Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / a58561b91779e9131786003111c59cdc2eae17b0 / . / docs / LangRef.html
blob: 6e794ec426b6c08da808eca7f7272ba9f2303fef [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>
6 <link rel="stylesheet" href="llvm.css" type="text/css">
7</head>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]8
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]9<body>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]10
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]11<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]12<ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]13 <li><a href="#abstract">Abstract</a></li>
14 <li><a href="#introduction">Introduction</a></li>
15 <li><a href="#identifiers">Identifiers</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]16 <li><a href="#typesystem">Type System</a>
17 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]18 <li><a href="#t_primitive">Primitive Types</a>
19 <ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]20 <li><a href="#t_classifications">Type Classifications</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]21 </ol>
22 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]23 <li><a href="#t_derived">Derived Types</a>
24 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]25 <li><a href="#t_array">Array Type</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]26 <li><a href="#t_function">Function Type</a></li>
27 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]28 <li><a href="#t_struct">Structure Type</a></li>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]29 <li><a href="#t_packed">Packed Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]30 </ol>
31 </li>
32 </ol>
33 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]34 <li><a href="#highlevel">High Level Structure</a>
35 <ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]36 <li><a href="#modulestructure">Module Structure</a></li>
37 <li><a href="#globalvars">Global Variables</a></li>
38 <li><a href="#functionstructure">Function Structure</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]39 </ol>
40 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]41 <li><a href="#instref">Instruction Reference</a>
42 <ol>
43 <li><a href="#terminators">Terminator Instructions</a>
44 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]45 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
46 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]47 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
48 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]49 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
50 </ol>
51 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]52 <li><a href="#binaryops">Binary Operations</a>
53 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]54 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
55 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
56 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
57 <li><a href="#i_div">'<tt>div</tt>' Instruction</a></li>
58 <li><a href="#i_rem">'<tt>rem</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]59 <li><a href="#i_setcc">'<tt>set<i>cc</i></tt>' Instructions</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]60 </ol>
61 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]62 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
63 <ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]64 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]65 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]66 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
67 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
68 <li><a href="#i_shr">'<tt>shr</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]69 </ol>
70 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]71 <li><a href="#memoryops">Memory Access Operations</a>
72 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]73 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
74 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
75 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
76 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
77 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
78 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
79 </ol>
80 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]81 <li><a href="#otherops">Other Operations</a>
82 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]83 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]84 <li><a href="#i_cast">'<tt>cast .. to</tt>' Instruction</a></li>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]85 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]86 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]87 <li><a href="#i_vanext">'<tt>vanext</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]88 <li><a href="#i_vaarg">'<tt>vaarg</tt>' Instruction</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]89 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]90 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]91 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]92 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]93 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]94 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]95 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
96 <ol>
97 <li><a href="#i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
98 <li><a href="#i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
99 <li><a href="#i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
100 </ol>
101 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]102 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
103 <ol>
104 <li><a href="#i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
105 <li><a href="#i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
106 <li><a href="#i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
107 </ol>
108 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]109 <li><a href="#int_codegen">Code Generator Intrinsics</a>
110 <ol>
111 <li><a href="#i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
112 <li><a href="#i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]113 </ol>
114 </li>
115 <li><a href="#int_os">Operating System Intrinsics</a>
116 <ol>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]117 <li><a href="#i_readport">'<tt>llvm.readport</tt>' Intrinsic</a></li>
118 <li><a href="#i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a></li>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]119 <li><a href="#i_readio">'<tt>llvm.readio</tt>' Intrinsic</a></li>
120 <li><a href="#i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a></li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]121 </ol>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]122 <li><a href="#int_libc">Standard C Library Intrinsics</a>
123 <ol>
124 <li><a href="#i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a></li>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]125 <li><a href="#i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a></li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]126 <li><a href="#i_memset">'<tt>llvm.memset</tt>' Intrinsic</a></li>
Alkis Evlogimenos96853722004-06-12 19:19:14 +0000[diff] [blame]127 <li><a href="#i_isunordered">'<tt>llvm.isunordered</tt>' Intrinsic</a></li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]128 </ol>
129 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]130 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]131 </ol>
132 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]133</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]134
135<div class="doc_author">
136 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
137 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]138</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]139
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]140<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]141<div class="doc_section"> <a name="abstract">Abstract </a></div>
142<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]143
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]144<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]145<p>This document is a reference manual for the LLVM assembly language.
146LLVM is an SSA based representation that provides type safety,
147low-level operations, flexibility, and the capability of representing
148'all' high-level languages cleanly. It is the common code
149representation used throughout all phases of the LLVM compilation
150strategy.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]151</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]152
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]153<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]154<div class="doc_section"> <a name="introduction">Introduction</a> </div>
155<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]156
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]157<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]158
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]159<p>The LLVM code representation is designed to be used in three
160different forms: as an in-memory compiler IR, as an on-disk bytecode
161representation (suitable for fast loading by a Just-In-Time compiler),
162and as a human readable assembly language representation. This allows
163LLVM to provide a powerful intermediate representation for efficient
164compiler transformations and analysis, while providing a natural means
165to debug and visualize the transformations. The three different forms
166of LLVM are all equivalent. This document describes the human readable
167representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]168
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]169<p>The LLVM representation aims to be a light-weight and low-level
170while being expressive, typed, and extensible at the same time. It
171aims to be a "universal IR" of sorts, by being at a low enough level
172that high-level ideas may be cleanly mapped to it (similar to how
173microprocessors are "universal IR's", allowing many source languages to
174be mapped to them). By providing type information, LLVM can be used as
175the target of optimizations: for example, through pointer analysis, it
176can be proven that a C automatic variable is never accessed outside of
177the current function... allowing it to be promoted to a simple SSA
178value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]179
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]180</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]181
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]182<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]183<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]184
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]185<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]186
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]187<p>It is important to note that this document describes 'well formed'
188LLVM assembly language. There is a difference between what the parser
189accepts and what is considered 'well formed'. For example, the
190following instruction is syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]191
192<pre>
193 %x = <a href="#i_add">add</a> int 1, %x
194</pre>
195
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]196<p>...because the definition of <tt>%x</tt> does not dominate all of
197its uses. The LLVM infrastructure provides a verification pass that may
198be used to verify that an LLVM module is well formed. This pass is
199automatically run by the parser after parsing input assembly, and by
200the optimizer before it outputs bytecode. The violations pointed out
201by the verifier pass indicate bugs in transformation passes or input to
202the parser.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]203
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]204<!-- Describe the typesetting conventions here. --> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]205
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]206<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]207<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]208<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]209
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]210<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]211
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]212<p>LLVM uses three different forms of identifiers, for different
213purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]214
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]215<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]216 <li>Numeric constants are represented as you would expect: 12, -3
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]217123.421, etc. Floating point constants have an optional hexadecimal
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]218notation.</li>
219 <li>Named values are represented as a string of characters with a '%'
220prefix. For example, %foo, %DivisionByZero,
221%a.really.long.identifier. The actual regular expression used is '<tt>%[a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
222Identifiers which require other characters in their names can be
223surrounded with quotes. In this way, anything except a <tt>"</tt>
224character can be used in a name.</li>
225 <li>Unnamed values are represented as an unsigned numeric value with
226a '%' prefix. For example, %12, %2, %44.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]227</ol>
John Criswell6794d922004-03-12 21:19:06 +0000[diff] [blame]228<p>LLVM requires that values start with a '%' sign for two reasons:
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]229Compilers don't need to worry about name clashes with reserved words,
230and the set of reserved words may be expanded in the future without
231penalty. Additionally, unnamed identifiers allow a compiler to quickly
232come up with a temporary variable without having to avoid symbol table
233conflicts.</p>
234<p>Reserved words in LLVM are very similar to reserved words in other
235languages. There are keywords for different opcodes ('<tt><a
236 href="#i_add">add</a></tt>', '<tt><a href="#i_cast">cast</a></tt>', '<tt><a
237 href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
238 href="#t_void">void</a></tt>', '<tt><a href="#t_uint">uint</a></tt>',
239etc...), and others. These reserved words cannot conflict with
240variable names, because none of them start with a '%' character.</p>
241<p>Here is an example of LLVM code to multiply the integer variable '<tt>%X</tt>'
242by 8:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]243<p>The easy way:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]244<pre> %result = <a href="#i_mul">mul</a> uint %X, 8<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]245<p>After strength reduction:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]246<pre> %result = <a href="#i_shl">shl</a> uint %X, ubyte 3<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]247<p>And the hard way:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]248<pre> <a href="#i_add">add</a> uint %X, %X <i>; yields {uint}:%0</i>
249 <a
250 href="#i_add">add</a> uint %0, %0 <i>; yields {uint}:%1</i>
251 %result = <a
252 href="#i_add">add</a> uint %1, %1<br></pre>
253<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
254important lexical features of LLVM:</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]255<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]256 <li>Comments are delimited with a '<tt>;</tt>' and go until the end
257of line.</li>
258 <li>Unnamed temporaries are created when the result of a computation
259is not assigned to a named value.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]260 <li>Unnamed temporaries are numbered sequentially</li>
261</ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]262<p>...and it also show a convention that we follow in this document.
