uint &lt;rw&gt;, uint &lt;locality&gt;)
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2665</pre>
2666
2667<h5>Overview:</h5>
2668
2669
2670<p>
2671The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2672a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
2673no
2674effect on the behavior of the program but can change its performance
Chris Lattner2a615362005-02-28 19:47:14 +0000[diff] [blame]2675characteristics.
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2676</p>
2677
2678<h5>Arguments:</h5>
2679
2680<p>
2681<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
2682determining if the fetch should be for a read (0) or write (1), and
2683<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattneraeffb4a2005-03-07 20:31:38 +0000[diff] [blame]2684locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2685<tt>locality</tt> arguments must be constant integers.
2686</p>
2687
2688<h5>Semantics:</h5>
2689
2690<p>
2691This intrinsic does not modify the behavior of the program. In particular,
2692prefetches cannot trap and do not produce a value. On targets that support this
2693intrinsic, the prefetch can provide hints to the processor cache for better
2694performance.
2695</p>
2696
2697</div>
2698
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2699<!-- _______________________________________________________________________ -->
2700<div class="doc_subsubsection">
2701 <a name="i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
2702</div>
2703
2704<div class="doc_text">
2705
2706<h5>Syntax:</h5>
2707<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2708 declare void %llvm.pcmarker( uint &lt;id&gt; )
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2709</pre>
2710
2711<h5>Overview:</h5>
2712
2713
2714<p>
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2715The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
2716(PC) in a region of
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2717code to simulators and other tools. The method is target specific, but it is
2718expected that the marker will use exported symbols to transmit the PC of the marker.
2719The marker makes no guaranties that it will remain with any specific instruction
2720after optimizations. It is possible that the presense of a marker will inhibit
2721optimizations. The intended use is to be inserted after optmizations to allow
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2722correlations of simulation runs.
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2723</p>
2724
2725<h5>Arguments:</h5>
2726
2727<p>
2728<tt>id</tt> is a numerical id identifying the marker.
2729</p>
2730
2731<h5>Semantics:</h5>
2732
2733<p>
2734This intrinsic does not modify the behavior of the program. Backends that do not
2735support this intrinisic may ignore it.
2736</p>
2737
2738</div>
2739
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2740
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2741<!-- ======================================================================= -->
2742<div class="doc_subsection">
2743 <a name="int_os">Operating System Intrinsics</a>
2744</div>
2745
2746<div class="doc_text">
2747<p>
2748These intrinsics are provided by LLVM to support the implementation of
2749operating system level code.
2750</p>
2751
2752</div>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2753
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2754<!-- _______________________________________________________________________ -->
2755<div class="doc_subsubsection">
2756 <a name="i_readport">'<tt>llvm.readport</tt>' Intrinsic</a>
2757</div>
2758
2759<div class="doc_text">
2760
2761<h5>Syntax:</h5>
2762<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2763 declare &lt;integer type&gt; %llvm.readport (&lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2764</pre>
2765
2766<h5>Overview:</h5>
2767
2768<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2769The '<tt>llvm.readport</tt>' intrinsic reads data from the specified hardware
2770I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2771</p>
2772
2773<h5>Arguments:</h5>
2774
2775<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2776The argument to this intrinsic indicates the hardware I/O address from which
2777to read the data. The address is in the hardware I/O address namespace (as
2778opposed to being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2779</p>
2780
2781<h5>Semantics:</h5>
2782
2783<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2784The '<tt>llvm.readport</tt>' intrinsic reads data from the hardware I/O port
2785specified by <i>address</i> and returns the value. The address and return
2786value must be integers, but the size is dependent upon the platform upon which
2787the program is code generated. For example, on x86, the address must be an
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2788unsigned 16-bit value, and the return value must be 8, 16, or 32 bits.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2789</p>
2790
2791</div>
2792
2793<!-- _______________________________________________________________________ -->
2794<div class="doc_subsubsection">
2795 <a name="i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a>
2796</div>
2797
2798<div class="doc_text">
2799
2800<h5>Syntax:</h5>
2801<pre>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2802 call void (&lt;integer type&gt;, &lt;integer type&gt;)*
2803 %llvm.writeport (&lt;integer type&gt; &lt;value&gt;,
2804 &lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2805</pre>
2806
2807<h5>Overview:</h5>
2808
2809<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2810The '<tt>llvm.writeport</tt>' intrinsic writes data to the specified hardware
2811I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2812</p>
2813
2814<h5>Arguments:</h5>
2815
2816<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2817The first argument is the value to write to the I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2818</p>
2819
2820<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2821The second argument indicates the hardware I/O address to which data should be
2822written. The address is in the hardware I/O address namespace (as opposed to
2823being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2824</p>
2825
2826<h5>Semantics:</h5>
2827
2828<p>
2829The '<tt>llvm.writeport</tt>' intrinsic writes <i>value</i> to the I/O port
2830specified by <i>address</i>. The address and value must be integers, but the
2831size is dependent upon the platform upon which the program is code generated.
