Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 0b118206bf3411722707f2e5cab8fd2eedcd50d6 / . / docs / LangRef.html
blob: 95ac99bc893e24409eaccefd0dfd7e8f4d724d11 [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>
Robert Bocchino3a558662006-01-05 17:37:02 +0000[diff] [blame]103 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]104 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]105 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]106 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></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 Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]109 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]110 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]111 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]112 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]113 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
114 <ol>
115 <li><a href="#i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
116 <li><a href="#i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
117 <li><a href="#i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
118 </ol>
119 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]120 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
121 <ol>
122 <li><a href="#i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
123 <li><a href="#i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
124 <li><a href="#i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
125 </ol>
126 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]127 <li><a href="#int_codegen">Code Generator Intrinsics</a>
128 <ol>
129 <li><a href="#i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
130 <li><a href="#i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]131 <li><a href="#i_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
132 <li><a href="#i_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]133 <li><a href="#i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]134 <li><a href="#i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]135 <li><a href="#i_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]136 </ol>
137 </li>
138 <li><a href="#int_os">Operating System Intrinsics</a>
139 <ol>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]140 <li><a href="#i_readport">'<tt>llvm.readport</tt>' Intrinsic</a></li>
141 <li><a href="#i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a></li>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]142 <li><a href="#i_readio">'<tt>llvm.readio</tt>' Intrinsic</a></li>
143 <li><a href="#i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a></li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]144 </ol>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]145 <li><a href="#int_libc">Standard C Library Intrinsics</a>
146 <ol>
147 <li><a href="#i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a></li>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]148 <li><a href="#i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a></li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]149 <li><a href="#i_memset">'<tt>llvm.memset</tt>' Intrinsic</a></li>
Alkis Evlogimenos96853722004-06-12 19:19:14 +0000[diff] [blame]150 <li><a href="#i_isunordered">'<tt>llvm.isunordered</tt>' Intrinsic</a></li>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]151 <li><a href="#i_sqrt">'<tt>llvm.sqrt</tt>' Intrinsic</a></li>
152
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]153 </ol>
154 </li>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]155 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]156 <ol>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]157 <li><a href="#i_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]158 <li><a href="#int_ctpop">'<tt>llvm.ctpop</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]159 <li><a href="#int_ctlz">'<tt>llvm.ctlz</tt>' Intrinsic </a></li>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]160 <li><a href="#int_cttz">'<tt>llvm.cttz</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]161 </ol>
162 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]163 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]164 </ol>
165 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]166</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]167
168<div class="doc_author">
169 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
170 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]171</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]172
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]173<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]174<div class="doc_section"> <a name="abstract">Abstract </a></div>
175<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]176
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]177<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]178<p>This document is a reference manual for the LLVM assembly language.
179LLVM is an SSA based representation that provides type safety,
180low-level operations, flexibility, and the capability of representing
181'all' high-level languages cleanly. It is the common code
182representation used throughout all phases of the LLVM compilation
183strategy.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]184</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]185
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]186<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]187<div class="doc_section"> <a name="introduction">Introduction</a> </div>
188<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]189
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]190<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]191
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]192<p>The LLVM code representation is designed to be used in three
193different forms: as an in-memory compiler IR, as an on-disk bytecode
194representation (suitable for fast loading by a Just-In-Time compiler),
195and as a human readable assembly language representation. This allows
196LLVM to provide a powerful intermediate representation for efficient
197compiler transformations and analysis, while providing a natural means
198to debug and visualize the transformations. The three different forms
199of LLVM are all equivalent. This document describes the human readable
200representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]201
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]202<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]203while being expressive, typed, and extensible at the same time. It
204aims to be a "universal IR" of sorts, by being at a low enough level
205that high-level ideas may be cleanly mapped to it (similar to how
206microprocessors are "universal IR's", allowing many source languages to
207be mapped to them). By providing type information, LLVM can be used as
208the target of optimizations: for example, through pointer analysis, it
209can be proven that a C automatic variable is never accessed outside of
210the current function... allowing it to be promoted to a simple SSA
211value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]212
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]213</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]214
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]215<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]216<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]217
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]218<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]219
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]220<p>It is important to note that this document describes 'well formed'
221LLVM assembly language. There is a difference between what the parser
222accepts and what is considered 'well formed'. For example, the
223following instruction is syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]224
225<pre>
226 %x = <a href="#i_add">add</a> int 1, %x
227</pre>
228
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]229<p>...because the definition of <tt>%x</tt> does not dominate all of
230its uses. The LLVM infrastructure provides a verification pass that may
231be used to verify that an LLVM module is well formed. This pass is
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]232automatically run by the parser after parsing input assembly and by
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]233the optimizer before it outputs bytecode. The violations pointed out
234by the verifier pass indicate bugs in transformation passes or input to
235the parser.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]236
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]237<!-- Describe the typesetting conventions here. --> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]238
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]239<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]240<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]241<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]242
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]243<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]244
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]245<p>LLVM uses three different forms of identifiers, for different
246purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]247
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]248<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]249 <li>Named values are represented as a string of characters with a '%' prefix.
250 For example, %foo, %DivisionByZero, %a.really.long.identifier. The actual
251 regular expression used is '<tt>%[a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
252 Identifiers which require other characters in their names can be surrounded
253 with quotes. In this way, anything except a <tt>"</tt> character can be used
254 in a name.</li>
255
256 <li>Unnamed values are represented as an unsigned numeric value with a '%'
257 prefix. For example, %12, %2, %44.</li>
258
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]259 <li>Constants, which are described in a <a href="#constants">section about
260 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]261</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]262
263<p>LLVM requires that values start with a '%' sign for two reasons: Compilers
264don't need to worry about name clashes with reserved words, and the set of
265reserved words may be expanded in the future without penalty. Additionally,
266unnamed identifiers allow a compiler to quickly come up with a temporary
267variable without having to avoid symbol table conflicts.</p>
268
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]269<p>Reserved words in LLVM are very similar to reserved words in other
270languages. There are keywords for different opcodes ('<tt><a
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]271href="#i_add">add</a></tt>', '<tt><a href="#i_cast">cast</a></tt>', '<tt><a
272href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
273href="#t_void">void</a></tt>', '<tt><a href="#t_uint">uint</a></tt>', etc...),
274and others. These reserved words cannot conflict with variable names, because
275none of them start with a '%' character.</p>
276
277<p>Here is an example of LLVM code to multiply the integer variable
278'<tt>%X</tt>' by 8:</p>
279
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]280<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]281
282<pre>
283 %result = <a href="#i_mul">mul</a> uint %X, 8
284</pre>
285
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]286<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]287
288<pre>
289 %result = <a href="#i_shl">shl</a> uint %X, ubyte 3
290</pre>
291
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]292<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]293
294<pre>
295 <a href="#i_add">add</a> uint %X, %X <i>; yields {uint}:%0</i>
296 <a href="#i_add">add</a> uint %0, %0 <i>; yields {uint}:%1</i>
297 %result = <a href="#i_add">add</a> uint %1, %1
298</pre>
299
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]300<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
301important lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]302
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]303<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]304
305 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
306 line.</li>
307
308 <li>Unnamed temporaries are created when the result of a computation is not
309 assigned to a named value.</li>
310
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]311 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]312
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]313</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]314
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]315<p>...and it also shows a convention that we follow in this document. When
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]316demonstrating instructions, we will follow an instruction with a comment that
317defines the type and name of value produced. Comments are shown in italic
318text.</p>
319
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]320</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]321
322<!-- *********************************************************************** -->
323<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
324<!-- *********************************************************************** -->
325
326<!-- ======================================================================= -->
327<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
328</div>
329
330<div class="doc_text">
331
332<p>LLVM programs are composed of "Module"s, each of which is a
333translation unit of the input programs. Each module consists of
334functions, global variables, and symbol table entries. Modules may be
335combined together with the LLVM linker, which merges function (and
336global variable) definitions, resolves forward declarations, and merges
337symbol table entries. Here is an example of the "hello world" module:</p>
338
339<pre><i>; Declare the string constant as a global constant...</i>
340<a href="#identifiers">%.LC0</a> = <a href="#linkage_internal">internal</a> <a
341 href="#globalvars">constant</a> <a href="#t_array">[13 x sbyte]</a> c"hello world\0A\00" <i>; [13 x sbyte]*</i>
342
343<i>; External declaration of the puts function</i>
344<a href="#functionstructure">declare</a> int %puts(sbyte*) <i>; int(sbyte*)* </i>
345
346<i>; Definition of main function</i>
347int %main() { <i>; int()* </i>
348 <i>; Convert [13x sbyte]* to sbyte *...</i>
349 %cast210 = <a
350 href="#i_getelementptr">getelementptr</a> [13 x sbyte]* %.LC0, long 0, long 0 <i>; sbyte*</i>
351
352 <i>; Call puts function to write out the string to stdout...</i>
353 <a
354 href="#i_call">call</a> int %puts(sbyte* %cast210) <i>; int</i>
355 <a
356 href="#i_ret">ret</a> int 0<br>}<br></pre>
357
358<p>This example is made up of a <a href="#globalvars">global variable</a>
359named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
360function, and a <a href="#functionstructure">function definition</a>
361for "<tt>main</tt>".</p>
362
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]363<p>In general, a module is made up of a list of global values,
364where both functions and global variables are global values. Global values are
365represented by a pointer to a memory location (in this case, a pointer to an
366array of char, and a pointer to a function), and have one of the following <a
367href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]368
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]369</div>
370
371<!-- ======================================================================= -->
372<div class="doc_subsection">
373 <a name="linkage">Linkage Types</a>
374</div>
375
376<div class="doc_text">
377
378<p>
379All Global Variables and Functions have one of the following types of linkage:
380</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]381
382<dl>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]383
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]384 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]385
386 <dd>Global values with internal linkage are only directly accessible by
387 objects in the current module. In particular, linking code into a module with
388 an internal global value may cause the internal to be renamed as necessary to
389 avoid collisions. Because the symbol is internal to the module, all
390 references can be updated. This corresponds to the notion of the
391 '<tt>static</tt>' keyword in C, or the idea of "anonymous namespaces" in C++.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]392 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]393
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]394 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]395
396 <dd>"<tt>linkonce</tt>" linkage is similar to <tt>internal</tt> linkage, with
397 the twist that linking together two modules defining the same
398 <tt>linkonce</tt> globals will cause one of the globals to be discarded. This
399 is typically used to implement inline functions. Unreferenced
400 <tt>linkonce</tt> globals are allowed to be discarded.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]401 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]402
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]403 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]404
405 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
406 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
407 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]408 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]409
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]410 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]411
412 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
413 pointer to array type. When two global variables with appending linkage are
414 linked together, the two global arrays are appended together. This is the
415 LLVM, typesafe, equivalent of having the system linker append together
416 "sections" with identical names when .o files are linked.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]417 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]418
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]419 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]420
421 <dd>If none of the above identifiers are used, the global is externally
422 visible, meaning that it participates in linkage and can be used to resolve
423 external symbol references.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]424 </dd>
425</dl>
426
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]427<p><a name="linkage_external">For example, since the "<tt>.LC0</tt>"
428variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
429variable and was linked with this one, one of the two would be renamed,
430preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
431external (i.e., lacking any linkage declarations), they are accessible
432outside of the current module. It is illegal for a function <i>declaration</i>
433to have any linkage type other than "externally visible".</a></p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]434
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]435</div>
436
437<!-- ======================================================================= -->
438<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]439 <a name="callingconv">Calling Conventions</a>
440</div>
441
442<div class="doc_text">
443
444<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
445and <a href="#i_invoke">invokes</a> can all have an optional calling convention
446specified for the call. The calling convention of any pair of dynamic
447caller/callee must match, or the behavior of the program is undefined. The
448following calling conventions are supported by LLVM, and more may be added in
449the future:</p>
450
451<dl>
452 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
453
454 <dd>This calling convention (the default if no other calling convention is
455 specified) matches the target C calling conventions. This calling convention
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]456 supports varargs function calls and tolerates some mismatch in the declared
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]457 prototype and implemented declaration of the function (as does normal C).
458 </dd>
459
460 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
461
462 <dd>This calling convention attempts to make calls as fast as possible
463 (e.g. by passing things in registers). This calling convention allows the
464 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattner8cdc5bc2005-05-06 23:08:23 +0000[diff] [blame]465 without having to conform to an externally specified ABI. Implementations of
466 this convention should allow arbitrary tail call optimization to be supported.
467 This calling convention does not support varargs and requires the prototype of
468 all callees to exactly match the prototype of the function definition.
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]469 </dd>
470
471 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
472
473 <dd>This calling convention attempts to make code in the caller as efficient
474 as possible under the assumption that the call is not commonly executed. As
475 such, these calls often preserve all registers so that the call does not break
476 any live ranges in the caller side. This calling convention does not support
477 varargs and requires the prototype of all callees to exactly match the
478 prototype of the function definition.
479 </dd>
480
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]481 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]482
483 <dd>Any calling convention may be specified by number, allowing
484 target-specific calling conventions to be used. Target specific calling
485 conventions start at 64.