263When demonstrating instructions, we will follow an instruction with a
264comment that defines the type and name of value produced. Comments are
265shown in italic text.</p>
266<p>The one non-intuitive notation for constants is the optional
267hexidecimal form of floating point constants. For example, the form '<tt>double
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2680x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2694.5e+15</tt>' which is also supported by the parser. The only time
270hexadecimal floating point constants are useful (and the only time that
271they are generated by the disassembler) is when an FP constant has to
272be emitted that is not representable as a decimal floating point number
273exactly. For example, NaN's, infinities, and other special cases are
274represented in their IEEE hexadecimal format so that assembly and
275disassembly do not cause any bits to change in the constants.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]276</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]277<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]278<div class="doc_section"> <a name="typesystem">Type System</a> </div>
279<!-- *********************************************************************** -->
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]280<div class="doc_text">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]281<p>The LLVM type system is one of the most important features of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]282intermediate representation. Being typed enables a number of
283optimizations to be performed on the IR directly, without having to do
284extra analyses on the side before the transformation. A strong type
285system makes it easier to read the generated code and enables novel
286analyses and transformations that are not feasible to perform on normal
287three address code representations.</p>
Chris Lattner7bae3952002-06-25 18:03:17 +0000[diff] [blame]288<!-- The written form for the type system was heavily influenced by the
289syntactic problems with types in the C language<sup><a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]290href="#rw_stroustrup">1</a></sup>.<p> --> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]291<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]292<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]293<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]294<p>The primitive types are the fundamental building blocks of the LLVM
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]295system. The current set of primitive types are as follows:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]296
297<table border="0" style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]298 <tbody>
299 <tr>
300 <td>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]301 <table border="1" cellspacing="0" cellpadding="4" style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]302 <tbody>
303 <tr>
304 <td><tt>void</tt></td>
305 <td>No value</td>
306 </tr>
307 <tr>
308 <td><tt>ubyte</tt></td>
309 <td>Unsigned 8 bit value</td>
310 </tr>
311 <tr>
312 <td><tt>ushort</tt></td>
313 <td>Unsigned 16 bit value</td>
314 </tr>
315 <tr>
316 <td><tt>uint</tt></td>
317 <td>Unsigned 32 bit value</td>
318 </tr>
319 <tr>
320 <td><tt>ulong</tt></td>
321 <td>Unsigned 64 bit value</td>
322 </tr>
323 <tr>
324 <td><tt>float</tt></td>
325 <td>32 bit floating point value</td>
326 </tr>
327 <tr>
328 <td><tt>label</tt></td>
329 <td>Branch destination</td>
330 </tr>
331 </tbody>
332 </table>
333 </td>
334 <td valign="top">
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]335 <table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]336 <tbody>
337 <tr>
338 <td><tt>bool</tt></td>
339 <td>True or False value</td>
340 </tr>
341 <tr>
342 <td><tt>sbyte</tt></td>
343 <td>Signed 8 bit value</td>
344 </tr>
345 <tr>
346 <td><tt>short</tt></td>
347 <td>Signed 16 bit value</td>
348 </tr>
349 <tr>
350 <td><tt>int</tt></td>
351 <td>Signed 32 bit value</td>
352 </tr>
353 <tr>
354 <td><tt>long</tt></td>
355 <td>Signed 64 bit value</td>
356 </tr>
357 <tr>
358 <td><tt>double</tt></td>
359 <td>64 bit floating point value</td>
360 </tr>
361 </tbody>
362 </table>
363 </td>
364 </tr>
365 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]366</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]367
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]368</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]369<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]370<div class="doc_subsubsection"> <a name="t_classifications">Type
371Classifications</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]372<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]373<p>These different primitive types fall into a few useful
374classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]375
376<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]377 <tbody>
378 <tr>
379 <td><a name="t_signed">signed</a></td>
380 <td><tt>sbyte, short, int, long, float, double</tt></td>
381 </tr>
382 <tr>
383 <td><a name="t_unsigned">unsigned</a></td>
384 <td><tt>ubyte, ushort, uint, ulong</tt></td>
385 </tr>
386 <tr>
387 <td><a name="t_integer">integer</a></td>
388 <td><tt>ubyte, sbyte, ushort, short, uint, int, ulong, long</tt></td>
389 </tr>
390 <tr>
391 <td><a name="t_integral">integral</a></td>
392 <td><tt>bool, ubyte, sbyte, ushort, short, uint, int, ulong, long</tt></td>
393 </tr>
394 <tr>
395 <td><a name="t_floating">floating point</a></td>
396 <td><tt>float, double</tt></td>
397 </tr>
398 <tr>
399 <td><a name="t_firstclass">first class</a></td>
400 <td><tt>bool, ubyte, sbyte, ushort, short,<br>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]401uint, int, ulong, long, float, double, <a href="#t_pointer">pointer</a>, <a href="#t_packed">packed</a></tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]402 </tr>
403 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]404</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]405
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]406<p>The <a href="#t_firstclass">first class</a> types are perhaps the
407most important. Values of these types are the only ones which can be
408produced by instructions, passed as arguments, or used as operands to
409instructions. This means that all structures and arrays must be
410manipulated either by pointer or by component.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]411</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]412<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]413<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]414<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]415<p>The real power in LLVM comes from the derived types in the system.
416This is what allows a programmer to represent arrays, functions,
417pointers, and other useful types. Note that these derived types may be
418recursive: For example, it is possible to have a two dimensional array.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]419</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]420<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]421<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]422<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]423<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]424<p>The array type is a very simple derived type that arranges elements
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]425sequentially in memory. The array type requires a size (number of
426elements) and an underlying data type.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]427<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]428<pre> [&lt;# elements&gt; x &lt;elementtype&gt;]<br></pre>
429<p>The number of elements is a constant integer value, elementtype may
430be any type with a size.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]431<h5>Examples:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]432<p> <tt>[40 x int ]</tt>: Array of 40 integer values.<br>
433<tt>[41 x int ]</tt>: Array of 41 integer values.<br>
434<tt>[40 x uint]</tt>: Array of 40 unsigned integer values.</p>
435<p> </p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]436<p>Here are some examples of multidimensional arrays:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]437
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]438<table border="0" cellpadding="0" cellspacing="0">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]439 <tbody>
440 <tr>
441 <td><tt>[3 x [4 x int]]</tt></td>
442 <td>: 3x4 array integer values.</td>
443 </tr>
444 <tr>
445 <td><tt>[12 x [10 x float]]</tt></td>
446 <td>: 12x10 array of single precision floating point values.</td>
447 </tr>
448 <tr>
449 <td><tt>[2 x [3 x [4 x uint]]]</tt></td>
450 <td>: 2x3x4 array of unsigned integer values.</td>
451 </tr>
452 </tbody>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]453</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]454
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]455</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]456<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]457<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]458<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]459<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]460<p>The function type can be thought of as a function signature. It
461consists of a return type and a list of formal parameter types.