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2832For example, on x86, the address must be an unsigned 16-bit value, and the
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2833value written must be 8, 16, or 32 bits in length.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2834</p>
2835
2836</div>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2837
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2838<!-- _______________________________________________________________________ -->
2839<div class="doc_subsubsection">
2840 <a name="i_readio">'<tt>llvm.readio</tt>' Intrinsic</a>
2841</div>
2842
2843<div class="doc_text">
2844
2845<h5>Syntax:</h5>
2846<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2847 declare &lt;result&gt; %llvm.readio (&lt;ty&gt; * &lt;pointer&gt;)
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2848</pre>
2849
2850<h5>Overview:</h5>
2851
2852<p>
2853The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
2854address.
2855</p>
2856
2857<h5>Arguments:</h5>
2858
2859<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2860The argument to this intrinsic is a pointer indicating the memory address from
2861which to read the data. The data must be a
2862<a href="#t_firstclass">first class</a> type.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2863</p>
2864
2865<h5>Semantics:</h5>
2866
2867<p>
2868The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2869location specified by <i>pointer</i> and returns the value. The argument must
2870be a pointer, and the return value must be a
2871<a href="#t_firstclass">first class</a> type. However, certain architectures
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2872may not support I/O on all first class types. For example, 32-bit processors
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2873may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2874</p>
2875
2876<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2877This intrinsic enforces an in-order memory model for llvm.readio and
2878llvm.writeio calls on machines that use dynamic scheduling. Dynamically
2879scheduled processors may execute loads and stores out of order, re-ordering at
2880run time accesses to memory mapped I/O registers. Using these intrinsics
2881ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2882</p>
2883
2884</div>
2885
2886<!-- _______________________________________________________________________ -->
2887<div class="doc_subsubsection">
2888 <a name="i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a>
2889</div>
2890
2891<div class="doc_text">
2892
2893<h5>Syntax:</h5>
2894<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2895 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]2896</pre>
2897
2898<h5>Overview:</h5>
2899
2900<p>
2901The '<tt>llvm.writeio</tt>' intrinsic writes data to the specified memory
2902mapped I/O address.
2903</p>
2904
2905<h5>Arguments:</h5>
2906
2907<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2908The first argument is the value to write to the memory mapped I/O location.
2909The second argument is a pointer indicating the memory address to which the
2910data should be written.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2911</p>
2912
2913<h5>Semantics:</h5>
2914
2915<p>
2916The '<tt>llvm.writeio</tt>' intrinsic writes <i>value</i> to the memory mapped
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2917I/O address specified by <i>pointer</i>. The value must be a
2918<a href="#t_firstclass">first class</a> type. However, certain architectures
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]2919may not support I/O on all first class types. For example, 32-bit processors
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2920may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2921</p>
2922
2923<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2924This intrinsic enforces an in-order memory model for llvm.readio and
2925llvm.writeio calls on machines that use dynamic scheduling. Dynamically
2926scheduled processors may execute loads and stores out of order, re-ordering at
2927run time accesses to memory mapped I/O registers. Using these intrinsics
2928ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2929</p>
2930
2931</div>
2932
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2933<!-- ======================================================================= -->
2934<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2935 <a name="int_libc">Standard C Library Intrinsics</a>
2936</div>
2937
2938<div class="doc_text">
2939<p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2940LLVM provides intrinsics for a few important standard C library functions.
2941These intrinsics allow source-language front-ends to pass information about the
2942alignment of the pointer arguments to the code generator, providing opportunity
2943for more efficient code generation.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2944</p>
2945
2946</div>
2947
2948<!-- _______________________________________________________________________ -->
2949<div class="doc_subsubsection">
2950 <a name="i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
2951</div>
2952
2953<div class="doc_text">
2954
2955<h5>Syntax:</h5>
2956<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]2957 declare void %llvm.memcpy(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
2958 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2959</pre>
2960
2961<h5>Overview:</h5>
2962
2963<p>
2964The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
2965location to the destination location.
2966</p>
2967
2968<p>
2969Note that, unlike the standard libc function, the <tt>llvm.memcpy</tt> intrinsic
2970does not return a value, and takes an extra alignment argument.
2971</p>
2972
2973<h5>Arguments:</h5>
2974
2975<p>
2976The first argument is a pointer to the destination, the second is a pointer to
2977the source. The third argument is an (arbitrarily sized) integer argument
2978specifying the number of bytes to copy, and the fourth argument is the alignment
2979of the source and destination locations.
2980</p>
2981
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]2982<p>
2983If the call to this intrinisic has an alignment value that is not 0 or 1, then
2984the caller guarantees that the size of the copy is a multiple of the alignment
2985and that both the source and destination pointers are aligned to that boundary.