486 </dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]487</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]488
489<p>More calling conventions can be added/defined on an as-needed basis, to
490support pascal conventions or any other well-known target-independent
491convention.</p>
492
493</div>
494
495<!-- ======================================================================= -->
496<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]497 <a name="globalvars">Global Variables</a>
498</div>
499
500<div class="doc_text">
501
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]502<p>Global variables define regions of memory allocated at compilation time
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]503instead of run-time. Global variables may optionally be initialized, may have
504an explicit section to be placed in, and may
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]505have an optional explicit alignment specified. A
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]506variable may be defined as a global "constant," which indicates that the
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]507contents of the variable will <b>never</b> be modified (enabling better
508optimization, allowing the global data to be placed in the read-only section of
509an executable, etc). Note that variables that need runtime initialization
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]510cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]511
512<p>
513LLVM explicitly allows <em>declarations</em> of global variables to be marked
514constant, even if the final definition of the global is not. This capability
515can be used to enable slightly better optimization of the program, but requires
516the language definition to guarantee that optimizations based on the
517'constantness' are valid for the translation units that do not include the
518definition.
519</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]520
521<p>As SSA values, global variables define pointer values that are in
522scope (i.e. they dominate) all basic blocks in the program. Global
523variables always define a pointer to their "content" type because they
524describe a region of memory, and all memory objects in LLVM are
525accessed through pointers.</p>
526
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]527<p>LLVM allows an explicit section to be specified for globals. If the target
528supports it, it will emit globals to the section specified.</p>
529
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]530<p>An explicit alignment may be specified for a global. If not present, or if
531the alignment is set to zero, the alignment of the global is set by the target
532to whatever it feels convenient. If an explicit alignment is specified, the
533global is forced to have at least that much alignment. All alignments must be
534a power of 2.</p>
535
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]536</div>
537
538
539<!-- ======================================================================= -->
540<div class="doc_subsection">
541 <a name="functionstructure">Functions</a>
542</div>
543
544<div class="doc_text">
545
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]546<p>LLVM function definitions consist of an optional <a href="#linkage">linkage
547type</a>, an optional <a href="#callingconv">calling convention</a>, a return
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]548type, a function name, a (possibly empty) argument list, an optional section,
549an optional alignment, an opening curly brace,
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]550a list of basic blocks, and a closing curly brace. LLVM function declarations
551are defined with the "<tt>declare</tt>" keyword, an optional <a
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]552href="#callingconv">calling convention</a>, a return type, a function name,
553a possibly empty list of arguments, and an optional alignment.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]554
555<p>A function definition contains a list of basic blocks, forming the CFG for
556the function. Each basic block may optionally start with a label (giving the
557basic block a symbol table entry), contains a list of instructions, and ends
558with a <a href="#terminators">terminator</a> instruction (such as a branch or
559function return).</p>
560
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]561<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]562executed on entrance to the function, and it is not allowed to have predecessor
563basic blocks (i.e. there can not be any branches to the entry block of a
564function). Because the block can have no predecessors, it also cannot have any
565<a href="#i_phi">PHI nodes</a>.</p>
566
567<p>LLVM functions are identified by their name and type signature. Hence, two
568functions with the same name but different parameter lists or return values are
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]569considered different functions, and LLVM will resolve references to each
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]570appropriately.</p>
571
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]572<p>LLVM allows an explicit section to be specified for functions. If the target
573supports it, it will emit functions to the section specified.</p>
574
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]575<p>An explicit alignment may be specified for a function. If not present, or if
576the alignment is set to zero, the alignment of the function is set by the target
577to whatever it feels convenient. If an explicit alignment is specified, the
578function is forced to have at least that much alignment. All alignments must be
579a power of 2.</p>
580
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]581</div>
582
583
584
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]585<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]586<div class="doc_section"> <a name="typesystem">Type System</a> </div>
587<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]588
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]589<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]590
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]591<p>The LLVM type system is one of the most important features of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]592intermediate representation. Being typed enables a number of
593optimizations to be performed on the IR directly, without having to do
594extra analyses on the side before the transformation. A strong type
595system makes it easier to read the generated code and enables novel
596analyses and transformations that are not feasible to perform on normal
597three address code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]598
599</div>
600
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]601<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]602<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]603<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]604<p>The primitive types are the fundamental building blocks of the LLVM
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]605system. The current set of primitive types is as follows:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]606
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]607<table class="layout">
608 <tr class="layout">
609 <td class="left">
610 <table>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]611 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]612 <tr><th>Type</th><th>Description</th></tr>
613 <tr><td><tt>void</tt></td><td>No value</td></tr>
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]614 <tr><td><tt>ubyte</tt></td><td>Unsigned 8-bit value</td></tr>
615 <tr><td><tt>ushort</tt></td><td>Unsigned 16-bit value</td></tr>
616 <tr><td><tt>uint</tt></td><td>Unsigned 32-bit value</td></tr>
617 <tr><td><tt>ulong</tt></td><td>Unsigned 64-bit value</td></tr>
618 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]619 <tr><td><tt>label</tt></td><td>Branch destination</td></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]620 </tbody>
621 </table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]622 </td>
623 <td class="right">
624 <table>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]625 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]626 <tr><th>Type</th><th>Description</th></tr>
627 <tr><td><tt>bool</tt></td><td>True or False value</td></tr>
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]628 <tr><td><tt>sbyte</tt></td><td>Signed 8-bit value</td></tr>
629 <tr><td><tt>short</tt></td><td>Signed 16-bit value</td></tr>
630 <tr><td><tt>int</tt></td><td>Signed 32-bit value</td></tr>
631 <tr><td><tt>long</tt></td><td>Signed 64-bit value</td></tr>
632 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]633 </tbody>
634 </table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]635 </td>
636 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]637</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]638</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]639
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]640<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]641<div class="doc_subsubsection"> <a name="t_classifications">Type
642Classifications</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]643<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]644<p>These different primitive types fall into a few useful
645classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]646
647<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]648 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]649 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]650 <tr>
651 <td><a name="t_signed">signed</a></td>
652 <td><tt>sbyte, short, int, long, float, double</tt></td>
653 </tr>
654 <tr>
655 <td><a name="t_unsigned">unsigned</a></td>
656 <td><tt>ubyte, ushort, uint, ulong</tt></td>
657 </tr>
658 <tr>
659 <td><a name="t_integer">integer</a></td>
660 <td><tt>ubyte, sbyte, ushort, short, uint, int, ulong, long</tt></td>
661 </tr>
662 <tr>
663 <td><a name="t_integral">integral</a></td>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]664 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long</tt>
665 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]666 </tr>
667 <tr>
668 <td><a name="t_floating">floating point</a></td>
669 <td><tt>float, double</tt></td>
670 </tr>
671 <tr>
672 <td><a name="t_firstclass">first class</a></td>
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]673 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long,<br>
674 float, double, <a href="#t_pointer">pointer</a>,
675 <a href="#t_packed">packed</a></tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]676 </tr>
677 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]678</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]679
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]680<p>The <a href="#t_firstclass">first class</a> types are perhaps the
681most important. Values of these types are the only ones which can be
682produced by instructions, passed as arguments, or used as operands to
683instructions. This means that all structures and arrays must be
684manipulated either by pointer or by component.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]685</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]686
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]687<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]688<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]689
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]690<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]691
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]692<p>The real power in LLVM comes from the derived types in the system.
693This is what allows a programmer to represent arrays, functions,
694pointers, and other useful types. Note that these derived types may be
695recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]696
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]697</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]698
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]699<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]700<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]701
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]702<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]703
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]704<h5>Overview:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]705
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]706<p>The array type is a very simple derived type that arranges elements
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]707sequentially in memory. The array type requires a size (number of
708elements) and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]709
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]710<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]711
712<pre>
713 [&lt;# elements&gt; x &lt;elementtype&gt;]
714</pre>
715
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]716<p>The number of elements is a constant integer value; elementtype may
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]717be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]718
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]719<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]720<table class="layout">
721 <tr class="layout">
722 <td class="left">
723 <tt>[40 x int ]</tt><br/>
724 <tt>[41 x int ]</tt><br/>
725 <tt>[40 x uint]</tt><br/>
726 </td>
727 <td class="left">
728 Array of 40 integer values.<br/>
729 Array of 41 integer values.<br/>
730 Array of 40 unsigned integer values.<br/>
731 </td>
732 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]733</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]734<p>Here are some examples of multidimensional arrays:</p>
735<table class="layout">
736 <tr class="layout">
737 <td class="left">
738 <tt>[3 x [4 x int]]</tt><br/>
739 <tt>[12 x [10 x float]]</tt><br/>
740 <tt>[2 x [3 x [4 x uint]]]</tt><br/>
741 </td>
742 <td class="left">
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]743 3x4 array of integer values.<br/>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]744 12x10 array of single precision floating point values.<br/>
745 2x3x4 array of unsigned integer values.<br/>
746 </td>
747 </tr>
748</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]749
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]750<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
751length array. Normally, accesses past the end of an array are undefined in
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]752LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
753As a special case, however, zero length arrays are recognized to be variable
754length. This allows implementation of 'pascal style arrays' with the LLVM
755type "{ int, [0 x float]}", for example.</p>
756
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]757</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]758
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]759<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]760<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]761<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]762<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]763<p>The function type can be thought of as a function signature. It
764consists of a return type and a list of formal parameter types.
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]765Function types are usually used to build virtual function tables
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]766(which are structures of pointers to functions), for indirect function
767calls, and when defining a function.</p>
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]768<p>
769The return type of a function type cannot be an aggregate type.
770</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]771<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]772<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]773<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]774specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]775which indicates that the function takes a variable number of arguments.
776Variable argument functions can access their arguments with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]777 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]778<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]779<table class="layout">
780 <tr class="layout">
781 <td class="left">
782 <tt>int (int)</tt> <br/>
783 <tt>float (int, int *) *</tt><br/>
784 <tt>int (sbyte *, ...)</tt><br/>
785 </td>
786 <td class="left">
787 function taking an <tt>int</tt>, returning an <tt>int</tt><br/>
788 <a href="#t_pointer">Pointer</a> to a function that takes an
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]789 <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]790 returning <tt>float</tt>.<br/>
791 A vararg function that takes at least one <a href="#t_pointer">pointer</a>
792 to <tt>sbyte</tt> (signed char in C), which returns an integer. This is
793 the signature for <tt>printf</tt> in LLVM.<br/>
794 </td>
795 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]796</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]797
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]798</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]799<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]800<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]801<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]802<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]803<p>The structure type is used to represent a collection of data members
804together in memory. The packing of the field types is defined to match
805the ABI of the underlying processor. The elements of a structure may
806be any type that has a size.</p>
807<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
808and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
809field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
810instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]811<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]812<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]813<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]814<table class="layout">
815 <tr class="layout">
816 <td class="left">
817 <tt>{ int, int, int }</tt><br/>
818 <tt>{ float, int (int) * }</tt><br/>
819 </td>
820 <td class="left">
821 a triple of three <tt>int</tt> values<br/>
822 A pair, where the first element is a <tt>float</tt> and the second element
823 is a <a href="#t_pointer">pointer</a> to a <a href="#t_function">function</a>
824 that takes an <tt>int</tt>, returning an <tt>int</tt>.<br/>
825 </td>
826 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]827</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]828</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]829
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]830<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]831<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]832<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]833<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]834<p>As in many languages, the pointer type represents a pointer or
835reference to another object, which must live in memory.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]836<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]837<pre> &lt;type&gt; *<br></pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]838<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]839<table class="layout">
840 <tr class="layout">
841 <td class="left">
842 <tt>[4x int]*</tt><br/>
843 <tt>int (int *) *</tt><br/>
844 </td>
845 <td class="left">
846 A <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a> of
847 four <tt>int</tt> values<br/>
848 A <a href="#t_pointer">pointer</a> to a <a
Chris Lattnera977c482005-02-19 02:22:14 +0000[diff] [blame]849 href="#t_function">function</a> that takes an <tt>int*</tt>, returning an
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]850 <tt>int</tt>.<br/>
851 </td>
852 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]853</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]854</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]855
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]856<!-- _______________________________________________________________________ -->
857<div class="doc_subsubsection"> <a name="t_packed">Packed Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]858<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]859
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]860<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]861
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]862<p>A packed type is a simple derived type that represents a vector
863of elements. Packed types are used when multiple primitive data
864are operated in parallel using a single instruction (SIMD).
865A packed type requires a size (number of
Chris Lattnerb8d172f2005-11-10 01:44:22 +0000[diff] [blame]866elements) and an underlying primitive data type. Vectors must have a power
867of two length (1, 2, 4, 8, 16 ...). Packed types are
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]868considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]869
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]870<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]871
872<pre>
873 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
874</pre>
875
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]876<p>The number of elements is a constant integer value; elementtype may
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]877be any integral or floating point type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]878
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]879<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]880
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]881<table class="layout">
882 <tr class="layout">
883 <td class="left">
884 <tt>&lt;4 x int&gt;</tt><br/>
885 <tt>&lt;8 x float&gt;</tt><br/>
886 <tt>&lt;2 x uint&gt;</tt><br/>
887 </td>
888 <td class="left">
889 Packed vector of 4 integer values.<br/>
890 Packed vector of 8 floating-point values.<br/>
891 Packed vector of 2 unsigned integer values.<br/>
892 </td>
893 </tr>
894</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]895</div>
896
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]897<!-- _______________________________________________________________________ -->
898<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
899<div class="doc_text">
900
901<h5>Overview:</h5>
902
903<p>Opaque types are used to represent unknown types in the system. This
904corresponds (for example) to the C notion of a foward declared structure type.