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]462Function types are usually used to build virtual function tables
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]463(which are structures of pointers to functions), for indirect function
464calls, and when defining a function.</p>
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]465<p>
466The return type of a function type cannot be an aggregate type.
467</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]468<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]469<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
470<p>Where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of
471type specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]472which indicates that the function takes a variable number of arguments.
473Variable argument functions can access their arguments with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]474 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]475<h5>Examples:</h5>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]476
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]477<table border="0" cellpadding="0" cellspacing="0">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]478 <tbody>
479 <tr>
480 <td><tt>int (int)</tt></td>
481 <td>: function taking an <tt>int</tt>, returning an <tt>int</tt></td>
482 </tr>
483 <tr>
484 <td><tt>float (int, int *) *</tt></td>
485 <td>: <a href="#t_pointer">Pointer</a> to a function that takes
486an <tt>int</tt> and a <a href="#t_pointer">pointer</a> to <tt>int</tt>,
487returning <tt>float</tt>.</td>
488 </tr>
489 <tr>
490 <td><tt>int (sbyte *, ...)</tt></td>
491 <td>: A vararg function that takes at least one <a
492 href="#t_pointer">pointer</a> to <tt>sbyte</tt> (signed char in C),
493which returns an integer. This is the signature for <tt>printf</tt>
494in LLVM.</td>
495 </tr>
496 </tbody>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]497</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]498
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]499</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]500<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]501<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]502<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]503<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]504<p>The structure type is used to represent a collection of data members
505together in memory. The packing of the field types is defined to match
506the ABI of the underlying processor. The elements of a structure may
507be any type that has a size.</p>
508<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
509and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
510field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
511instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]512<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]513<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]514<h5>Examples:</h5>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]515
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]516<table border="0" cellpadding="0" cellspacing="0">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]517 <tbody>
518 <tr>
519 <td><tt>{ int, int, int }</tt></td>
520 <td>: a triple of three <tt>int</tt> values</td>
521 </tr>
522 <tr>
523 <td><tt>{ float, int (int) * }</tt></td>
524 <td>: A pair, where the first element is a <tt>float</tt> and the
525second element is a <a href="#t_pointer">pointer</a> to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]526 href="#t_function">function</a> that takes an <tt>int</tt>, returning
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]527an <tt>int</tt>.</td>
528 </tr>
529 </tbody>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]530</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]531
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]532</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]533<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]534<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]535<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]536<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]537<p>As in many languages, the pointer type represents a pointer or
538reference to another object, which must live in memory.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]539<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]540<pre> &lt;type&gt; *<br></pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]541<h5>Examples:</h5>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]542
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]543<table border="0" cellpadding="0" cellspacing="0">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]544 <tbody>
545 <tr>
546 <td><tt>[4x int]*</tt></td>
547 <td>: <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a>
548of four <tt>int</tt> values</td>
549 </tr>
550 <tr>
551 <td><tt>int (int *) *</tt></td>
552 <td>: A <a href="#t_pointer">pointer</a> to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]553 href="#t_function">function</a> that takes an <tt>int</tt>, returning
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]554an <tt>int</tt>.</td>
555 </tr>
556 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]557</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]558</div>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]559<!-- _______________________________________________________________________ -->
560<div class="doc_subsubsection"> <a name="t_packed">Packed Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]561<div class="doc_text">
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]562<h5>Overview:</h5>
563<p>A packed type is a simple derived type that represents a vector
564of elements. Packed types are used when multiple primitive data
565are operated in parallel using a single instruction (SIMD).
566A packed type requires a size (number of
567elements) and an underlying primitive data type. Packed types are
568considered <a href="#t_firstclass">first class</a>.</p>
569<h5>Syntax:</h5>
570<pre> &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;<br></pre>
571<p>The number of elements is a constant integer value, elementtype may
572be any integral or floating point type.</p>
573<h5>Examples:</h5>
574<p> <tt>&lt;4 x int&gt;</tt>: Packed vector of 4 integer values.<br>
575<tt>&lt;8 x float&gt;</tt>: Packed vector of 8 floating-point values.<br>
576<tt>&lt;2 x uint&gt;</tt>: Packed vector of 2 unsigned integer values.</p>
577<p> </p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]578</div>
579
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]580<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]581<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
582<!-- *********************************************************************** --><!-- ======================================================================= -->
583<div class="doc_subsection"> <a name="modulestructure">Module Structure</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]584<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]585<p>LLVM programs are composed of "Module"s, each of which is a
586translation unit of the input programs. Each module consists of
587functions, global variables, and symbol table entries. Modules may be
588combined together with the LLVM linker, which merges function (and
589global variable) definitions, resolves forward declarations, and merges
590symbol table entries. Here is an example of the "hello world" module:</p>
591<pre><i>; Declare the string constant as a global constant...</i>
592<a href="#identifiers">%.LC0</a> = <a href="#linkage_internal">internal</a> <a
593 href="#globalvars">constant</a> <a href="#t_array">[13 x sbyte]</a> c"hello world\0A\00" <i>; [13 x sbyte]*</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]594
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]595<i>; External declaration of the puts function</i>
596<a href="#functionstructure">declare</a> int %puts(sbyte*) <i>; int(sbyte*)* </i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]597
598<i>; Definition of main function</i>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]599int %main() { <i>; int()* </i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]600 <i>; Convert [13x sbyte]* to sbyte *...</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]601 %cast210 = <a
602 href="#i_getelementptr">getelementptr</a> [13 x sbyte]* %.LC0, long 0, long 0 <i>; sbyte*</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]603
604 <i>; Call puts function to write out the string to stdout...</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]605 <a
606 href="#i_call">call</a> int %puts(sbyte* %cast210) <i>; int</i>
607 <a
608 href="#i_ret">ret</a> int 0<br>}<br></pre>
609<p>This example is made up of a <a href="#globalvars">global variable</a>
610named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
611function, and a <a href="#functionstructure">function definition</a>
612for "<tt>main</tt>".</p>
613<a name="linkage"> In general, a module is made up of a list of global
614values, where both functions and global variables are global values.
615Global values are represented by a pointer to a memory location (in
616this case, a pointer to an array of char, and a pointer to a function),
617and have one of the following linkage types:</a>
618<p> </p>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]619<dl>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]620 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]621 <dd>Global values with internal linkage are only directly accessible
622by objects in the current module. In particular, linking code into a
623module with an internal global value may cause the internal to be
624renamed as necessary to avoid collisions. Because the symbol is
625internal to the module, all references can be updated. This
626corresponds to the notion of the '<tt>static</tt>' keyword in C, or the
627idea of "anonymous namespaces" in C++.
628 <p> </p>
629 </dd>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]630 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]631 <dd>"<tt>linkonce</tt>" linkage is similar to <tt>internal</tt>
632linkage, with the twist that linking together two modules defining the
633same <tt>linkonce</tt> globals will cause one of the globals to be
634discarded. This is typically used to implement inline functions.
635Unreferenced <tt>linkonce</tt> globals are allowed to be discarded.
636 <p> </p>
637 </dd>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]638 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]639 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt>
640linkage, except that unreferenced <tt>weak</tt> globals may not be
641discarded. This is used to implement constructs in C such as "<tt>int
642X;</tt>" at global scope.
643 <p> </p>
644 </dd>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]645 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]646 <dd>"<tt>appending</tt>" linkage may only be applied to global
647variables of pointer to array type. When two global variables with
648appending linkage are linked together, the two global arrays are
649appended together. This is the LLVM, typesafe, equivalent of having
650the system linker append together "sections" with identical names when
651.o files are linked.
652 <p> </p>
653 </dd>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]654 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]655 <dd>If none of the above identifiers are used, the global is
656externally visible, meaning that it participates in linkage and can be
657used to resolve external symbol references.