2986</p>
2987
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2988<h5>Semantics:</h5>
2989
2990<p>
2991The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
2992location to the destination location, which are not allowed to overlap. It
2993copies "len" bytes of memory over. If the argument is known to be aligned to
2994some boundary, this can be specified as the fourth argument, otherwise it should
2995be set to 0 or 1.
2996</p>
2997</div>
2998
2999
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3000<!-- _______________________________________________________________________ -->
3001<div class="doc_subsubsection">
3002 <a name="i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
3003</div>
3004
3005<div class="doc_text">
3006
3007<h5>Syntax:</h5>
3008<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3009 declare void %llvm.memmove(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
3010 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3011</pre>
3012
3013<h5>Overview:</h5>
3014
3015<p>
3016The '<tt>llvm.memmove</tt>' intrinsic moves a block of memory from the source
3017location to the destination location. It is similar to the '<tt>llvm.memcpy</tt>'
3018intrinsic but allows the two memory locations to overlap.
3019</p>
3020
3021<p>
3022Note that, unlike the standard libc function, the <tt>llvm.memmove</tt> intrinsic
3023does not return a value, and takes an extra alignment argument.
3024</p>
3025
3026<h5>Arguments:</h5>
3027
3028<p>
3029The first argument is a pointer to the destination, the second is a pointer to
3030the source. The third argument is an (arbitrarily sized) integer argument
3031specifying the number of bytes to copy, and the fourth argument is the alignment
3032of the source and destination locations.
3033</p>
3034
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]3035<p>
3036If the call to this intrinisic has an alignment value that is not 0 or 1, then
3037the caller guarantees that the size of the copy is a multiple of the alignment
3038and that both the source and destination pointers are aligned to that boundary.
3039</p>
3040
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3041<h5>Semantics:</h5>
3042
3043<p>
3044The '<tt>llvm.memmove</tt>' intrinsic copies a block of memory from the source
3045location to the destination location, which may overlap. It
3046copies "len" bytes of memory over. If the argument is known to be aligned to
3047some boundary, this can be specified as the fourth argument, otherwise it should
3048be set to 0 or 1.
3049</p>
3050</div>
3051
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3052
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3053<!-- _______________________________________________________________________ -->
3054<div class="doc_subsubsection">
3055 <a name="i_memset">'<tt>llvm.memset</tt>' Intrinsic</a>
3056</div>
3057
3058<div class="doc_text">
3059
3060<h5>Syntax:</h5>
3061<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3062 declare void %llvm.memset(sbyte* &lt;dest&gt;, ubyte &lt;val&gt;,
3063 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3064</pre>
3065
3066<h5>Overview:</h5>
3067
3068<p>
3069The '<tt>llvm.memset</tt>' intrinsic fills a block of memory with a particular
3070byte value.
3071</p>
3072
3073<p>
3074Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
3075does not return a value, and takes an extra alignment argument.
3076</p>
3077
3078<h5>Arguments:</h5>
3079
3080<p>
3081The first argument is a pointer to the destination to fill, the second is the
3082byte value to fill it with, the third argument is an (arbitrarily sized) integer
3083argument specifying the number of bytes to fill, and the fourth argument is the
3084known alignment of destination location.
3085</p>
3086
3087<p>
3088If the call to this intrinisic has an alignment value that is not 0 or 1, then
3089the caller guarantees that the size of the copy is a multiple of the alignment
3090and that the destination pointer is aligned to that boundary.
3091</p>
3092
3093<h5>Semantics:</h5>
3094
3095<p>
3096The '<tt>llvm.memset</tt>' intrinsic fills "len" bytes of memory starting at the
3097destination location. If the argument is known to be aligned to some boundary,
3098this can be specified as the fourth argument, otherwise it should be set to 0 or
30991.
3100</p>
3101</div>
3102
3103
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]3104<!-- _______________________________________________________________________ -->
3105<div class="doc_subsubsection">
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]3106 <a name="i_isunordered">'<tt>llvm.isunordered</tt>' Intrinsic</a>
3107</div>
3108
3109<div class="doc_text">
3110
3111<h5>Syntax:</h5>
3112<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]3113 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]3114</pre>
3115
3116<h5>Overview:</h5>
3117
3118<p>
3119The '<tt>llvm.isunordered</tt>' intrinsic returns true if either or both of the
3120specified floating point values is a NAN.
3121</p>
3122
3123<h5>Arguments:</h5>
3124
3125<p>
3126The arguments are floating point numbers of the same type.
3127</p>
3128
3129<h5>Semantics:</h5>
3130
3131<p>
3132If either or both of the arguments is a SNAN or QNAN, it returns true, otherwise
3133false.