905In LLVM, opaque types can eventually be resolved to any type (not just a
906structure type).</p>
907
908<h5>Syntax:</h5>
909
910<pre>
911 opaque
912</pre>
913
914<h5>Examples:</h5>
915
916<table class="layout">
917 <tr class="layout">
918 <td class="left">
919 <tt>opaque</tt>
920 </td>
921 <td class="left">
922 An opaque type.<br/>
923 </td>
924 </tr>
925</table>
926</div>
927
928
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]929<!-- *********************************************************************** -->
930<div class="doc_section"> <a name="constants">Constants</a> </div>
931<!-- *********************************************************************** -->
932
933<div class="doc_text">
934
935<p>LLVM has several different basic types of constants. This section describes
936them all and their syntax.</p>
937
938</div>
939
940<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]941<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]942
943<div class="doc_text">
944
945<dl>
946 <dt><b>Boolean constants</b></dt>
947
948 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
949 constants of the <tt><a href="#t_primitive">bool</a></tt> type.
950 </dd>
951
952 <dt><b>Integer constants</b></dt>
953
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]954 <dd>Standard integers (such as '4') are constants of the <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]955 href="#t_integer">integer</a> type. Negative numbers may be used with signed
956 integer types.
957 </dd>
958
959 <dt><b>Floating point constants</b></dt>
960
961 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
962 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]963 notation (see below). Floating point constants must have a <a
964 href="#t_floating">floating point</a> type. </dd>
965
966 <dt><b>Null pointer constants</b></dt>
967
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]968 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]969 and must be of <a href="#t_pointer">pointer type</a>.</dd>
970
971</dl>
972
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]973<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]974of floating point constants. For example, the form '<tt>double
9750x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
9764.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]977(and the only time that they are generated by the disassembler) is when a
978floating point constant must be emitted but it cannot be represented as a
979decimal floating point number. For example, NaN's, infinities, and other
980special values are represented in their IEEE hexadecimal format so that
981assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]982
983</div>
984
985<!-- ======================================================================= -->
986<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
987</div>
988
989<div class="doc_text">
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]990<p>Aggregate constants arise from aggregation of simple constants
991and smaller aggregate constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]992
993<dl>
994 <dt><b>Structure constants</b></dt>
995
996 <dd>Structure constants are represented with notation similar to structure
997 type definitions (a comma separated list of elements, surrounded by braces
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]998 (<tt>{}</tt>)). For example: "<tt>{ int 4, float 17.0, int* %G }</tt>",
999 where "<tt>%G</tt>" is declared as "<tt>%G = external global int</tt>". Structure constants
1000 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1001 types of elements must match those specified by the type.
1002 </dd>
1003
1004 <dt><b>Array constants</b></dt>
1005
1006 <dd>Array constants are represented with notation similar to array type
1007 definitions (a comma separated list of elements, surrounded by square brackets
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1008 (<tt>[]</tt>)). For example: "<tt>[ int 42, int 11, int 74 ]</tt>". Array
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1009 constants must have <a href="#t_array">array type</a>, and the number and
1010 types of elements must match those specified by the type.
1011 </dd>
1012
1013 <dt><b>Packed constants</b></dt>
1014
1015 <dd>Packed constants are represented with notation similar to packed type
1016 definitions (a comma separated list of elements, surrounded by
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1017 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]1018 int 11, int 74, int 100 &gt;</tt>". Packed constants must have <a
1019 href="#t_packed">packed type</a>, and the number and types of elements must
1020 match those specified by the type.
1021 </dd>
1022
1023 <dt><b>Zero initialization</b></dt>
1024
1025 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
1026 value to zero of <em>any</em> type, including scalar and aggregate types.
1027 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]1028 large arrays) and is always exactly equivalent to using explicit zero
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1029 initializers.
1030 </dd>
1031</dl>
1032
1033</div>
1034
1035<!-- ======================================================================= -->
1036<div class="doc_subsection">
1037 <a name="globalconstants">Global Variable and Function Addresses</a>
1038</div>
1039
1040<div class="doc_text">
1041
1042<p>The addresses of <a href="#globalvars">global variables</a> and <a
1043href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1044constants. These constants are explicitly referenced when the <a
1045href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1046href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1047file:</p>
1048
1049<pre>
1050 %X = global int 17
1051 %Y = global int 42
1052 %Z = global [2 x int*] [ int* %X, int* %Y ]
1053</pre>
1054
1055</div>
1056
1057<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1058<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1059<div class="doc_text">
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1060 <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]1061 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1062 a constant is permitted.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1063
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1064 <p>Undefined values indicate to the compiler that the program is well defined
1065 no matter what value is used, giving the compiler more freedom to optimize.
1066 </p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1067</div>
1068
1069<!-- ======================================================================= -->
1070<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1071</div>
1072
1073<div class="doc_text">
1074
1075<p>Constant expressions are used to allow expressions involving other constants
1076to be used as constants. Constant expressions may be of any <a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1077href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1078that does not have side effects (e.g. load and call are not supported). The
1079following is the syntax for constant expressions:</p>
1080
1081<dl>
1082 <dt><b><tt>cast ( CST to TYPE )</tt></b></dt>
1083
1084 <dd>Cast a constant to another type.</dd>
1085
1086 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1087
1088 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1089 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1090 instruction, the index list may have zero or more indexes, which are required
1091 to make sense for the type of "CSTPTR".</dd>
1092
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1093 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
1094
1095 <dd>Perform the <a href="#i_select">select operation</a> on
1096 constants.
1097
1098 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
1099
1100 <dd>Perform the <a href="#i_extractelement">extractelement
1101 operation</a> on constants.
1102
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]1103 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
1104
1105 <dd>Perform the <a href="#i_insertelement">insertelement
1106 operation</a> on constants.
1107
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1108 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1109
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1110 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1111 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1112 binary</a> operations. The constraints on operands are the same as those for
1113 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1114 values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1115</dl>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1116</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]1117
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1118<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1119<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1120<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1121
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1122<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1123
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1124<p>The LLVM instruction set consists of several different
1125classifications of instructions: <a href="#terminators">terminator
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1126instructions</a>, <a href="#binaryops">binary instructions</a>,
1127<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1128 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1129instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1130
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1131</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1132
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1133<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1134<div class="doc_subsection"> <a name="terminators">Terminator
1135Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1136
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1137<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1138
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1139<p>As mentioned <a href="#functionstructure">previously</a>, every
1140basic block in a program ends with a "Terminator" instruction, which
1141indicates which block should be executed after the current block is
1142finished. These terminator instructions typically yield a '<tt>void</tt>'
1143value: they produce control flow, not values (the one exception being
1144the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1145<p>There are six different terminator instructions: the '<a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1146 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1147instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1148the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1149 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1150 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1151
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1152</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1153
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1154<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1155<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1156Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1157<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1158<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1159<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]1160 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1161</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1162<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1163<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1164value) from a function back to the caller.</p>
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]1165<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1166returns a value and then causes control flow, and one that just causes
1167control flow to occur.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1168<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1169<p>The '<tt>ret</tt>' instruction may return any '<a
1170 href="#t_firstclass">first class</a>' type. Notice that a function is
1171not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1172instruction inside of the function that returns a value that does not
1173match the return type of the function.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1174<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1175<p>When the '<tt>ret</tt>' instruction is executed, control flow
1176returns back to the calling function's context. If the caller is a "<a
John Criswellfa081872004-06-25 15:16:57 +0000[diff] [blame]1177 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1178the instruction after the call. If the caller was an "<a
1179 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1180at the beginning of the "normal" destination block. If the instruction
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1181returns a value, that value shall set the call or invoke instruction's
1182return value.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1183<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1184<pre> ret int 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1185 ret void <i>; Return from a void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1186</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1187</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1188<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1189<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1190<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1191<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1192<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]1193</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1194<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1195<p>The '<tt>br</tt>' instruction is used to cause control flow to
1196transfer to a different basic block in the current function. There are
1197two forms of this instruction, corresponding to a conditional branch
1198and an unconditional branch.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1199<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1200<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
1201single '<tt>bool</tt>' value and two '<tt>label</tt>' values. The
1202unconditional form of the '<tt>br</tt>' instruction takes a single '<tt>label</tt>'
1203value as a target.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1204<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1205<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>bool</tt>'
1206argument is evaluated. If the value is <tt>true</tt>, control flows
1207to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1208control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1209<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1210<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
1211 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]1212</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1213<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1214<div class="doc_subsubsection">
1215 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1216</div>
1217
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1218<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1219<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1220
1221<pre>
1222 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1223</pre>
1224
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1225<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1226
1227<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1228several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1229instruction, allowing a branch to occur to one of many possible
1230destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1231
1232
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1233<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1234
1235<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1236comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1237an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1238table is not allowed to contain duplicate constant entries.</p>
1239
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1240<h5>Semantics:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1241
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1242<p>The <tt>switch</tt> instruction specifies a table of values and
1243destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1244table is searched for the given value. If the value is found, control flow is
1245transfered to the corresponding destination; otherwise, control flow is
1246transfered to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1247
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1248<h5>Implementation:</h5>
1249
1250<p>Depending on properties of the target machine and the particular
1251<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1252ways. For example, it could be generated as a series of chained conditional
1253branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1254
1255<h5>Example:</h5>
1256
1257<pre>
1258 <i>; Emulate a conditional br instruction</i>
1259 %Val = <a href="#i_cast">cast</a> bool %value to int
1260 switch int %Val, label %truedest [int 0, label %falsedest ]
1261
1262 <i>; Emulate an unconditional br instruction</i>
1263 switch uint 0, label %dest [ ]
1264
1265 <i>; Implement a jump table:</i>
1266 switch uint %val, label %otherwise [ uint 0, label %onzero
1267 uint 1, label %onone
1268 uint 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1269</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1270</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1271
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1272<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1273<div class="doc_subsubsection">
1274 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1275</div>
1276
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1277<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1278
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1279<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1280
1281<pre>
1282 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)
1283 to label &lt;normal label&gt; except label &lt;exception label&gt;
1284</pre>
1285
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1286<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1287
1288<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1289function, with the possibility of control flow transfer to either the
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1290'<tt>normal</tt>' label or the
1291'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1292"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1293"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1294href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1295continued at the dynamically nearest "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1296
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1297<h5>Arguments:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1298
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1299<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1300
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1301<ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1302 <li>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1303 The optional "cconv" marker indicates which <a href="callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1304 convention</a> the call should use. If none is specified, the call defaults
1305 to using C calling conventions.
1306 </li>
1307 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1308 function value being invoked. In most cases, this is a direct function
1309 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1310 an arbitrary pointer to function value.
1311 </li>
1312
1313 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1314 function to be invoked. </li>
1315
1316 <li>'<tt>function args</tt>': argument list whose types match the function
1317 signature argument types. If the function signature indicates the function
1318 accepts a variable number of arguments, the extra arguments can be
1319 specified. </li>
1320
1321 <li>'<tt>normal label</tt>': the label reached when the called function
1322 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1323
1324 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1325 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1326
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1327</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1328
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1329<h5>Semantics:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1330
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1331<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1332href="#i_call">call</a></tt>' instruction in most regards. The primary
1333difference is that it establishes an association with a label, which is used by
1334the runtime library to unwind the stack.</p>
1335
1336<p>This instruction is used in languages with destructors to ensure that proper
1337cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1338exception. Additionally, this is important for implementation of
1339'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1340
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1341<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1342<pre>
1343 %retval = invoke int %Test(int 15) to label %Continue
1344 except label %TestCleanup <i>; {int}:retval set</i>
1345 %retval = invoke <a href="#callingconv">coldcc</a> int %Test(int 15) to label %Continue
1346 except label %TestCleanup <i>; {int}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1347</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1348</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1349
1350
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1351<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1352
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1353<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1354Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1355
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1356<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1357
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1358<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1359<pre>
1360 unwind
1361</pre>
1362
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1363<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1364
1365<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1366at the first callee in the dynamic call stack which used an <a
1367href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1368primarily used to implement exception handling.</p>
1369
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1370<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1371
1372<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
1373immediately halt. The dynamic call stack is then searched for the first <a
1374href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
1375execution continues at the "exceptional" destination block specified by the
1376<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
1377dynamic call chain, undefined behavior results.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1378</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1379
1380<!-- _______________________________________________________________________ -->
1381
1382<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
1383Instruction</a> </div>
1384
1385<div class="doc_text">
1386
1387<h5>Syntax:</h5>
1388<pre>
1389 unreachable
1390</pre>
1391
1392<h5>Overview:</h5>
1393
1394<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
1395instruction is used to inform the optimizer that a particular portion of the
1396code is not reachable. This can be used to indicate that the code after a
1397no-return function cannot be reached, and other facts.</p>
1398
1399<h5>Semantics:</h5>
1400
1401<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
1402</div>
1403
1404
1405
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1406<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1407<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1408<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1409<p>Binary operators are used to do most of the computation in a
1410program. They require two operands, execute an operation on them, and
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1411produce a single value. The operands might represent
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1412multiple data, as is the case with the <a href="#t_packed">packed</a> data type.