658 <p> </p>
659 </dd>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]660</dl>
661<p> </p>
662<p><a name="linkage_external">For example, since the "<tt>.LC0</tt>"
663variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
664variable and was linked with this one, one of the two would be renamed,
665preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
666external (i.e., lacking any linkage declarations), they are accessible
667outside of the current module. It is illegal for a function <i>declaration</i>
668to have any linkage type other than "externally visible".</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]669</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]670
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]671<!-- ======================================================================= -->
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]672<div class="doc_subsection">
673 <a name="globalvars">Global Variables</a>
674</div>
675
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]676<div class="doc_text">
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]677
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]678<p>Global variables define regions of memory allocated at compilation
679time instead of run-time. Global variables may optionally be
680initialized. A variable may be defined as a global "constant", which
681indicates that the contents of the variable will never be modified
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]682(opening options for optimization).</p>
683
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]684<p>As SSA values, global variables define pointer values that are in
685scope (i.e. they dominate) for all basic blocks in the program. Global
686variables always define a pointer to their "content" type because they
687describe a region of memory, and all memory objects in LLVM are
688accessed through pointers.</p>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]689
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]690</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]691
692
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]693<!-- ======================================================================= -->
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]694<div class="doc_subsection">
695 <a name="functionstructure">Functions</a>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]696</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]697
698<div class="doc_text">
699
700<p>LLVM function definitions are composed of a (possibly empty) argument list,
701an opening curly brace, a list of basic blocks, and a closing curly brace. LLVM
702function declarations are defined with the "<tt>declare</tt>" keyword, a
703function name, and a function signature.</p>
704
705<p>A function definition contains a list of basic blocks, forming the CFG for
706the function. Each basic block may optionally start with a label (giving the
707basic block a symbol table entry), contains a list of instructions, and ends
708with a <a href="#terminators">terminator</a> instruction (such as a branch or
709function return).</p>
710
711<p>The first basic block in program is special in two ways: it is immediately
712executed on entrance to the function, and it is not allowed to have predecessor
713basic blocks (i.e. there can not be any branches to the entry block of a
714function). Because the block can have no predecessors, it also cannot have any
715<a href="#i_phi">PHI nodes</a>.</p>
716
717<p>LLVM functions are identified by their name and type signature. Hence, two
718functions with the same name but different parameter lists or return values are
719considered different functions, and LLVM will resolves references to each
720appropriately.</p>
721
722</div>
723
724
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]725<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]726<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
727<!-- *********************************************************************** -->
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]728<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]729<p>The LLVM instruction set consists of several different
730classifications of instructions: <a href="#terminators">terminator
731instructions</a>, <a href="#binaryops">binary instructions</a>, <a
732 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
733instructions</a>.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]734</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]735<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]736<div class="doc_subsection"> <a name="terminators">Terminator
737Instructions</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]738<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]739<p>As mentioned <a href="#functionstructure">previously</a>, every
740basic block in a program ends with a "Terminator" instruction, which
741indicates which block should be executed after the current block is
742finished. These terminator instructions typically yield a '<tt>void</tt>'
743value: they produce control flow, not values (the one exception being
744the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]745<p>There are five different terminator instructions: the '<a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]746 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
747instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
748the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, and the '<a
749 href="#i_unwind"><tt>unwind</tt></a>' instruction.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]750</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]751<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]752<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
753Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]754<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]755<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]756<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]757 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]758</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]759<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]760<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
761value) from a function, back to the caller.</p>
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]762<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]763returns a value and then causes control flow, and one that just causes
764control flow to occur.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]765<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]766<p>The '<tt>ret</tt>' instruction may return any '<a
767 href="#t_firstclass">first class</a>' type. Notice that a function is
768not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
769instruction inside of the function that returns a value that does not
770match the return type of the function.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]771<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]772<p>When the '<tt>ret</tt>' instruction is executed, control flow
773returns back to the calling function's context. If the caller is a "<a
John Criswellfa081872004-06-25 15:16:57 +0000[diff] [blame]774 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]775the instruction after the call. If the caller was an "<a
776 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
777at the beginning "normal" of the destination block. If the instruction
778returns a value, that value shall set the call or invoke instruction's
779return value.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]780<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]781<pre> ret int 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]782 ret void <i>; Return from a void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]783</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]784</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]785<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]786<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]787<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]788<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]789<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]790</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]791<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]792<p>The '<tt>br</tt>' instruction is used to cause control flow to
793transfer to a different basic block in the current function. There are
794two forms of this instruction, corresponding to a conditional branch
795and an unconditional branch.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]796<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]797<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
798single '<tt>bool</tt>' value and two '<tt>label</tt>' values. The
799unconditional form of the '<tt>br</tt>' instruction takes a single '<tt>label</tt>'
800value as a target.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]801<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]802<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>bool</tt>'
803argument is evaluated. If the value is <tt>true</tt>, control flows
804to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
805control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]806<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]807<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
808 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]809</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]810<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]811<div class="doc_subsubsection">
812 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
813</div>
814
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]815<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]816<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]817
818<pre>
819 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
820</pre>
821
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]822<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]823
824<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
825several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]826instruction, allowing a branch to occur to one of many possible
827destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]828
829
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]830<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]831
832<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
833comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
834an array of pairs of comparison value constants and '<tt>label</tt>'s. The
835table is not allowed to contain duplicate constant entries.</p>
836
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]837<h5>Semantics:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]838
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]839<p>The <tt>switch</tt> instruction specifies a table of values and
840destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]841table is searched for the given value. If the value is found, control flow is
842transfered to the corresponding destination; otherwise, control flow is
843transfered to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]844
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]845<h5>Implementation:</h5>
846
847<p>Depending on properties of the target machine and the particular
848<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]849ways. For example, it could be generated as a series of chained conditional
850branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]851
852<h5>Example:</h5>
853
854<pre>
855 <i>; Emulate a conditional br instruction</i>
856 %Val = <a href="#i_cast">cast</a> bool %value to int
857 switch int %Val, label %truedest [int 0, label %falsedest ]
858
859 <i>; Emulate an unconditional br instruction</i>
860 switch uint 0, label %dest [ ]
861
862 <i>; Implement a jump table:</i>
863 switch uint %val, label %otherwise [ uint 0, label %onzero
864 uint 1, label %onone
865 uint 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]866</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]867</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]868<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]869<div class="doc_subsubsection"> <a name="i_invoke">'<tt>invoke</tt>'
870Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]871<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]872<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]873<pre> &lt;result&gt; = invoke &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)<br> to label &lt;normal label&gt; except label &lt;exception label&gt;<br></pre>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]874<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]875<p>The '<tt>invoke</tt>' instruction causes control to transfer to a
876specified function, with the possibility of control flow transfer to
877either the '<tt>normal</tt>' <tt>label</tt> label or the '<tt>exception</tt>'<tt>label</tt>.
878If the callee function returns with the "<tt><a href="#i_ret">ret</a></tt>"
879instruction, control flow will return to the "normal" label. If the
880callee (or any indirect callees) returns with the "<a href="#i_unwind"><tt>unwind</tt></a>"
881instruction, control is interrupted, and continued at the dynamically
882nearest "except" label.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]883<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]884<p>This instruction requires several arguments:</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]885<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]886 <li>'<tt>ptr to function ty</tt>': shall be the signature of the
887pointer to function value being invoked. In most cases, this is a
888direct function invocation, but indirect <tt>invoke</tt>s are just as
889possible, branching off an arbitrary pointer to function value. </li>
890 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer
891to a function to be invoked. </li>
892 <li>'<tt>function args</tt>': argument list whose types match the
893function signature argument types. If the function signature indicates
894the function accepts a variable number of arguments, the extra
895arguments can be specified. </li>
896 <li>'<tt>normal label</tt>': the label reached when the called
897function executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
898 <li>'<tt>exception label</tt>': the label reached when a callee
899returns with the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]900</ol>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]901<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]902<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]903 href="#i_call">call</a></tt>' instruction in most regards. The
904primary difference is that it establishes an association with a label,
905which is used by the runtime library to unwind the stack.</p>
906<p>This instruction is used in languages with destructors to ensure
907that proper cleanup is performed in the case of either a <tt>longjmp</tt>
908or a thrown exception. Additionally, this is important for
909implementation of '<tt>catch</tt>' clauses in high-level languages that
910support them.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]911<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]912<pre> %retval = invoke int %Test(int 15)<br> to label %Continue<br> except label %TestCleanup <i>; {int}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]913</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]914</div>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]915<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]916<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
917Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]918<div class="doc_text">
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]919<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]920<pre> unwind<br></pre>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]921<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]922<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing
923control flow at the first callee in the dynamic call stack which used
924an <a href="#i_invoke"><tt>invoke</tt></a> instruction to perform the
925call. This is primarily used to implement exception handling.</p>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]926<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]927<p>The '<tt>unwind</tt>' intrinsic causes execution of the current
928function to immediately halt. The dynamic call stack is then searched
929for the first <a href="#i_invoke"><tt>invoke</tt></a> instruction on
930the call stack. Once found, execution continues at the "exceptional"
931destination block specified by the <tt>invoke</tt> instruction. If
932there is no <tt>invoke</tt> instruction in the dynamic call chain,
933undefined behavior results.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]934</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]935<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]936<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]937<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]938<p>Binary operators are used to do most of the computation in a
939program. They require two operands, execute an operation on them, and
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]940produce a single value. Although, that single value might represent
941multiple data, as is the case with the <a href="#t_packed">packed</a> data type.