3134</p>
3135</div>
3136
3137
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]3138<!-- _______________________________________________________________________ -->
3139<div class="doc_subsubsection">
3140 <a name="i_sqrt">'<tt>llvm.sqrt</tt>' Intrinsic</a>
3141</div>
3142
3143<div class="doc_text">
3144
3145<h5>Syntax:</h5>
3146<pre>
3147 declare &lt;float or double&gt; %llvm.sqrt(&lt;float or double&gt; Val)
3148</pre>
3149
3150<h5>Overview:</h5>
3151
3152<p>
3153The '<tt>llvm.sqrt</tt>' intrinsic returns the sqrt of the specified operand,
3154returning the same value as the libm '<tt>sqrt</tt>' function would. Unlike
3155<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
3156negative numbers (which allows for better optimization).
3157</p>
3158
3159<h5>Arguments:</h5>
3160
3161<p>
3162The argument and return value are floating point numbers of the same type.
3163</p>
3164
3165<h5>Semantics:</h5>
3166
3167<p>
3168This function returns the sqrt of the specified operand if it is a positive
3169floating point number.
3170</p>
3171</div>
3172
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3173<!-- ======================================================================= -->
3174<div class="doc_subsection">
3175 <a name="int_count">Bit Counting Intrinsics</a>
3176</div>
3177
3178<div class="doc_text">
3179<p>
3180LLVM provides intrinsics for a few important bit counting operations.
3181These allow efficient code generation for some algorithms.
3182</p>
3183
3184</div>
3185
3186<!-- _______________________________________________________________________ -->
3187<div class="doc_subsubsection">
3188 <a name="int_ctpop">'<tt>llvm.ctpop</tt>' Intrinsic</a>
3189</div>
3190
3191<div class="doc_text">
3192
3193<h5>Syntax:</h5>
3194<pre>
3195 declare int %llvm.ctpop(int &lt;src&gt;)
3196
3197</pre>
3198
3199<h5>Overview:</h5>
3200
3201<p>
3202The '<tt>llvm.ctpop</tt>' intrinsic counts the number of ones in a variable.
3203</p>
3204
3205<h5>Arguments:</h5>
3206
3207<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3208The only argument is the value to be counted. The argument may be of any
3209integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3210</p>
3211
3212<h5>Semantics:</h5>
3213
3214<p>
3215The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
3216</p>
3217</div>
3218
3219<!-- _______________________________________________________________________ -->
3220<div class="doc_subsubsection">
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3221 <a name="int_ctlz">'<tt>llvm.ctlz</tt>' Intrinsic</a>
3222</div>
3223
3224<div class="doc_text">
3225
3226<h5>Syntax:</h5>
3227<pre>
3228 declare int %llvm.ctlz(int &lt;src&gt;)
3229
3230</pre>
3231
3232<h5>Overview:</h5>
3233
3234<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3235The '<tt>llvm.ctlz</tt>' intrinsic counts the number of leading zeros in a
3236variable.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3237</p>
3238
3239<h5>Arguments:</h5>
3240
3241<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]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.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3244</p>
3245
3246<h5>Semantics:</h5>
3247
3248<p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3249The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most 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(int 2) = 30</tt>.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3252</p>
3253</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]3254
3255
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3256
3257<!-- _______________________________________________________________________ -->
3258<div class="doc_subsubsection">
3259 <a name="int_cttz">'<tt>llvm.cttz</tt>' Intrinsic</a>
3260</div>
3261
3262<div class="doc_text">
3263
3264<h5>Syntax:</h5>
3265<pre>
3266 declare int %llvm.cttz(int &lt;src&gt;)
3267
3268</pre>
3269
3270<h5>Overview:</h5>
3271
3272<p>
3273The '<tt>llvm.cttz</tt>' intrinsic counts the number of trailing zeros.
3274</p>
3275
3276<h5>Arguments:</h5>
3277
3278<p>
3279The only argument is the value to be counted. The argument may be of any
3280integer type. The return type must match the argument type.
3281</p>
3282
3283<h5>Semantics:</h5>
3284
3285<p>
3286The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
3287in a variable. If the src == 0 then the result is the size in bits of the type
3288of src. For example, <tt>llvm.cttz(2) = 1</tt>.
3289</p>
3290</div>
3291
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3292<!-- ======================================================================= -->
3293<div class="doc_subsection">
3294 <a name="int_debugger">Debugger Intrinsics</a>
3295</div>
3296
3297<div class="doc_text">
3298<p>
3299The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
3300are described in the <a
3301href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
3302Debugging</a> document.
3303</p>
3304</div>
3305
3306
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3307<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3308<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]3309<address>
3310 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
3311 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
3312 <a href="http://validator.w3.org/check/referer"><img
3313 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
3314
3315 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
3316 <a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br>
3317 Last modified: $Date$
3318</address>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3319</body>
3320</html>