1413The result value of a binary operator is not
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1414necessarily the same type as its operands.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1415<p>There are several different binary operators:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1416</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1417<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1418<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
1419Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1420<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1421<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1422<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]1423</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1424<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1425<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1426<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1427<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1428 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
1429 This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1430Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1431<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1432<p>The value produced is the integer or floating point sum of the two
1433operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1434<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1435<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]1436</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1437</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1438<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1439<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
1440Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1441<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1442<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1443<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]1444</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1445<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1446<p>The '<tt>sub</tt>' instruction returns the difference of its two
1447operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1448<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
1449instruction present in most other intermediate representations.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1450<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1451<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1452 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1453values.
1454This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1455Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1456<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1457<p>The value produced is the integer or floating point difference of
1458the two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1459<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1460<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]1461 &lt;result&gt; = sub int 0, %val <i>; yields {int}:result = -%var</i>
1462</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1463</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1464<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1465<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
1466Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1467<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1468<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1469<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]1470</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1471<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1472<p>The '<tt>mul</tt>' instruction returns the product of its two
1473operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1474<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1475<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1476 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1477values.
1478This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1479Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1480<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1481<p>The value produced is the integer or floating point product of the
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1482two operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1483<p>There is no signed vs unsigned multiplication. The appropriate
1484action is taken based on the type of the operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1485<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1486<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]1487</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1488</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1489<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1490<div class="doc_subsubsection"> <a name="i_div">'<tt>div</tt>'
1491Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1492<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1493<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1494<pre> &lt;result&gt; = div &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1495</pre>
1496<h5>Overview:</h5>
1497<p>The '<tt>div</tt>' instruction returns the quotient of its two
1498operands.</p>
1499<h5>Arguments:</h5>
1500<p>The two arguments to the '<tt>div</tt>' instruction must be either <a
1501 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1502values.
1503This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1504Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1505<h5>Semantics:</h5>
1506<p>The value produced is the integer or floating point quotient of the
1507two operands.</p>
1508<h5>Example:</h5>
1509<pre> &lt;result&gt; = div int 4, %var <i>; yields {int}:result = 4 / %var</i>
1510</pre>
1511</div>
1512<!-- _______________________________________________________________________ -->
1513<div class="doc_subsubsection"> <a name="i_rem">'<tt>rem</tt>'
1514Instruction</a> </div>
1515<div class="doc_text">
1516<h5>Syntax:</h5>
1517<pre> &lt;result&gt; = rem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1518</pre>
1519<h5>Overview:</h5>
1520<p>The '<tt>rem</tt>' instruction returns the remainder from the
1521division of its two operands.</p>
1522<h5>Arguments:</h5>
1523<p>The two arguments to the '<tt>rem</tt>' instruction must be either <a
1524 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1525values.
1526This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1527Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1528<h5>Semantics:</h5>
1529<p>This returns the <i>remainder</i> of a division (where the result
1530has the same sign as the divisor), not the <i>modulus</i> (where the
1531result has the same sign as the dividend) of a value. For more
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1532information about the difference, see <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1533 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
1534Math Forum</a>.</p>
1535<h5>Example:</h5>
1536<pre> &lt;result&gt; = rem int 4, %var <i>; yields {int}:result = 4 % %var</i>
1537</pre>
1538</div>
1539<!-- _______________________________________________________________________ -->
1540<div class="doc_subsubsection"> <a name="i_setcc">'<tt>set<i>cc</i></tt>'
1541Instructions</a> </div>
1542<div class="doc_text">
1543<h5>Syntax:</h5>
1544<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]1545 &lt;result&gt; = setne &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1546 &lt;result&gt; = setlt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1547 &lt;result&gt; = setgt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1548 &lt;result&gt; = setle &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1549 &lt;result&gt; = setge &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1550</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1551<h5>Overview:</h5>
1552<p>The '<tt>set<i>cc</i></tt>' family of instructions returns a boolean
1553value based on a comparison of their two operands.</p>
1554<h5>Arguments:</h5>
1555<p>The two arguments to the '<tt>set<i>cc</i></tt>' instructions must
1556be of <a href="#t_firstclass">first class</a> type (it is not possible
1557to compare '<tt>label</tt>'s, '<tt>array</tt>'s, '<tt>structure</tt>'
1558or '<tt>void</tt>' values, etc...). Both arguments must have identical
1559types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1560<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1561<p>The '<tt>seteq</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1562value if both operands are equal.<br>
1563The '<tt>setne</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1564value if both operands are unequal.<br>
1565The '<tt>setlt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1566value if the first operand is less than the second operand.<br>
1567The '<tt>setgt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1568value if the first operand is greater than the second operand.<br>
1569The '<tt>setle</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1570value if the first operand is less than or equal to the second operand.<br>
1571The '<tt>setge</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1572value if the first operand is greater than or equal to the second
1573operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1574<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1575<pre> &lt;result&gt; = seteq int 4, 5 <i>; yields {bool}:result = false</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1576 &lt;result&gt; = setne float 4, 5 <i>; yields {bool}:result = true</i>
1577 &lt;result&gt; = setlt uint 4, 5 <i>; yields {bool}:result = true</i>
1578 &lt;result&gt; = setgt sbyte 4, 5 <i>; yields {bool}:result = false</i>
1579 &lt;result&gt; = setle sbyte 4, 5 <i>; yields {bool}:result = true</i>
1580 &lt;result&gt; = setge sbyte 4, 5 <i>; yields {bool}:result = false</i>
1581</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1582</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1583<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1584<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
1585Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1586<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1587<p>Bitwise binary operators are used to do various forms of
1588bit-twiddling in a program. They are generally very efficient
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1589instructions and can commonly be strength reduced from other
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1590instructions. They require two operands, execute an operation on them,
1591and produce a single value. The resulting value of the bitwise binary
1592operators is always the same type as its first operand.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1593</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1594<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1595<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
1596Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1597<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1598<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1599<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]1600</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1601<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1602<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
1603its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1604<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1605<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1606 href="#t_integral">integral</a> values. Both arguments must have
1607identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1608<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1609<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1610<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1611<div style="align: center">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1612<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1613 <tbody>
1614 <tr>
1615 <td>In0</td>
1616 <td>In1</td>
1617 <td>Out</td>
1618 </tr>
1619 <tr>
1620 <td>0</td>
1621 <td>0</td>
1622 <td>0</td>
1623 </tr>
1624 <tr>
1625 <td>0</td>
1626 <td>1</td>
1627 <td>0</td>
1628 </tr>
1629 <tr>
1630 <td>1</td>
1631 <td>0</td>
1632 <td>0</td>
1633 </tr>
1634 <tr>
1635 <td>1</td>
1636 <td>1</td>
1637 <td>1</td>
1638 </tr>
1639 </tbody>
1640</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1641</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1642<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1643<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]1644 &lt;result&gt; = and int 15, 40 <i>; yields {int}:result = 8</i>
1645 &lt;result&gt; = and int 4, 8 <i>; yields {int}:result = 0</i>
1646</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1647</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1648<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1649<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1650<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1651<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1652<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]1653</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1654<h5>Overview:</h5>
1655<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
1656or of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1657<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1658<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1659 href="#t_integral">integral</a> values. Both arguments must have
1660identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1661<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1662<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1663<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1664<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1665<table border="1" cellspacing="0" cellpadding="4">
1666 <tbody>
1667 <tr>
1668 <td>In0</td>
1669 <td>In1</td>
1670 <td>Out</td>
1671 </tr>
1672 <tr>
1673 <td>0</td>
1674 <td>0</td>
1675 <td>0</td>
1676 </tr>
1677 <tr>
1678 <td>0</td>
1679 <td>1</td>
1680 <td>1</td>
1681 </tr>
1682 <tr>
1683 <td>1</td>
1684 <td>0</td>
1685 <td>1</td>
1686 </tr>
1687 <tr>
1688 <td>1</td>
1689 <td>1</td>
1690 <td>1</td>
1691 </tr>
1692 </tbody>
1693</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1694</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1695<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1696<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]1697 &lt;result&gt; = or int 15, 40 <i>; yields {int}:result = 47</i>
1698 &lt;result&gt; = or int 4, 8 <i>; yields {int}:result = 12</i>
1699</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1700</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1701<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1702<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
1703Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1704<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1705<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1706<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]1707</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1708<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1709<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
1710or of its two operands. The <tt>xor</tt> is used to implement the
1711"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1712<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1713<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1714 href="#t_integral">integral</a> values. Both arguments must have
1715identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1716<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1717<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1718<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1719<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1720<table border="1" cellspacing="0" cellpadding="4">
1721 <tbody>
1722 <tr>
1723 <td>In0</td>
1724 <td>In1</td>
1725 <td>Out</td>
1726 </tr>
1727 <tr>
1728 <td>0</td>
1729 <td>0</td>
1730 <td>0</td>
1731 </tr>
1732 <tr>
1733 <td>0</td>
1734 <td>1</td>
1735 <td>1</td>
1736 </tr>
1737 <tr>
1738 <td>1</td>
1739 <td>0</td>
1740 <td>1</td>
1741 </tr>
1742 <tr>
1743 <td>1</td>
1744 <td>1</td>
1745 <td>0</td>
1746 </tr>
1747 </tbody>
1748</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1749</div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1750<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1751<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1752<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]1753 &lt;result&gt; = xor int 15, 40 <i>; yields {int}:result = 39</i>
1754 &lt;result&gt; = xor int 4, 8 <i>; yields {int}:result = 12</i>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1755 &lt;result&gt; = xor int %V, -1 <i>; yields {int}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1756</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1757</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1758<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1759<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
1760Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1761<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1762<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1763<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]1764</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1765<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1766<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
1767the left a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1768<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1769<p>The first argument to the '<tt>shl</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1770 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1771type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1772<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1773<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]1774<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1775<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]1776 &lt;result&gt; = shl int 4, ubyte 2 <i>; yields {int}:result = 16</i>
1777 &lt;result&gt; = shl int 1, ubyte 10 <i>; yields {int}:result = 1024</i>
1778</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1779</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1780<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1781<div class="doc_subsubsection"> <a name="i_shr">'<tt>shr</tt>'
1782Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1783<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1784<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1785<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]1786</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1787<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1788<p>The '<tt>shr</tt>' instruction returns the first operand shifted to
1789the right a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1790<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1791<p>The first argument to the '<tt>shr</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1792 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1793type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1794<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1795<p>If the first argument is a <a href="#t_signed">signed</a> type, the
1796most significant bit is duplicated in the newly free'd bit positions.