942The result value of a binary operator is not
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]943necessarily the same type as its operands.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]944<p>There are several different binary operators:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]945</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]946<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]947<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
948Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]949<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]950<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]951<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]952</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]953<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]954<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]955<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]956<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]957 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
958 This instruction can also take <a href="#t_packed">packed</a> versions of the values.
959Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]960<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]961<p>The value produced is the integer or floating point sum of the two
962operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]963<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]964<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]965</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]966</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]967<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]968<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
969Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]970<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]971<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]972<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]973</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]974<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]975<p>The '<tt>sub</tt>' instruction returns the difference of its two
976operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]977<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
978instruction present in most other intermediate representations.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]979<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]980<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]981 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]982values.
983This instruction can also take <a href="#t_packed">packed</a> versions of the values.
984Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]985<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]986<p>The value produced is the integer or floating point difference of
987the two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]988<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]989<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]990 &lt;result&gt; = sub int 0, %val <i>; yields {int}:result = -%var</i>
991</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]992</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]993<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]994<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
995Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]996<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]997<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]998<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]999</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1000<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1001<p>The '<tt>mul</tt>' instruction returns the product of its two
1002operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1003<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1004<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1005 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]1006values.
1007This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1008Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1009<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1010<p>The value produced is the integer or floating point product of the
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1011two operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1012<p>There is no signed vs unsigned multiplication. The appropriate
1013action is taken based on the type of the operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1014<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1015<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]1016</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1017</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1018<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1019<div class="doc_subsubsection"> <a name="i_div">'<tt>div</tt>'
1020Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1021<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1022<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1023<pre> &lt;result&gt; = div &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1024</pre>
1025<h5>Overview:</h5>
1026<p>The '<tt>div</tt>' instruction returns the quotient of its two
1027operands.</p>
1028<h5>Arguments:</h5>
1029<p>The two arguments to the '<tt>div</tt>' instruction must be either <a
1030 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]1031values.
1032This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1033Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1034<h5>Semantics:</h5>
1035<p>The value produced is the integer or floating point quotient of the
1036two operands.</p>
1037<h5>Example:</h5>
1038<pre> &lt;result&gt; = div int 4, %var <i>; yields {int}:result = 4 / %var</i>
1039</pre>
1040</div>
1041<!-- _______________________________________________________________________ -->
1042<div class="doc_subsubsection"> <a name="i_rem">'<tt>rem</tt>'
1043Instruction</a> </div>
1044<div class="doc_text">
1045<h5>Syntax:</h5>
1046<pre> &lt;result&gt; = rem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1047</pre>
1048<h5>Overview:</h5>
1049<p>The '<tt>rem</tt>' instruction returns the remainder from the
1050division of its two operands.</p>
1051<h5>Arguments:</h5>
1052<p>The two arguments to the '<tt>rem</tt>' instruction must be either <a
1053 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame^]1054values.
1055This instruction can also take <a href="#t_packed">packed</a> versions of the values.
1056Both arguments must have identical types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1057<h5>Semantics:</h5>
1058<p>This returns the <i>remainder</i> of a division (where the result
1059has the same sign as the divisor), not the <i>modulus</i> (where the
1060result has the same sign as the dividend) of a value. For more
1061information about the difference, see: <a
1062 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
1063Math Forum</a>.</p>
1064<h5>Example:</h5>
1065<pre> &lt;result&gt; = rem int 4, %var <i>; yields {int}:result = 4 % %var</i>
1066</pre>
1067</div>
1068<!-- _______________________________________________________________________ -->
1069<div class="doc_subsubsection"> <a name="i_setcc">'<tt>set<i>cc</i></tt>'
1070Instructions</a> </div>
1071<div class="doc_text">
1072<h5>Syntax:</h5>
1073<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]1074 &lt;result&gt; = setne &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1075 &lt;result&gt; = setlt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1076 &lt;result&gt; = setgt &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1077 &lt;result&gt; = setle &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1078 &lt;result&gt; = setge &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {bool}:result</i>
1079</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1080<h5>Overview:</h5>
1081<p>The '<tt>set<i>cc</i></tt>' family of instructions returns a boolean
1082value based on a comparison of their two operands.</p>
1083<h5>Arguments:</h5>
1084<p>The two arguments to the '<tt>set<i>cc</i></tt>' instructions must
1085be of <a href="#t_firstclass">first class</a> type (it is not possible
1086to compare '<tt>label</tt>'s, '<tt>array</tt>'s, '<tt>structure</tt>'
1087or '<tt>void</tt>' values, etc...). Both arguments must have identical
1088types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1089<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1090<p>The '<tt>seteq</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1091value if both operands are equal.<br>
1092The '<tt>setne</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1093value if both operands are unequal.<br>
1094The '<tt>setlt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1095value if the first operand is less than the second operand.<br>
1096The '<tt>setgt</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1097value if the first operand is greater than the second operand.<br>
1098The '<tt>setle</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1099value if the first operand is less than or equal to the second operand.<br>
1100The '<tt>setge</tt>' instruction yields a <tt>true</tt> '<tt>bool</tt>'
1101value if the first operand is greater than or equal to the second
1102operand.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1103<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1104<pre> &lt;result&gt; = seteq int 4, 5 <i>; yields {bool}:result = false</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1105 &lt;result&gt; = setne float 4, 5 <i>; yields {bool}:result = true</i>
1106 &lt;result&gt; = setlt uint 4, 5 <i>; yields {bool}:result = true</i>
1107 &lt;result&gt; = setgt sbyte 4, 5 <i>; yields {bool}:result = false</i>
1108 &lt;result&gt; = setle sbyte 4, 5 <i>; yields {bool}:result = true</i>
1109 &lt;result&gt; = setge sbyte 4, 5 <i>; yields {bool}:result = false</i>
1110</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1111</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1112<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1113<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
1114Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1115<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1116<p>Bitwise binary operators are used to do various forms of
1117bit-twiddling in a program. They are generally very efficient
1118instructions, and can commonly be strength reduced from other
1119instructions. They require two operands, execute an operation on them,
1120and produce a single value. The resulting value of the bitwise binary
1121operators is always the same type as its first operand.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1122</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1123<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1124<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
1125Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1126<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1127<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1128<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]1129</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1130<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1131<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
1132its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1133<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1134<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1135 href="#t_integral">integral</a> values. Both arguments must have
1136identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1137<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1138<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1139<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1140<div style="align: center">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1141<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1142 <tbody>
1143 <tr>
1144 <td>In0</td>
1145 <td>In1</td>
1146 <td>Out</td>
1147 </tr>
1148 <tr>
1149 <td>0</td>
1150 <td>0</td>
1151 <td>0</td>
1152 </tr>
1153 <tr>
1154 <td>0</td>
1155 <td>1</td>
1156 <td>0</td>
1157 </tr>
1158 <tr>
1159 <td>1</td>
1160 <td>0</td>
1161 <td>0</td>
1162 </tr>
1163 <tr>
1164 <td>1</td>
1165 <td>1</td>
1166 <td>1</td>
1167 </tr>
1168 </tbody>
1169</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1170</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1171<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1172<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]1173 &lt;result&gt; = and int 15, 40 <i>; yields {int}:result = 8</i>
1174 &lt;result&gt; = and int 4, 8 <i>; yields {int}:result = 0</i>
1175</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1176</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1177<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1178<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1179<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1180<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1181<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]1182</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1183<h5>Overview:</h5>