1797If the first argument is unsigned, zero bits shall fill the empty
1798positions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1799<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1800<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]1801 &lt;result&gt; = shr uint 4, ubyte 1 <i>; yields {uint}:result = 2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1802 &lt;result&gt; = shr int 4, ubyte 2 <i>; yields {int}:result = 1</i>
Chris Lattner8c6bb902003-06-18 21:30:51 +0000[diff] [blame]1803 &lt;result&gt; = shr sbyte 4, ubyte 3 <i>; yields {sbyte}:result = 0</i>
1804 &lt;result&gt; = shr sbyte -2, ubyte 1 <i>; yields {sbyte}:result = -1</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1805</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1806</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1807
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1808<!-- ======================================================================= -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1809<div class="doc_subsection">
1810 <a name="memoryops">Memory Access Operations</a>
1811</div>
1812
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1813<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1814
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1815<p>A key design point of an SSA-based representation is how it
1816represents memory. In LLVM, no memory locations are in SSA form, which
1817makes things very simple. This section describes how to read, write,
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1818allocate, and free memory in LLVM.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1819
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1820</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1821
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1822<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1823<div class="doc_subsubsection">
1824 <a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
1825</div>
1826
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1827<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1828
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1829<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1830
1831<pre>
1832 &lt;result&gt; = malloc &lt;type&gt;[, uint &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1833</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1834
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1835<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1836
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1837<p>The '<tt>malloc</tt>' instruction allocates memory from the system
1838heap and returns a pointer to it.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1839
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1840<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1841
1842<p>The '<tt>malloc</tt>' instruction allocates
1843<tt>sizeof(&lt;type&gt;)*NumElements</tt>
John Criswell6e4ca612004-02-24 16:13:56 +0000[diff] [blame]1844bytes of memory from the operating system and returns a pointer of the
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1845appropriate type to the program. If "NumElements" is specified, it is the
1846number of elements allocated. If an alignment is specified, the value result
1847of the allocation is guaranteed to be aligned to at least that boundary. If
1848not specified, or if zero, the target can choose to align the allocation on any
1849convenient boundary.</p>
1850
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1851<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1852
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1853<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1854
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1855<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
1856a pointer is returned.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1857
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1858<h5>Example:</h5>
1859
1860<pre>
1861 %array = malloc [4 x ubyte ] <i>; yields {[%4 x ubyte]*}:array</i>
1862
1863 %size = <a 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]1864 %array1 = malloc ubyte, uint 4 <i>; yields {ubyte*}:array1</i>
1865 %array2 = malloc [12 x ubyte], uint %size <i>; yields {[12 x ubyte]*}:array2</i>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1866 %array3 = malloc int, uint 4, align 1024 <i>; yields {int*}:array3</i>
1867 %array4 = malloc int, align 1024 <i>; yields {int*}:array4</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1868</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1869</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1870
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1871<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1872<div class="doc_subsubsection">
1873 <a name="i_free">'<tt>free</tt>' Instruction</a>
1874</div>
1875
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1876<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1877
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1878<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1879
1880<pre>
1881 free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1882</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1883
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1884<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1885
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1886<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1887memory heap to be reallocated in the future.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1888
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1889<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1890
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1891<p>'<tt>value</tt>' shall be a pointer value that points to a value
1892that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
1893instruction.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1894
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1895<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1896
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1897<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1898after this instruction executes.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1899
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1900<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1901
1902<pre>
1903 %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]1904 free [4 x ubyte]* %array
1905</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1906</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1907
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1908<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1909<div class="doc_subsubsection">
1910 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
1911</div>
1912
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1913<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1914
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1915<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1916
1917<pre>
1918 &lt;result&gt; = alloca &lt;type&gt;[, uint &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1919</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1920
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1921<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1922
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1923<p>The '<tt>alloca</tt>' instruction allocates memory on the current
1924stack frame of the procedure that is live until the current function
1925returns to its caller.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1926
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1927<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1928
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1929<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1930bytes of memory on the runtime stack, returning a pointer of the
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1931appropriate type to the program. If "NumElements" is specified, it is the
1932number of elements allocated. If an alignment is specified, the value result
1933of the allocation is guaranteed to be aligned to at least that boundary. If
1934not specified, or if zero, the target can choose to align the allocation on any
1935convenient boundary.</p>
1936
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1937<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1938
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1939<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1940
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1941<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1942memory is automatically released when the function returns. The '<tt>alloca</tt>'
1943instruction is commonly used to represent automatic variables that must
1944have an address available. When the function returns (either with the <tt><a
John Criswelldae2e932005-05-12 16:55:34 +0000[diff] [blame]1945 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]1946instructions), the memory is reclaimed.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1947
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1948<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1949
1950<pre>
1951 %ptr = alloca int <i>; yields {int*}:ptr</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1952 %ptr = alloca int, uint 4 <i>; yields {int*}:ptr</i>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1953 %ptr = alloca int, uint 4, align 1024 <i>; yields {int*}:ptr</i>
1954 %ptr = alloca int, align 1024 <i>; yields {int*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1955</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1956</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]1957
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1958<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1959<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
1960Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1961<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1962<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1963<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]1964<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1965<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1966<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1967<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1968address from which to load. The pointer must point to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1969 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1970marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1971the number or order of execution of this <tt>load</tt> with other
1972volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
1973instructions. </p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1974<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1975<p>The location of memory pointed to is loaded.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1976<h5>Examples:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1977<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1978 <a
1979 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1980 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1981</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1982</div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1983<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1984<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
1985Instruction</a> </div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1986<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1987<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]1988 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]1989</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1990<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1991<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1992<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1993<p>There are two arguments to the '<tt>store</tt>' instruction: a value
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1994to 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]1995operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1996operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1997optimizer is not allowed to modify the number or order of execution of
1998this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
1999 href="#i_store">store</a></tt> instructions.</p>
2000<h5>Semantics:</h5>
2001<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
2002at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2003<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2004<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
2005 <a
2006 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2007 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
2008</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2009<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2010<div class="doc_subsubsection">
2011 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
2012</div>
2013
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2014<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2015<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2016<pre>
2017 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
2018</pre>
2019
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2020<h5>Overview:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2021
2022<p>
2023The '<tt>getelementptr</tt>' instruction is used to get the address of a
2024subelement of an aggregate data structure.</p>
2025
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2026<h5>Arguments:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2027
2028<p>This instruction takes a list of integer constants that indicate what
2029elements of the aggregate object to index to. The actual types of the arguments
2030provided depend on the type of the first pointer argument. The
2031'<tt>getelementptr</tt>' instruction is used to index down through the type
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2032levels of a structure or to a specific index in an array. When indexing into a
2033structure, only <tt>uint</tt>
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]2034integer constants are allowed. When indexing into an array or pointer,
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2035<tt>int</tt> and <tt>long</tt> indexes are allowed of any sign.</p>
2036
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2037<p>For example, let's consider a C code fragment and how it gets
2038compiled to LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2039
2040<pre>
2041 struct RT {
2042 char A;
2043 int B[10][20];
2044 char C;
2045 };
2046 struct ST {
2047 int X;
2048 double Y;
2049 struct RT Z;
2050 };
2051
2052 int *foo(struct ST *s) {
2053 return &amp;s[1].Z.B[5][13];
2054 }
2055</pre>
2056
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2057<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2058
2059<pre>
2060 %RT = type { sbyte, [10 x [20 x int]], sbyte }
2061 %ST = type { int, double, %RT }
2062
Brian Gaeke7283e7c2004-07-02 21:08:14 +0000[diff] [blame]2063 implementation
2064
2065 int* %foo(%ST* %s) {
2066 entry:
2067 %reg = getelementptr %ST* %s, int 1, uint 2, uint 1, int 5, int 13
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2068 ret int* %reg
2069 }
2070</pre>
2071
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2072<h5>Semantics:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2073
2074<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]2075on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]2076and <a href="#t_array">array</a> types require <tt>uint</tt>, <tt>int</tt>,
2077<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]2078types require <tt>uint</tt> <b>constants</b>.</p>
2079
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2080<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]2081type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ int, double, %RT
2082}</tt>' type, a structure. The second index indexes into the third element of
2083the structure, yielding a '<tt>%RT</tt>' = '<tt>{ sbyte, [10 x [20 x int]],
2084sbyte }</tt>' type, another structure. The third index indexes into the second
2085element of the structure, yielding a '<tt>[10 x [20 x int]]</tt>' type, an
2086array. The two dimensions of the array are subscripted into, yielding an
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2087'<tt>int</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2088to this element, thus computing a value of '<tt>int*</tt>' type.</p>
2089
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2090<p>Note that it is perfectly legal to index partially through a
2091structure, returning a pointer to an inner element. Because of this,
2092the LLVM code for the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2093
2094<pre>
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]2095 int* %foo(%ST* %s) {
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2096 %t1 = getelementptr %ST* %s, int 1 <i>; yields %ST*:%t1</i>
2097 %t2 = getelementptr %ST* %t1, int 0, uint 2 <i>; yields %RT*:%t2</i>
2098 %t3 = getelementptr %RT* %t2, int 0, uint 1 <i>; yields [10 x [20 x int]]*:%t3</i>
2099 %t4 = getelementptr [10 x [20 x int]]* %t3, int 0, int 5 <i>; yields [20 x int]*:%t4</i>
2100 %t5 = getelementptr [20 x int]* %t4, int 0, int 13 <i>; yields int*:%t5</i>
2101 ret int* %t5
2102 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2103</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]2104
2105<p>Note that it is undefined to access an array out of bounds: array and
2106pointer indexes must always be within the defined bounds of the array type.
2107The one exception for this rules is zero length arrays. These arrays are
2108defined to be accessible as variable length arrays, which requires access
2109beyond the zero'th element.</p>
2110
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2111<h5>Example:</h5>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]2112
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2113<pre>
2114 <i>; yields [12 x ubyte]*:aptr</i>
2115 %aptr = getelementptr {int, [12 x ubyte]}* %sptr, long 0, uint 1
2116</pre>
2117
2118</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2119<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2120<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2121<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]2122<p>The instructions in this category are the "miscellaneous"
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2123instructions, which defy better classification.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2124</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2125<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2126<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
2127Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2128<div class="doc_text">
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2129<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2130<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]2131<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2132<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
2133the SSA graph representing the function.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2134<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2135<p>The type of the incoming values are specified with the first type
2136field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
2137as arguments, with one pair for each predecessor basic block of the
2138current block. Only values of <a href="#t_firstclass">first class</a>
2139type may be used as the value arguments to the PHI node. Only labels
2140may be used as the label arguments.</p>
2141<p>There must be no non-phi instructions between the start of a basic
2142block and the PHI instructions: i.e. PHI instructions must be first in
2143a basic block.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2144<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2145<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the
2146value specified by the parameter, depending on which basic block we
2147came from in the last <a href="#terminators">terminator</a> instruction.</p>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2148<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2149<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]2150</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2151
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2152<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2153<div class="doc_subsubsection">
2154 <a name="i_cast">'<tt>cast .. to</tt>' Instruction</a>
2155</div>
2156
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2157<div class="doc_text">
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2158
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2159<h5>Syntax:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2160
2161<pre>
2162 &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]2163</pre>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2164
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2165<h5>Overview:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2166
2167<p>
2168The '<tt>cast</tt>' instruction is used as the primitive means to convert
2169integers to floating point, change data type sizes, and break type safety (by
2170casting pointers).
2171</p>
2172
2173
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2174<h5>Arguments:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2175
2176<p>
2177The '<tt>cast</tt>' instruction takes a value to cast, which must be a first
2178class value, and a type to cast it to, which must also be a <a
2179href="#t_firstclass">first class</a> type.
2180</p>
2181
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2182<h5>Semantics:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2183
2184<p>
2185This instruction follows the C rules for explicit casts when determining how the
2186data being cast must change to fit in its new container.
2187</p>
2188
2189<p>
2190When casting to bool, any value that would be considered true in the context of
2191a C '<tt>if</tt>' condition is converted to the boolean '<tt>true</tt>' values,
2192all else are '<tt>false</tt>'.
2193</p>
2194
2195<p>
2196When extending an integral value from a type of one signness to another (for
2197example '<tt>sbyte</tt>' to '<tt>ulong</tt>'), the value is sign-extended if the
2198<b>source</b> value is signed, and zero-extended if the source value is
2199unsigned. <tt>bool</tt> values are always zero extended into either zero or
2200one.
2201</p>
2202
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2203<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2204
2205<pre>
2206 %X = cast int 257 to ubyte <i>; yields ubyte:1</i>
Chris Lattner7bae3952002-06-25 18:03:17 +0000[diff] [blame]2207 %Y = cast int 123 to bool <i>; yields bool:true</i>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2208</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2209</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2210
2211<!-- _______________________________________________________________________ -->
2212<div class="doc_subsubsection">
2213 <a name="i_select">'<tt>select</tt>' Instruction</a>
2214</div>
2215
2216<div class="doc_text">
2217
2218<h5>Syntax:</h5>
2219
2220<pre>
2221 &lt;result&gt; = select bool &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
2222</pre>
2223
2224<h5>Overview:</h5>
2225
2226<p>
2227The '<tt>select</tt>' instruction is used to choose one value based on a
2228condition, without branching.
2229</p>
2230
2231
2232<h5>Arguments:</h5>
2233
2234<p>
2235The '<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.
2236</p>
2237
2238<h5>Semantics:</h5>
2239
2240<p>
2241If the boolean condition evaluates to true, the instruction returns the first
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2242value argument; otherwise, it returns the second value argument.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2243</p>
2244
2245<h5>Example:</h5>
2246
2247<pre>
2248 %X = select bool true, ubyte 17, ubyte 42 <i>; yields ubyte:17</i>
2249</pre>
2250</div>
2251
2252
Robert Bocchino3a558662006-01-05 17:37:02 +0000[diff] [blame]2253<!-- _______________________________________________________________________ -->
2254<div class="doc_subsubsection">
2255 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
2256</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2257
Robert Bocchino3a558662006-01-05 17:37:02 +0000[diff] [blame]2258<div class="doc_text">
2259
2260<h5>Syntax:</h5>
2261
2262<pre>
2263 &lt;result&gt; = extractelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, uint &lt;idx&gt; <i>; yields &lt;ty&gt;</i>
2264</pre>
2265
2266<h5>Overview:</h5>
2267
2268<p>
2269The '<tt>extractelement</tt>' instruction extracts a single scalar
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]2270element from a packed vector at a specified index.
Robert Bocchino3a558662006-01-05 17:37:02 +0000[diff] [blame]2271</p>
2272
2273
2274<h5>Arguments:</h5>
2275
2276<p>
2277The first operand of an '<tt>extractelement</tt>' instruction is a
2278value of <a href="#t_packed">packed</a> type. The second operand is
2279an index indicating the position from which to extract the element.
2280The index may be a variable.</p>
2281
2282<h5>Semantics:</h5>
2283
2284<p>
2285The result is a scalar of the same type as the element type of
2286<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
2287<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
2288results are undefined.
2289</p>
2290
2291<h5>Example:</h5>
2292
2293<pre>
2294 %result = extractelement &lt;4 x int&gt; %vec, uint 0 <i>; yields int</i>
2295</pre>
2296</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]2297
2298
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]2299<!-- _______________________________________________________________________ -->
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2300<div class="doc_subsubsection">
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]2301 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
2302</div>
2303
2304<div class="doc_text">
2305
2306<h5>Syntax:</h5>
2307
2308<pre>
2309 &lt;result&gt; = insertelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, &lt;ty&gt; &lt;elt&gt, uint &lt;idx&gt; <i>; yields &lt;n x &lt;ty&gt;&gt;</i>
2310</pre>
2311
2312<h5>Overview:</h5>
2313
2314<p>
2315The '<tt>insertelement</tt>' instruction inserts a scalar
2316element into a packed vector at a specified index.