1184<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
1185or of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1186<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1187<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1188 href="#t_integral">integral</a> values. Both arguments must have
1189identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1190<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1191<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1192<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1193<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1194<table border="1" cellspacing="0" cellpadding="4">
1195 <tbody>
1196 <tr>
1197 <td>In0</td>
1198 <td>In1</td>
1199 <td>Out</td>
1200 </tr>
1201 <tr>
1202 <td>0</td>
1203 <td>0</td>
1204 <td>0</td>
1205 </tr>
1206 <tr>
1207 <td>0</td>
1208 <td>1</td>
1209 <td>1</td>
1210 </tr>
1211 <tr>
1212 <td>1</td>
1213 <td>0</td>
1214 <td>1</td>
1215 </tr>
1216 <tr>
1217 <td>1</td>
1218 <td>1</td>
1219 <td>1</td>
1220 </tr>
1221 </tbody>
1222</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1223</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1224<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1225<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]1226 &lt;result&gt; = or int 15, 40 <i>; yields {int}:result = 47</i>
1227 &lt;result&gt; = or int 4, 8 <i>; yields {int}:result = 12</i>
1228</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1229</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1230<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1231<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
1232Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1233<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1234<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1235<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]1236</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1237<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1238<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
1239or of its two operands. The <tt>xor</tt> is used to implement the
1240"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1241<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1242<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1243 href="#t_integral">integral</a> values. Both arguments must have
1244identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1245<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1246<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1247<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1248<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1249<table border="1" cellspacing="0" cellpadding="4">
1250 <tbody>
1251 <tr>
1252 <td>In0</td>
1253 <td>In1</td>
1254 <td>Out</td>
1255 </tr>
1256 <tr>
1257 <td>0</td>
1258 <td>0</td>
1259 <td>0</td>
1260 </tr>
1261 <tr>
1262 <td>0</td>
1263 <td>1</td>
1264 <td>1</td>
1265 </tr>
1266 <tr>
1267 <td>1</td>
1268 <td>0</td>
1269 <td>1</td>
1270 </tr>
1271 <tr>
1272 <td>1</td>
1273 <td>1</td>
1274 <td>0</td>
1275 </tr>
1276 </tbody>
1277</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1278</div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1279<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1280<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1281<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]1282 &lt;result&gt; = xor int 15, 40 <i>; yields {int}:result = 39</i>
1283 &lt;result&gt; = xor int 4, 8 <i>; yields {int}:result = 12</i>
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1284 &lt;result&gt; = xor int %V, -1 <i>; yields {int}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1285</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1286</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1287<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1288<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
1289Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1290<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1291<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1292<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]1293</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1294<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1295<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
1296the left a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1297<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1298<p>The first argument to the '<tt>shl</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1299 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1300type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1301<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1302<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]1303<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1304<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]1305 &lt;result&gt; = shl int 4, ubyte 2 <i>; yields {int}:result = 16</i>
1306 &lt;result&gt; = shl int 1, ubyte 10 <i>; yields {int}:result = 1024</i>
1307</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1308</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1309<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1310<div class="doc_subsubsection"> <a name="i_shr">'<tt>shr</tt>'
1311Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1312<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1313<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1314<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]1315</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1316<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1317<p>The '<tt>shr</tt>' instruction returns the first operand shifted to
1318the right a specified number of bits.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1319<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1320<p>The first argument to the '<tt>shr</tt>' instruction must be an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1321 href="#t_integer">integer</a> type. The second argument must be an '<tt>ubyte</tt>'
1322type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1323<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1324<p>If the first argument is a <a href="#t_signed">signed</a> type, the
1325most significant bit is duplicated in the newly free'd bit positions.
1326If the first argument is unsigned, zero bits shall fill the empty
1327positions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1328<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1329<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]1330 &lt;result&gt; = shr uint 4, ubyte 1 <i>; yields {uint}:result = 2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1331 &lt;result&gt; = shr int 4, ubyte 2 <i>; yields {int}:result = 1</i>
Chris Lattner8c6bb902003-06-18 21:30:51 +0000[diff] [blame]1332 &lt;result&gt; = shr sbyte 4, ubyte 3 <i>; yields {sbyte}:result = 0</i>
1333 &lt;result&gt; = shr sbyte -2, ubyte 1 <i>; yields {sbyte}:result = -1</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1334</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1335</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1336<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1337<div class="doc_subsection"> <a name="memoryops">Memory Access
1338Operations</a></div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1339<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1340<p>A key design point of an SSA-based representation is how it
1341represents memory. In LLVM, no memory locations are in SSA form, which
1342makes things very simple. This section describes how to read, write,
1343allocate and free memory in LLVM.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1344</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1345<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1346<div class="doc_subsubsection"> <a name="i_malloc">'<tt>malloc</tt>'
1347Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1348<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1349<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1350<pre> &lt;result&gt; = malloc &lt;type&gt;, uint &lt;NumElements&gt; <i>; yields {type*}:result</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1351 &lt;result&gt; = malloc &lt;type&gt; <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1352</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1353<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1354<p>The '<tt>malloc</tt>' instruction allocates memory from the system
1355heap and returns a pointer to it.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1356<h5>Arguments:</h5>
John Criswell6e4ca612004-02-24 16:13:56 +0000[diff] [blame]1357<p>The '<tt>malloc</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
1358bytes of memory from the operating system and returns a pointer of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1359appropriate type to the program. The second form of the instruction is
1360a shorter version of the first instruction that defaults to allocating
1361one element.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1362<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1363<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1364<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
1365a pointer is returned.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1366<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1367<pre> %array = malloc [4 x ubyte ] <i>; yields {[%4 x ubyte]*}:array</i>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1368
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1369 %size = <a
1370 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]1371 %array1 = malloc ubyte, uint 4 <i>; yields {ubyte*}:array1</i>
1372 %array2 = malloc [12 x ubyte], uint %size <i>; yields {[12 x ubyte]*}:array2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1373</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1374</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1375<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1376<div class="doc_subsubsection"> <a name="i_free">'<tt>free</tt>'
1377Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1378<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1379<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1380<pre> free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1381</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1382<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1383<p>The '<tt>free</tt>' instruction returns memory back to the unused
1384memory heap, to be reallocated in the future.</p>
1385<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1386<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1387<p>'<tt>value</tt>' shall be a pointer value that points to a value
1388that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
1389instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1390<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1391<p>Access to the memory pointed to by the pointer is not longer defined
1392after this instruction executes.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1393<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1394<pre> %array = <a href="#i_malloc">malloc</a> [4 x ubyte] <i>; yields {[4 x ubyte]*}:array</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1395 free [4 x ubyte]* %array
1396</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1397</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1398<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1399<div class="doc_subsubsection"> <a name="i_alloca">'<tt>alloca</tt>'
1400Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1401<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1402<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1403<pre> &lt;result&gt; = alloca &lt;type&gt;, uint &lt;NumElements&gt; <i>; yields {type*}:result</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1404 &lt;result&gt; = alloca &lt;type&gt; <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1405</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1406<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1407<p>The '<tt>alloca</tt>' instruction allocates memory on the current
1408stack frame of the procedure that is live until the current function
1409returns to its caller.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1410<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1411<p>The the '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
1412bytes of memory on the runtime stack, returning a pointer of the
1413appropriate type to the program. The second form of the instruction is
1414a shorter version of the first that defaults to allocating one element.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1415<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1416<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1417<p>Memory is allocated, a pointer is returned. '<tt>alloca</tt>'d
1418memory is automatically released when the function returns. The '<tt>alloca</tt>'
1419instruction is commonly used to represent automatic variables that must
1420have an address available. When the function returns (either with the <tt><a
1421 href="#i_ret">ret</a></tt> or <tt><a href="#i_invoke">invoke</a></tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1422instructions), the memory is reclaimed.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1423<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1424<pre> %ptr = alloca int <i>; yields {int*}:ptr</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1425 %ptr = alloca int, uint 4 <i>; yields {int*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1426</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1427</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1428<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1429<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
1430Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1431<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1432<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1433<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]1434<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1435<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1436<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1437<p>The argument to the '<tt>load</tt>' instruction specifies the memory
1438address to load from. The pointer must point to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1439 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1440marked as <tt>volatile</tt> then the optimizer is not allowed to modify
1441the number or order of execution of this <tt>load</tt> with other
1442volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
1443instructions. </p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1444<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1445<p>The location of memory pointed to is loaded.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1446<h5>Examples:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1447<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1448 <a
1449 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1450 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1451</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1452</div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1453<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1454<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
1455Instruction</a> </div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1456<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1457<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]1458 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]1459</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1460<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1461<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1462<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1463<p>There are two arguments to the '<tt>store</tt>' instruction: a value
1464to store and an address to store it into. The type of the '<tt>&lt;pointer&gt;</tt>'
1465operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
1466operand. If the <tt>store</tt> is marked as <tt>volatile</tt> then the
1467optimizer is not allowed to modify the number or order of execution of
1468this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
1469 href="#i_store">store</a></tt> instructions.</p>
1470<h5>Semantics:</h5>
1471<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
1472at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1473<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1474<pre> %ptr = <a href="#i_alloca">alloca</a> int <i>; yields {int*}:ptr</i>
1475 <a
1476 href="#i_store">store</a> int 3, int* %ptr <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1477 %val = load int* %ptr <i>; yields {int}:val = int 3</i>
1478</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]1479<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1480<div class="doc_subsubsection">
1481 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
1482</div>
1483
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1484<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1485<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1486<pre>
1487 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
1488</pre>
1489
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1490<h5>Overview:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1491
1492<p>
1493The '<tt>getelementptr</tt>' instruction is used to get the address of a
1494subelement of an aggregate data structure.</p>
1495
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1496<h5>Arguments:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1497
1498<p>This instruction takes a list of integer constants that indicate what
1499elements of the aggregate object to index to. The actual types of the arguments
1500provided depend on the type of the first pointer argument. The
1501'<tt>getelementptr</tt>' instruction is used to index down through the type
1502levels of a structure. When indexing into a structure, only <tt>uint</tt>
1503integer constants are allowed. When indexing into an array or pointer
1504<tt>int</tt> and <tt>long</tt> indexes are allowed of any sign.</p>
1505
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1506<p>For example, let's consider a C code fragment and how it gets
1507compiled to LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1508
1509<pre>
1510 struct RT {
1511 char A;
1512 int B[10][20];
1513 char C;
1514 };
1515 struct ST {
1516 int X;
1517 double Y;
1518 struct RT Z;
1519 };
1520
1521 int *foo(struct ST *s) {
1522 return &amp;s[1].Z.B[5][13];
1523 }
1524</pre>
1525
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1526<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1527
1528<pre>
1529 %RT = type { sbyte, [10 x [20 x int]], sbyte }
1530 %ST = type { int, double, %RT }
1531
Brian Gaeke7283e7c2004-07-02 21:08:14 +0000[diff] [blame]1532 implementation
1533
1534 int* %foo(%ST* %s) {
1535 entry:
1536 %reg = getelementptr %ST* %s, int 1, uint 2, uint 1, int 5, int 13
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1537 ret int* %reg
1538 }
1539</pre>
1540
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1541<h5>Semantics:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1542
1543<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]1544on the pointer type that is being index into. <a href="#t_pointer">Pointer</a>
1545and <a href="#t_array">array</a> types require <tt>uint</tt>, <tt>int</tt>,
1546<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]1547types require <tt>uint</tt> <b>constants</b>.</p>
1548
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1549<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]1550type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ int, double, %RT
1551}</tt>' type, a structure. The second index indexes into the third element of
1552the structure, yielding a '<tt>%RT</tt>' = '<tt>{ sbyte, [10 x [20 x int]],
1553sbyte }</tt>' type, another structure. The third index indexes into the second
1554element of the structure, yielding a '<tt>[10 x [20 x int]]</tt>' type, an
1555array. The two dimensions of the array are subscripted into, yielding an
1556'<tt>int</tt>' type. The '<tt>getelementptr</tt>' instruction return a pointer
1557to this element, thus computing a value of '<tt>int*</tt>' type.</p>
1558
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1559<p>Note that it is perfectly legal to index partially through a
1560structure, returning a pointer to an inner element. Because of this,
1561the LLVM code for the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1562
1563<pre>
1564 int* "foo"(%ST* %s) {
1565 %t1 = getelementptr %ST* %s, int 1 <i>; yields %ST*:%t1</i>
1566 %t2 = getelementptr %ST* %t1, int 0, uint 2 <i>; yields %RT*:%t2</i>
1567 %t3 = getelementptr %RT* %t2, int 0, uint 1 <i>; yields [10 x [20 x int]]*:%t3</i>
1568 %t4 = getelementptr [10 x [20 x int]]* %t3, int 0, int 5 <i>; yields [20 x int]*:%t4</i>
1569 %t5 = getelementptr [20 x int]* %t4, int 0, int 13 <i>; yields int*:%t5</i>
1570 ret int* %t5
1571 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1572</pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1573<h5>Example:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]1574<pre>
1575 <i>; yields [12 x ubyte]*:aptr</i>
1576 %aptr = getelementptr {int, [12 x ubyte]}* %sptr, long 0, uint 1
1577</pre>
1578
1579</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1580<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1581<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1582<div class="doc_text">
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]1583<p>The instructions in this category are the "miscellaneous"
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1584instructions, which defy better classification.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1585</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1586<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1587<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
1588Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1589<div class="doc_text">
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]1590<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1591<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]1592<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1593<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
1594the SSA graph representing the function.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]1595<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1596<p>The type of the incoming values are specified with the first type
1597field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
1598as arguments, with one pair for each predecessor basic block of the
1599current block. Only values of <a href="#t_firstclass">first class</a>
1600type may be used as the value arguments to the PHI node. Only labels
1601may be used as the label arguments.</p>
1602<p>There must be no non-phi instructions between the start of a basic
1603block and the PHI instructions: i.e. PHI instructions must be first in
1604a basic block.</p>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]1605<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1606<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the
1607value specified by the parameter, depending on which basic block we
1608came from in the last <a href="#terminators">terminator</a> instruction.</p>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1609<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1610<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]1611</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1612
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1613<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1614<div class="doc_subsubsection">
1615 <a name="i_cast">'<tt>cast .. to</tt>' Instruction</a>
1616</div>
1617
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1618<div class="doc_text">
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1619
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1620<h5>Syntax:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1621
1622<pre>
1623 &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]1624</pre>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1625
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1626<h5>Overview:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1627
1628<p>
1629The '<tt>cast</tt>' instruction is used as the primitive means to convert
1630integers to floating point, change data type sizes, and break type safety (by
1631casting pointers).
1632</p>
1633
1634
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1635<h5>Arguments:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1636
1637<p>
1638The '<tt>cast</tt>' instruction takes a value to cast, which must be a first
1639class value, and a type to cast it to, which must also be a <a
1640href="#t_firstclass">first class</a> type.
1641</p>
1642
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1643<h5>Semantics:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1644
1645<p>
1646This instruction follows the C rules for explicit casts when determining how the
1647data being cast must change to fit in its new container.
1648</p>
1649
1650<p>
1651When casting to bool, any value that would be considered true in the context of
1652a C '<tt>if</tt>' condition is converted to the boolean '<tt>true</tt>' values,
1653all else are '<tt>false</tt>'.