2317</p>
2318
2319
2320<h5>Arguments:</h5>
2321
2322<p>
2323The first operand of an '<tt>insertelement</tt>' instruction is a
2324value of <a href="#t_packed">packed</a> type. The second operand is a
2325scalar value whose type must equal the element type of the first
2326operand. The third operand is an index indicating the position at
2327which to insert the value. The index may be a variable.</p>
2328
2329<h5>Semantics:</h5>
2330
2331<p>
2332The result is a packed vector of the same type as <tt>val</tt>. Its
2333element values are those of <tt>val</tt> except at position
2334<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
2335exceeds the length of <tt>val</tt>, the results are undefined.
2336</p>
2337
2338<h5>Example:</h5>
2339
2340<pre>
2341 %result = insertelement &lt;4 x int&gt; %vec, int 1, uint 0 <i>; yields &lt;4 x int&gt;</i>
2342</pre>
2343</div>
2344
2345
2346<!-- _______________________________________________________________________ -->
2347<div class="doc_subsubsection">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2348 <a name="i_call">'<tt>call</tt>' Instruction</a>
2349</div>
2350
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2351<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2352
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2353<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2354<pre>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2355 &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]2356</pre>
2357
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2358<h5>Overview:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2359
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2360<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2361
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2362<h5>Arguments:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2363
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2364<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2365
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2366<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2367 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2368 <p>The optional "tail" marker indicates whether the callee function accesses
2369 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2370 function call is eligible for tail call optimization. Note that calls may
2371 be marked "tail" even if they do not occur before a <a
2372 href="#i_ret"><tt>ret</tt></a> instruction.
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2373 </li>
2374 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2375 <p>The optional "cconv" marker indicates which <a href="callingconv">calling
2376 convention</a> the call should use. If none is specified, the call defaults
2377 to using C calling conventions.
2378 </li>
2379 <li>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2380 <p>'<tt>ty</tt>': shall be the signature of the pointer to function value
2381 being invoked. The argument types must match the types implied by this
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2382 signature. This type can be omitted if the function is not varargs and
2383 if the function type does not return a pointer to a function.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2384 </li>
2385 <li>
2386 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
2387 be invoked. In most cases, this is a direct function invocation, but
2388 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2389 to function value.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2390 </li>
2391 <li>
2392 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencera7e302a2005-05-01 22:22:57 +0000[diff] [blame]2393 function signature argument types. All arguments must be of
2394 <a href="#t_firstclass">first class</a> type. If the function signature
2395 indicates the function accepts a variable number of arguments, the extra
2396 arguments can be specified.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2397 </li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2398</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2399
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2400<h5>Semantics:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2401
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2402<p>The '<tt>call</tt>' instruction is used to cause control flow to
2403transfer to a specified function, with its incoming arguments bound to
2404the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
2405instruction in the called function, control flow continues with the
2406instruction after the function call, and the return value of the
2407function is bound to the result argument. This is a simpler case of
2408the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2409
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2410<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2411
2412<pre>
2413 %retval = call int %test(int %argc)
2414 call int(sbyte*, ...) *%printf(sbyte* %msg, int 12, sbyte 42);
2415 %X = tail call int %foo()
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2416 %Y = tail call <a href="#callingconv">fastcc</a> int %foo()
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]2417</pre>
2418
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2419</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2420
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2421<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2422<div class="doc_subsubsection">
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]2423 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2424</div>
2425
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2426<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2427
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2428<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2429
2430<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2431 &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]2432</pre>
2433
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2434<h5>Overview:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2435
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2436<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2437the "variable argument" area of a function call. It is used to implement the
2438<tt>va_arg</tt> macro in C.</p>
2439
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2440<h5>Arguments:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2441
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2442<p>This instruction takes a <tt>va_list*</tt> value and the type of
2443the argument. It returns a value of the specified argument type and
Jeff Cohen25d4f7e2005-11-11 02:15:27 +0000[diff] [blame]2444increments the <tt>va_list</tt> to point to the next argument. Again, the
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2445actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2446
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2447<h5>Semantics:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2448
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2449<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
2450type from the specified <tt>va_list</tt> and causes the
2451<tt>va_list</tt> to point to the next argument. For more information,
2452see the variable argument handling <a href="#int_varargs">Intrinsic
2453Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2454
2455<p>It is legal for this instruction to be called in a function which does not
2456take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2457function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2458
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2459<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2460href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2461argument.</p>
2462
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]2463<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]2464
2465<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
2466
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2467</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2468
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2469<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2470<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
2471<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2472
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2473<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2474
2475<p>LLVM supports the notion of an "intrinsic function". These functions have
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2476well known names and semantics and are required to follow certain
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2477restrictions. Overall, these instructions represent an extension mechanism for
2478the LLVM language that does not require changing all of the transformations in
2479LLVM to add to the language (or the bytecode reader/writer, the parser,
2480etc...).</p>
2481
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2482<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
2483prefix is reserved in LLVM for intrinsic names; thus, functions may not be named
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2484this. Intrinsic functions must always be external functions: you cannot define
2485the body of intrinsic functions. Intrinsic functions may only be used in call
2486or invoke instructions: it is illegal to take the address of an intrinsic
2487function. Additionally, because intrinsic functions are part of the LLVM
2488language, it is required that they all be documented here if any are added.</p>
2489
2490
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2491<p>To learn how to add an intrinsic function, please see the <a
Chris Lattner590cff32005-05-11 03:35:57 +0000[diff] [blame]2492href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2493</p>
2494
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2495</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2496
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2497<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2498<div class="doc_subsection">
2499 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
2500</div>
2501
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2502<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2503
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2504<p>Variable argument support is defined in LLVM with the <a
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]2505 href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2506intrinsic functions. These functions are related to the similarly
2507named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2508
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2509<p>All of these functions operate on arguments that use a
2510target-specific value type "<tt>va_list</tt>". The LLVM assembly
2511language reference manual does not define what this type is, so all
2512transformations should be prepared to handle intrinsics with any type
2513used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2514
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2515<p>This example shows how the <a href="#i_vanext"><tt>vanext</tt></a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2516instruction and the variable argument handling intrinsic functions are
2517used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2518
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2519<pre>
2520int %test(int %X, ...) {
2521 ; Initialize variable argument processing
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2522 %ap = alloca sbyte*
2523 call void %<a href="#i_va_start">llvm.va_start</a>(sbyte** %ap)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2524
2525 ; Read a single integer argument
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2526 %tmp = va_arg sbyte** %ap, int
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2527
2528 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2529 %aq = alloca sbyte*
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2530 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]2531 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte** %aq)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2532
2533 ; Stop processing of arguments.
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2534 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte** %ap)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2535 ret int %tmp
2536}
2537</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2538</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2539
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2540<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2541<div class="doc_subsubsection">
2542 <a name="i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
2543</div>
2544
2545
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2546<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2547<h5>Syntax:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2548<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]2549<h5>Overview:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2550<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
2551<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
2552href="#i_va_arg">va_arg</a></tt>.</p>
2553
2554<h5>Arguments:</h5>
2555
2556<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
2557
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2558<h5>Semantics:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2559
2560<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
2561macro available in C. In a target-dependent way, it initializes the
2562<tt>va_list</tt> element the argument points to, so that the next call to
2563<tt>va_arg</tt> will produce the first variable argument passed to the function.
2564Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
2565last argument of the function, the compiler can figure that out.</p>
2566
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2567</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2568
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2569<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2570<div class="doc_subsubsection">
2571 <a name="i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
2572</div>
2573
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2574<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2575<h5>Syntax:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2576<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]2577<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2578<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>&lt;arglist&gt;</tt>
2579which has been initialized previously with <tt><a href="#i_va_start">llvm.va_start</a></tt>
2580or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2581<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2582<p>The argument is a <tt>va_list</tt> to destroy.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2583<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2584<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]2585macro available in C. In a target-dependent way, it destroys the <tt>va_list</tt>.
2586Calls to <a href="#i_va_start"><tt>llvm.va_start</tt></a> and <a
2587 href="#i_va_copy"><tt>llvm.va_copy</tt></a> must be matched exactly
2588with calls to <tt>llvm.va_end</tt>.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2589</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2590
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2591<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2592<div class="doc_subsubsection">
2593 <a name="i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
2594</div>
2595
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2596<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2597
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2598<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2599
2600<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2601 declare void %llvm.va_copy(&lt;va_list&gt;* &lt;destarglist&gt;,
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]2602 &lt;va_list&gt;* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2603</pre>
2604
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2605<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2606
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2607<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position from
2608the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2609
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2610<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2611
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2612<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]2613The 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]2614
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2615
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]2616<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2617
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]2618<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt> macro
2619available in C. In a target-dependent way, it copies the source
2620<tt>va_list</tt> element into the destination list. This intrinsic is necessary
2621because 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]2622arbitrarily complex and require memory allocation, for example.</p>
2623
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2624</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2625
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2626<!-- ======================================================================= -->
2627<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2628 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
2629</div>
2630
2631<div class="doc_text">
2632
2633<p>
2634LLVM support for <a href="GarbageCollection.html">Accurate Garbage
2635Collection</a> requires the implementation and generation of these intrinsics.
2636These intrinsics allow identification of <a href="#i_gcroot">GC roots on the
2637stack</a>, as well as garbage collector implementations that require <a
2638href="#i_gcread">read</a> and <a href="#i_gcwrite">write</a> barriers.
2639Front-ends for type-safe garbage collected languages should generate these
2640intrinsics to make use of the LLVM garbage collectors. For more details, see <a
2641href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
2642</p>
2643</div>
2644
2645<!-- _______________________________________________________________________ -->
2646<div class="doc_subsubsection">
2647 <a name="i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
2648</div>
2649
2650<div class="doc_text">
2651
2652<h5>Syntax:</h5>
2653
2654<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2655 declare void %llvm.gcroot(&lt;ty&gt;** %ptrloc, &lt;ty2&gt;* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2656</pre>
2657
2658<h5>Overview:</h5>
2659
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2660<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]2661the code generator, and allows some metadata to be associated with it.</p>
2662
2663<h5>Arguments:</h5>
2664
2665<p>The first argument specifies the address of a stack object that contains the
2666root pointer. The second pointer (which must be either a constant or a global
2667value address) contains the meta-data to be associated with the root.</p>
2668
2669<h5>Semantics:</h5>
2670
2671<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
2672location. At compile-time, the code generator generates information to allow
2673the runtime to find the pointer at GC safe points.
2674</p>
2675
2676</div>
2677
2678
2679<!-- _______________________________________________________________________ -->
2680<div class="doc_subsubsection">
2681 <a name="i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
2682</div>
2683
2684<div class="doc_text">
2685
2686<h5>Syntax:</h5>
2687
2688<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2689 declare sbyte* %llvm.gcread(sbyte** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2690</pre>
2691
2692<h5>Overview:</h5>
2693
2694<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
2695locations, allowing garbage collector implementations that require read
2696barriers.</p>
2697
2698<h5>Arguments:</h5>
2699
2700<p>The argument is the address to read from, which should be an address
2701allocated from the garbage collector.</p>
2702
2703<h5>Semantics:</h5>
2704
2705<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
2706instruction, but may be replaced with substantially more complex code by the
2707garbage collector runtime, as needed.</p>
2708
2709</div>
2710
2711
2712<!-- _______________________________________________________________________ -->
2713<div class="doc_subsubsection">
2714 <a name="i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
2715</div>
2716
2717<div class="doc_text">
2718
2719<h5>Syntax:</h5>
2720
2721<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2722 declare void %llvm.gcwrite(sbyte* %P1, sbyte** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]2723</pre>
2724
2725<h5>Overview:</h5>
2726
2727<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
2728locations, allowing garbage collector implementations that require write
2729barriers (such as generational or reference counting collectors).</p>
2730
2731<h5>Arguments:</h5>
2732
2733<p>The first argument is the reference to store, and the second is the heap
2734location to store to.</p>
2735
2736<h5>Semantics:</h5>
2737
2738<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
2739instruction, but may be replaced with substantially more complex code by the
2740garbage collector runtime, as needed.</p>
2741
2742</div>
2743
2744
2745
2746<!-- ======================================================================= -->
2747<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2748 <a name="int_codegen">Code Generator Intrinsics</a>
2749</div>
2750
2751<div class="doc_text">
2752<p>
2753These intrinsics are provided by LLVM to expose special features that may only
2754be implemented with code generator support.
2755</p>
2756
2757</div>
2758
2759<!-- _______________________________________________________________________ -->
2760<div class="doc_subsubsection">
2761 <a name="i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
2762</div>
2763
2764<div class="doc_text">
2765
2766<h5>Syntax:</h5>
2767<pre>
Chris Lattnerfcf39d42006-01-13 01:20:27 +0000[diff] [blame]2768 declare sbyte *%llvm.returnaddress(uint &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2769</pre>
2770
2771<h5>Overview:</h5>
2772
2773<p>
2774The '<tt>llvm.returnaddress</tt>' intrinsic returns a target-specific value
2775indicating the return address of the current function or one of its callers.
2776</p>
2777
2778<h5>Arguments:</h5>
2779
2780<p>
2781The argument to this intrinsic indicates which function to return the address
2782for. Zero indicates the calling function, one indicates its caller, etc. The
2783argument is <b>required</b> to be a constant integer value.
2784</p>
2785
2786<h5>Semantics:</h5>
2787
2788<p>
2789The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
2790the return address of the specified call frame, or zero if it cannot be
2791identified. The value returned by this intrinsic is likely to be incorrect or 0
2792for arguments other than zero, so it should only be used for debugging purposes.
2793</p>
2794
2795<p>
2796Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]2797aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2798source-language caller.
2799</p>
2800</div>
2801
2802
2803<!-- _______________________________________________________________________ -->
2804<div class="doc_subsubsection">
2805 <a name="i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
2806</div>
2807
2808<div class="doc_text">
2809
2810<h5>Syntax:</h5>
2811<pre>
Chris Lattnerfcf39d42006-01-13 01:20:27 +0000[diff] [blame]2812 declare sbyte *%llvm.frameaddress(uint &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2813</pre>
2814
2815<h5>Overview:</h5>
2816
2817<p>
2818The '<tt>llvm.frameaddress</tt>' intrinsic returns the target-specific frame
2819pointer value for the specified stack frame.
2820</p>
2821
2822<h5>Arguments:</h5>
2823
2824<p>
2825The argument to this intrinsic indicates which function to return the frame
2826pointer for. Zero indicates the calling function, one indicates its caller,
2827etc. The argument is <b>required</b> to be a constant integer value.
2828</p>
2829
2830<h5>Semantics:</h5>
2831
2832<p>
2833The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
2834the frame address of the specified call frame, or zero if it cannot be
2835identified. The value returned by this intrinsic is likely to be incorrect or 0
2836for arguments other than zero, so it should only be used for debugging purposes.
2837</p>
2838
2839<p>
2840Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]2841aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2842source-language caller.
2843</p>
2844</div>
2845
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2846<!-- _______________________________________________________________________ -->
2847<div class="doc_subsubsection">
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]2848 <a name="i_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
2849</div>
2850
2851<div class="doc_text">
2852
2853<h5>Syntax:</h5>
2854<pre>
2855 declare sbyte *%llvm.stacksave()
2856</pre>
2857
2858<h5>Overview:</h5>
2859
2860<p>
2861The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
2862the function stack, for use with <a href="#i_stackrestore">
2863<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
2864features like scoped automatic variable sized arrays in C99.
2865</p>
2866
2867<h5>Semantics:</h5>
2868
2869<p>
2870This intrinsic returns a opaque pointer value that can be passed to <a
2871href="#i_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
2872<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
2873<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
2874state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
2875practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
2876that were allocated after the <tt>llvm.stacksave</tt> was executed.
2877</p>
2878
2879</div>
2880
2881<!-- _______________________________________________________________________ -->
2882<div class="doc_subsubsection">
2883 <a name="i_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
2884</div>
2885
2886<div class="doc_text">
2887
2888<h5>Syntax:</h5>
2889<pre>
2890 declare void %llvm.stackrestore(sbyte* %ptr)
2891</pre>
2892
2893<h5>Overview:</h5>
2894
2895<p>
2896The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
2897the function stack to the state it was in when the corresponding <a
2898href="#llvm.stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
2899useful for implementing language features like scoped automatic variable sized
2900arrays in C99.
2901</p>
2902
2903<h5>Semantics:</h5>
2904
2905<p>
2906See the description for <a href="#i_stacksave"><tt>llvm.stacksave</tt></a>.
2907</p>
2908
2909</div>
2910
2911
2912<!-- _______________________________________________________________________ -->
2913<div class="doc_subsubsection">
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2914 <a name="i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
2915</div>
2916
2917<div class="doc_text">
2918
2919<h5>Syntax:</h5>
2920<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2921 declare void %llvm.prefetch(sbyte * &lt;address&gt;,
2922 uint &lt;rw&gt;, uint &lt;locality&gt;)
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2923</pre>
2924
2925<h5>Overview:</h5>
2926
2927
2928<p>
2929The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2930a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
2931no
2932effect on the behavior of the program but can change its performance
Chris Lattner2a615362005-02-28 19:47:14 +0000[diff] [blame]2933characteristics.
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2934</p>
2935
2936<h5>Arguments:</h5>
2937
2938<p>
2939<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
2940determining if the fetch should be for a read (0) or write (1), and
2941<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattneraeffb4a2005-03-07 20:31:38 +0000[diff] [blame]2942locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]2943<tt>locality</tt> arguments must be constant integers.
2944</p>
2945
2946<h5>Semantics:</h5>
2947
2948<p>
2949This intrinsic does not modify the behavior of the program. In particular,
2950prefetches cannot trap and do not produce a value. On targets that support this
2951intrinsic, the prefetch can provide hints to the processor cache for better
2952performance.
2953</p>
2954
2955</div>
2956
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2957<!-- _______________________________________________________________________ -->
2958<div class="doc_subsubsection">
2959 <a name="i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
2960</div>
2961
2962<div class="doc_text">
2963
2964<h5>Syntax:</h5>
2965<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]2966 declare void %llvm.pcmarker( uint &lt;id&gt; )
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2967</pre>
2968
2969<h5>Overview:</h5>
2970
2971
2972<p>
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2973The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
2974(PC) in a region of
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2975code to simulators and other tools. The method is target specific, but it is
2976expected that the marker will use exported symbols to transmit the PC of the marker.
Jeff Cohen25d4f7e2005-11-11 02:15:27 +0000[diff] [blame]2977The marker makes no guarantees that it will remain with any specific instruction
Chris Lattnerd07c3f42005-11-15 06:07:55 +0000[diff] [blame]2978after optimizations. It is possible that the presence of a marker will inhibit
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2979optimizations. The intended use is to be inserted after optmizations to allow
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2980correlations of simulation runs.
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]2981</p>
2982
2983<h5>Arguments:</h5>
2984
2985<p>
2986<tt>id</tt> is a numerical id identifying the marker.
2987</p>
2988
2989<h5>Semantics:</h5>
2990
2991<p>
2992This intrinsic does not modify the behavior of the program. Backends that do not
2993support this intrinisic may ignore it.
2994</p>
2995
2996</div>
2997
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]2998<!-- _______________________________________________________________________ -->
2999<div class="doc_subsubsection">
3000 <a name="i_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
3001</div>
3002
3003<div class="doc_text">
3004
3005<h5>Syntax:</h5>
3006<pre>
3007 declare ulong %llvm.readcyclecounter( )
3008</pre>
3009
3010<h5>Overview:</h5>
3011
3012
3013<p>
3014The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
3015counter register (or similar low latency, high accuracy clocks) on those targets
3016that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
3017As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
3018should only be used for small timings.
3019</p>
3020
3021<h5>Semantics:</h5>
3022
3023<p>
3024When directly supported, reading the cycle counter should not modify any memory.
3025Implementations are allowed to either return a application specific value or a
3026system wide value. On backends without support, this is lowered to a constant 0.
3027</p>
3028
3029</div>
3030
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]3031
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]3032<!-- ======================================================================= -->
3033<div class="doc_subsection">
3034 <a name="int_os">Operating System Intrinsics</a>
3035</div>
3036
3037<div class="doc_text">
3038<p>
3039These intrinsics are provided by LLVM to support the implementation of
3040operating system level code.
3041</p>
3042
3043</div>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3044
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3045<!-- _______________________________________________________________________ -->
3046<div class="doc_subsubsection">
3047 <a name="i_readport">'<tt>llvm.readport</tt>' Intrinsic</a>
3048</div>
3049
3050<div class="doc_text">
3051
3052<h5>Syntax:</h5>
3053<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]3054 declare &lt;integer type&gt; %llvm.readport (&lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3055</pre>
3056
3057<h5>Overview:</h5>
3058
3059<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]3060The '<tt>llvm.readport</tt>' intrinsic reads data from the specified hardware
3061I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3062</p>
3063
3064<h5>Arguments:</h5>
3065
3066<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]3067The argument to this intrinsic indicates the hardware I/O address from which
3068to read the data. The address is in the hardware I/O address namespace (as
3069opposed to being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3070</p>
3071
3072<h5>Semantics:</h5>
3073
3074<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]3075The '<tt>llvm.readport</tt>' intrinsic reads data from the hardware I/O port
3076specified by <i>address</i> and returns the value. The address and return
3077value must be integers, but the size is dependent upon the platform upon which
3078the program is code generated. For example, on x86, the address must be an
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]3079unsigned 16-bit value, and the return value must be 8, 16, or 32 bits.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3080</p>
3081
3082</div>
3083
3084<!-- _______________________________________________________________________ -->
3085<div class="doc_subsubsection">
3086 <a name="i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a>
3087</div>
3088
3089<div class="doc_text">
3090
3091<h5>Syntax:</h5>
3092<pre>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]3093 call void (&lt;integer type&gt;, &lt;integer type&gt;)*
3094 %llvm.writeport (&lt;integer type&gt; &lt;value&gt;,
3095 &lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3096</pre>
3097
3098<h5>Overview:</h5>
3099
3100<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]3101The '<tt>llvm.writeport</tt>' intrinsic writes data to the specified hardware
3102I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3103</p>
3104
3105<h5>Arguments:</h5>
3106
3107<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3108The first argument is the value to write to the I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3109</p>
3110
3111<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3112The second argument indicates the hardware I/O address to which data should be
3113written. The address is in the hardware I/O address namespace (as opposed to
3114being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3115</p>
3116
3117<h5>Semantics:</h5>
3118
3119<p>
3120The '<tt>llvm.writeport</tt>' intrinsic writes <i>value</i> to the I/O port
3121specified by <i>address</i>. The address and value must be integers, but the
3122size is dependent upon the platform upon which the program is code generated.
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]3123For example, on x86, the address must be an unsigned 16-bit value, and the
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]3124value written must be 8, 16, or 32 bits in length.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]3125</p>
3126
3127</div>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3128
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3129<!-- _______________________________________________________________________ -->
3130<div class="doc_subsubsection">
3131 <a name="i_readio">'<tt>llvm.readio</tt>' Intrinsic</a>
3132</div>
3133
3134<div class="doc_text">
3135
3136<h5>Syntax:</h5>
3137<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]3138 declare &lt;result&gt; %llvm.readio (&lt;ty&gt; * &lt;pointer&gt;)
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3139</pre>
3140
3141<h5>Overview:</h5>
3142
3143<p>
3144The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
3145address.
3146</p>
3147
3148<h5>Arguments:</h5>
3149
3150<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3151The argument to this intrinsic is a pointer indicating the memory address from
3152which to read the data. The data must be a
3153<a href="#t_firstclass">first class</a> type.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3154</p>
3155
3156<h5>Semantics:</h5>
3157
3158<p>
3159The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3160location specified by <i>pointer</i> and returns the value. The argument must
3161be a pointer, and the return value must be a
3162<a href="#t_firstclass">first class</a> type. However, certain architectures
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]3163may not support I/O on all first class types. For example, 32-bit processors
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3164may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3165</p>
3166
3167<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3168This intrinsic enforces an in-order memory model for llvm.readio and
3169llvm.writeio calls on machines that use dynamic scheduling. Dynamically
3170scheduled processors may execute loads and stores out of order, re-ordering at
3171run time accesses to memory mapped I/O registers. Using these intrinsics
3172ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3173</p>
3174
3175</div>
3176
3177<!-- _______________________________________________________________________ -->
3178<div class="doc_subsubsection">
3179 <a name="i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a>
3180</div>
3181
3182<div class="doc_text">
3183
3184<h5>Syntax:</h5>
3185<pre>
Reid Spencera8d451e2005-04-26 20:50:44 +0000[diff] [blame]3186 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]3187</pre>
3188
3189<h5>Overview:</h5>
3190
3191<p>
3192The '<tt>llvm.writeio</tt>' intrinsic writes data to the specified memory
3193mapped I/O address.
3194</p>
3195
3196<h5>Arguments:</h5>
3197
3198<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3199The first argument is the value to write to the memory mapped I/O location.
3200The second argument is a pointer indicating the memory address to which the
3201data should be written.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3202</p>
3203
3204<h5>Semantics:</h5>
3205
3206<p>
3207The '<tt>llvm.writeio</tt>' intrinsic writes <i>value</i> to the memory mapped
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3208I/O address specified by <i>pointer</i>. The value must be a
3209<a href="#t_firstclass">first class</a> type. However, certain architectures
Misha Brukmancfa87bc2005-04-22 18:02:52 +0000[diff] [blame]3210may not support I/O on all first class types. For example, 32-bit processors
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3211may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3212</p>
3213
3214<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]3215This intrinsic enforces an in-order memory model for llvm.readio and
3216llvm.writeio calls on machines that use dynamic scheduling. Dynamically
3217scheduled processors may execute loads and stores out of order, re-ordering at
3218run time accesses to memory mapped I/O registers. Using these intrinsics
3219ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]3220</p>
3221
3222</div>
3223
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3224<!-- ======================================================================= -->
3225<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3226 <a name="int_libc">Standard C Library Intrinsics</a>
3227</div>
3228
3229<div class="doc_text">
3230<p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3231LLVM provides intrinsics for a few important standard C library functions.
3232These intrinsics allow source-language front-ends to pass information about the
3233alignment of the pointer arguments to the code generator, providing opportunity
3234for more efficient code generation.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3235</p>
3236
3237</div>
3238
3239<!-- _______________________________________________________________________ -->
3240<div class="doc_subsubsection">
3241 <a name="i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
3242</div>
3243
3244<div class="doc_text">
3245
3246<h5>Syntax:</h5>
3247<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3248 declare void %llvm.memcpy(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
3249 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3250</pre>
3251
3252<h5>Overview:</h5>
3253
3254<p>
3255The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
3256location to the destination location.
3257</p>
3258
3259<p>
3260Note that, unlike the standard libc function, the <tt>llvm.memcpy</tt> intrinsic
3261does not return a value, and takes an extra alignment argument.
3262</p>
3263
3264<h5>Arguments:</h5>
3265
3266<p>
3267The first argument is a pointer to the destination, the second is a pointer to
3268the source. The third argument is an (arbitrarily sized) integer argument
3269specifying the number of bytes to copy, and the fourth argument is the alignment
3270of the source and destination locations.
3271</p>
3272
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]3273<p>
3274If the call to this intrinisic has an alignment value that is not 0 or 1, then
3275the caller guarantees that the size of the copy is a multiple of the alignment
3276and that both the source and destination pointers are aligned to that boundary.
3277</p>
3278
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3279<h5>Semantics:</h5>
3280
3281<p>
3282The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
3283location to the destination location, which are not allowed to overlap. It
3284copies "len" bytes of memory over. If the argument is known to be aligned to
3285some boundary, this can be specified as the fourth argument, otherwise it should
3286be set to 0 or 1.
3287</p>
3288</div>
3289
3290
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3291<!-- _______________________________________________________________________ -->
3292<div class="doc_subsubsection">
3293 <a name="i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
3294</div>
3295
3296<div class="doc_text">
3297
3298<h5>Syntax:</h5>
3299<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3300 declare void %llvm.memmove(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
3301 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3302</pre>
3303
3304<h5>Overview:</h5>
3305
3306<p>
3307The '<tt>llvm.memmove</tt>' intrinsic moves a block of memory from the source
3308location to the destination location. It is similar to the '<tt>llvm.memcpy</tt>'
3309intrinsic but allows the two memory locations to overlap.
3310</p>
3311
3312<p>
3313Note that, unlike the standard libc function, the <tt>llvm.memmove</tt> intrinsic
3314does not return a value, and takes an extra alignment argument.
3315</p>
3316
3317<h5>Arguments:</h5>
3318
3319<p>
3320The first argument is a pointer to the destination, the second is a pointer to
3321the source. The third argument is an (arbitrarily sized) integer argument
3322specifying the number of bytes to copy, and the fourth argument is the alignment
3323of the source and destination locations.
3324</p>
3325
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]3326<p>
3327If the call to this intrinisic has an alignment value that is not 0 or 1, then
3328the caller guarantees that the size of the copy is a multiple of the alignment
3329and that both the source and destination pointers are aligned to that boundary.
3330</p>
3331
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]3332<h5>Semantics:</h5>
3333
3334<p>
3335The '<tt>llvm.memmove</tt>' intrinsic copies a block of memory from the source
3336location to the destination location, which may overlap. It
3337copies "len" bytes of memory over. If the argument is known to be aligned to
3338some boundary, this can be specified as the fourth argument, otherwise it should
3339be set to 0 or 1.
3340</p>
3341</div>
3342
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3343
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3344<!-- _______________________________________________________________________ -->
3345<div class="doc_subsubsection">
3346 <a name="i_memset">'<tt>llvm.memset</tt>' Intrinsic</a>
3347</div>
3348
3349<div class="doc_text">
3350
3351<h5>Syntax:</h5>
3352<pre>
Reid Spencerd4622352005-04-26 20:41:16 +0000[diff] [blame]3353 declare void %llvm.memset(sbyte* &lt;dest&gt;, ubyte &lt;val&gt;,
3354 uint &lt;len&gt;, uint &lt;align&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]3355</pre>
3356
3357<h5>Overview:</h5>
3358
3359<p>
3360The '<tt>llvm.memset</tt>' intrinsic fills a block of memory with a particular
3361byte value.
3362</p>
3363
3364<p>
3365Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
3366does not return a value, and takes an extra alignment argument.
3367</p>
3368
3369<h5>Arguments:</h5>
3370
3371<p>
3372The first argument is a pointer to the destination to fill, the second is the
3373byte value to fill it with, the third argument is an (arbitrarily sized) integer
3374argument specifying the number of bytes to fill, and the fourth argument is the
3375known alignment of destination location.
3376</p>
3377
3378<p>
3379If the call to this intrinisic has an alignment value that is not 0 or 1, then
3380the caller guarantees that the size of the copy is a multiple of the alignment
3381and that the destination pointer is aligned to that boundary.
3382</p>
3383
3384<h5>Semantics:</h5>
3385
3386<p>
3387The '<tt>llvm.memset</tt>' intrinsic fills "len" bytes of memory starting at the
3388destination location. If the argument is known to be aligned to some boundary,
3389this can be specified as the fourth argument, otherwise it should be set to 0 or
33901.
3391</p>
3392</div>
3393
3394
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]3395<!-- _______________________________________________________________________ -->
3396<div class="doc_subsubsection">
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3397 <a name="i_isunordered">'<tt>llvm.isunordered.*</tt>' Intrinsic</a>
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]3398</div>
3399
3400<div class="doc_text">
3401
3402<h5>Syntax:</h5>
3403<pre>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3404 declare bool %llvm.isunordered.f32(float Val1, float Val2)
3405 declare bool %llvm.isunordered.f64(double Val1, double Val2)
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]3406</pre>
3407
3408<h5>Overview:</h5>
3409
3410<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3411The '<tt>llvm.isunordered</tt>' intrinsics return true if either or both of the
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]3412specified floating point values is a NAN.
3413</p>
3414
3415<h5>Arguments:</h5>
3416
3417<p>
3418The arguments are floating point numbers of the same type.
3419</p>
3420
3421<h5>Semantics:</h5>
3422
3423<p>
3424If either or both of the arguments is a SNAN or QNAN, it returns true, otherwise
3425false.
3426</p>
3427</div>
3428
3429
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]3430<!-- _______________________________________________________________________ -->
3431<div class="doc_subsubsection">
3432 <a name="i_sqrt">'<tt>llvm.sqrt</tt>' Intrinsic</a>
3433</div>
3434
3435<div class="doc_text">
3436
3437<h5>Syntax:</h5>
3438<pre>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3439 declare double %llvm.sqrt.f32(float Val)
3440 declare double %llvm.sqrt.f64(double Val)
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]3441</pre>
3442
3443<h5>Overview:</h5>
3444
3445<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3446The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]3447returning the same value as the libm '<tt>sqrt</tt>' function would. Unlike
3448<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
3449negative numbers (which allows for better optimization).
3450</p>
3451
3452<h5>Arguments:</h5>
3453
3454<p>
3455The argument and return value are floating point numbers of the same type.
3456</p>
3457
3458<h5>Semantics:</h5>
3459
3460<p>
3461This function returns the sqrt of the specified operand if it is a positive
3462floating point number.
3463</p>
3464</div>
3465
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3466<!-- ======================================================================= -->
3467<div class="doc_subsection">
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]3468 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3469</div>
3470
3471<div class="doc_text">
3472<p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]3473LLVM provides intrinsics for a few important bit manipulation operations.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3474These allow efficient code generation for some algorithms.
3475</p>
3476
3477</div>
3478
3479<!-- _______________________________________________________________________ -->
3480<div class="doc_subsubsection">
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]3481 <a name="i_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
3482</div>
3483
3484<div class="doc_text">
3485
3486<h5>Syntax:</h5>
3487<pre>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3488 declare ushort %llvm.bswap.i8( ubyte &lt;id&gt; )
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]3489 declare ushort %llvm.bswap.i16( ushort &lt;id&gt; )
3490 declare uint %llvm.bswap.i32( uint &lt;id&gt; )
3491 declare ulong %llvm.bswap.i64( ulong &lt;id&gt; )
3492</pre>
3493
3494<h5>Overview:</h5>
3495
3496<p>
3497The '<tt>llvm.bwsap</tt>' family of intrinsics is used to byteswap a 16, 32 or
349864 bit quantity. These are useful for performing operations on data that is not
3499in the target's native byte order.
3500</p>
3501
3502<h5>Semantics:</h5>
3503
3504<p>
3505The llvm.bswap.16 intrinsic returns a ushort value that has the high and low
3506byte of the input ushort swapped. Similarly, the llvm.bswap.i32 intrinsic
3507returns a uint value that has the four bytes of the input uint swapped, so that
3508if the input bytes are numbered 0, 1, 2, 3 then the returned uint will have its
3509bytes in 3, 2, 1, 0 order. The llvm.bswap.i64 intrinsic extends this concept
3510to 64 bits.
3511</p>
3512
3513</div>
3514
3515<!-- _______________________________________________________________________ -->
3516<div class="doc_subsubsection">
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3517 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3518</div>
3519
3520<div class="doc_text">
3521
3522<h5>Syntax:</h5>
3523<pre>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3524 declare sbyte %llvm.ctpop.i8(sbyte &lt;src&gt;)
3525 declare short %llvm.ctpop.i16(short &lt;src&gt;)
3526 declare int %llvm.ctpop.i32(int &lt;src&gt;)
3527 declare long %llvm.ctpop.i64(long &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3528</pre>
3529
3530<h5>Overview:</h5>
3531
3532<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3533The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of ones in a
3534variable.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3535</p>
3536
3537<h5>Arguments:</h5>
3538
3539<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3540The only argument is the value to be counted. The argument may be of any
3541integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3542</p>
3543
3544<h5>Semantics:</h5>
3545
3546<p>
3547The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
3548</p>
3549</div>
3550
3551<!-- _______________________________________________________________________ -->
3552<div class="doc_subsubsection">
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3553 <a name="int_ctlz">'<tt>llvm.ctlz</tt>' Intrinsic</a>
3554</div>
3555
3556<div class="doc_text">
3557
3558<h5>Syntax:</h5>
3559<pre>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3560 declare sbyte %llvm.ctlz.i8(sbyte &lt;src&gt;)
3561 declare short %llvm.ctlz.i16(short &lt;src&gt;)
3562 declare int %llvm.ctlz.i32(int &lt;src&gt;)
3563 declare long %llvm.ctlz.i64(long &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3564</pre>
3565
3566<h5>Overview:</h5>
3567
3568<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3569The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
3570leading zeros in a variable.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3571</p>
3572
3573<h5>Arguments:</h5>
3574
3575<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]3576The only argument is the value to be counted. The argument may be of any
3577integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3578</p>
3579
3580<h5>Semantics:</h5>
3581
3582<p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3583The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
3584in a variable. If the src == 0 then the result is the size in bits of the type
3585of src. For example, <tt>llvm.cttz(int 2) = 30</tt>.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]3586</p>
3587</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]3588
3589
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3590
3591<!-- _______________________________________________________________________ -->
3592<div class="doc_subsubsection">
3593 <a name="int_cttz">'<tt>llvm.cttz</tt>' Intrinsic</a>
3594</div>
3595
3596<div class="doc_text">
3597
3598<h5>Syntax:</h5>
3599<pre>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3600 declare sbyte %llvm.cttz.i8(sbyte &lt;src&gt;)
3601 declare short %llvm.cttz.i16(short &lt;src&gt;)
3602 declare int %llvm.cttz.i32(int &lt;src&gt;)
3603 declare long %llvm.cttz.i64(long &lt;src&gt;)
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3604</pre>
3605
3606<h5>Overview:</h5>
3607
3608<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame^]3609The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
3610trailing zeros.
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]3611</p>
3612
3613<h5>Arguments:</h5>
3614
3615<p>
3616The only argument is the value to be counted. The argument may be of any
3617integer type. The return type must match the argument type.
3618</p>
3619
3620<h5>Semantics:</h5>
3621
3622<p>
3623The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
3624in a variable. If the src == 0 then the result is the size in bits of the type
3625of src. For example, <tt>llvm.cttz(2) = 1</tt>.
3626</p>
3627</div>
3628
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3629<!-- ======================================================================= -->
3630<div class="doc_subsection">
3631 <a name="int_debugger">Debugger Intrinsics</a>
3632</div>
3633
3634<div class="doc_text">
3635<p>
3636The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
3637are described in the <a
3638href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
3639Debugging</a> document.
3640</p>
3641</div>
3642
3643
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3644<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3645<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]3646<address>
3647 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
3648 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
3649 <a href="http://validator.w3.org/check/referer"><img
3650 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
3651
3652 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
3653 <a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br>
3654 Last modified: $Date$
3655</address>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3656</body>
3657</html>