1654</p>
1655
1656<p>
1657When extending an integral value from a type of one signness to another (for
1658example '<tt>sbyte</tt>' to '<tt>ulong</tt>'), the value is sign-extended if the
1659<b>source</b> value is signed, and zero-extended if the source value is
1660unsigned. <tt>bool</tt> values are always zero extended into either zero or
1661one.
1662</p>
1663
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]1664<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1665
1666<pre>
1667 %X = cast int 257 to ubyte <i>; yields ubyte:1</i>
Chris Lattner7bae3952002-06-25 18:03:17 +0000[diff] [blame]1668 %Y = cast int 123 to bool <i>; yields bool:true</i>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]1669</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1670</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]1671
1672<!-- _______________________________________________________________________ -->
1673<div class="doc_subsubsection">
1674 <a name="i_select">'<tt>select</tt>' Instruction</a>
1675</div>
1676
1677<div class="doc_text">
1678
1679<h5>Syntax:</h5>
1680
1681<pre>
1682 &lt;result&gt; = select bool &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
1683</pre>
1684
1685<h5>Overview:</h5>
1686
1687<p>
1688The '<tt>select</tt>' instruction is used to choose one value based on a
1689condition, without branching.
1690</p>
1691
1692
1693<h5>Arguments:</h5>
1694
1695<p>
1696The '<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.
1697</p>
1698
1699<h5>Semantics:</h5>
1700
1701<p>
1702If the boolean condition evaluates to true, the instruction returns the first
1703value argument, otherwise it returns the second value argument.
1704</p>
1705
1706<h5>Example:</h5>
1707
1708<pre>
1709 %X = select bool true, ubyte 17, ubyte 42 <i>; yields ubyte:17</i>
1710</pre>
1711</div>
1712
1713
1714
1715
1716
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]1717<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1718<div class="doc_subsubsection"> <a name="i_call">'<tt>call</tt>'
1719Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1720<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1721<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1722<pre> &lt;result&gt; = call &lt;ty&gt;* &lt;fnptrval&gt;(&lt;param list&gt;)<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1723<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1724<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1725<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1726<p>This instruction requires several arguments:</p>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1727<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1728 <li>
1729 <p>'<tt>ty</tt>': shall be the signature of the pointer to function
1730value being invoked. The argument types must match the types implied
1731by this signature.</p>
1732 </li>
1733 <li>
1734 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a
1735function to be invoked. In most cases, this is a direct function
1736invocation, but indirect <tt>call</tt>s are just as possible,
1737calling an arbitrary pointer to function values.</p>
1738 </li>
1739 <li>
1740 <p>'<tt>function args</tt>': argument list whose types match the
1741function signature argument types. If the function signature
1742indicates the function accepts a variable number of arguments, the
1743extra arguments can be specified.</p>
1744 </li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1745</ol>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1746<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1747<p>The '<tt>call</tt>' instruction is used to cause control flow to
1748transfer to a specified function, with its incoming arguments bound to
1749the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
1750instruction in the called function, control flow continues with the
1751instruction after the function call, and the return value of the
1752function is bound to the result argument. This is a simpler case of
1753the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1754<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1755<pre> %retval = call int %test(int %argc)<br> call int(sbyte*, ...) *%printf(sbyte* %msg, int 12, sbyte 42);<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1756</div>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1757<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1758<div class="doc_subsubsection"> <a name="i_vanext">'<tt>vanext</tt>'
1759Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1760<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1761<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1762<pre> &lt;resultarglist&gt; = vanext &lt;va_list&gt; &lt;arglist&gt;, &lt;argty&gt;<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1763<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1764<p>The '<tt>vanext</tt>' instruction is used to access arguments passed
1765through the "variable argument" area of a function call. It is used to
1766implement the <tt>va_arg</tt> macro in C.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1767<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1768<p>This instruction takes a <tt>valist</tt> value and the type of the
1769argument. It returns another <tt>valist</tt>.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1770<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1771<p>The '<tt>vanext</tt>' instruction advances the specified <tt>valist</tt>
1772past an argument of the specified type. In conjunction with the <a
1773 href="#i_vaarg"><tt>vaarg</tt></a> instruction, it is used to implement
1774the <tt>va_arg</tt> macro available in C. For more information, see
1775the variable argument handling <a href="#int_varargs">Intrinsic
1776Functions</a>.</p>
1777<p>It is legal for this instruction to be called in a function which
1778does not take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1779function.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1780<p><tt>vanext</tt> is an LLVM instruction instead of an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1781 href="#intrinsics">intrinsic function</a> because it takes an type as
1782an argument.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1783<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1784<p>See the <a href="#int_varargs">variable argument processing</a>
1785section.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1786</div>
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]1787<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1788<div class="doc_subsubsection"> <a name="i_vaarg">'<tt>vaarg</tt>'
1789Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1790<div class="doc_text">
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]1791<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1792<pre> &lt;resultval&gt; = vaarg &lt;va_list&gt; &lt;arglist&gt;, &lt;argty&gt;<br></pre>
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]1793<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1794<p>The '<tt>vaarg</tt>' instruction is used to access arguments passed
1795through the "variable argument" area of a function call. It is used to
1796implement the <tt>va_arg</tt> macro in C.</p>
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]1797<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1798<p>This instruction takes a <tt>valist</tt> value and the type of the
1799argument. It returns a value of the specified argument type.</p>
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]1800<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1801<p>The '<tt>vaarg</tt>' instruction loads an argument of the specified
1802type from the specified <tt>va_list</tt>. In conjunction with the <a
1803 href="#i_vanext"><tt>vanext</tt></a> instruction, it is used to
1804implement the <tt>va_arg</tt> macro available in C. For more
1805information, see the variable argument handling <a href="#int_varargs">Intrinsic
1806Functions</a>.</p>
1807<p>It is legal for this instruction to be called in a function which
1808does not take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1809function.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1810<p><tt>vaarg</tt> is an LLVM instruction instead of an <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1811 href="#intrinsics">intrinsic function</a> because it takes an type as
1812an argument.</p>
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]1813<h5>Example:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1814<p>See the <a href="#int_varargs">variable argument processing</a>
1815section.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1816</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1817
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1818<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1819<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
1820<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1821
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1822<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]1823
1824<p>LLVM supports the notion of an "intrinsic function". These functions have
1825well known names and semantics, and are required to follow certain
1826restrictions. Overall, these instructions represent an extension mechanism for
1827the LLVM language that does not require changing all of the transformations in
1828LLVM to add to the language (or the bytecode reader/writer, the parser,
1829etc...).</p>
1830
1831<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix, this
1832prefix is reserved in LLVM for intrinsic names, thus functions may not be named
1833this. Intrinsic functions must always be external functions: you cannot define
1834the body of intrinsic functions. Intrinsic functions may only be used in call
1835or invoke instructions: it is illegal to take the address of an intrinsic
1836function. Additionally, because intrinsic functions are part of the LLVM
1837language, it is required that they all be documented here if any are added.</p>
1838
1839
1840<p>
1841Adding an intrinsic to LLVM is straight-forward if it is possible to express the
1842concept in LLVM directly (ie, code generator support is not _required_). To do
1843this, extend the default implementation of the IntrinsicLowering class to handle
1844the intrinsic. Code generators use this class to lower intrinsics they do not
1845understand to raw LLVM instructions that they do.
1846</p>
1847
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1848</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1849
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1850<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1851<div class="doc_subsection">
1852 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
1853</div>
1854
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1855<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1856
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1857<p>Variable argument support is defined in LLVM with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1858 href="#i_vanext"><tt>vanext</tt></a> instruction and these three
1859intrinsic functions. These functions are related to the similarly
1860named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1861
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1862<p>All of these functions operate on arguments that use a
1863target-specific value type "<tt>va_list</tt>". The LLVM assembly
1864language reference manual does not define what this type is, so all
1865transformations should be prepared to handle intrinsics with any type
1866used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1867
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1868<p>This example shows how the <a href="#i_vanext"><tt>vanext</tt></a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1869instruction and the variable argument handling intrinsic functions are
1870used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1871
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]1872<pre>
1873int %test(int %X, ...) {
1874 ; Initialize variable argument processing
1875 %ap = call sbyte* %<a href="#i_va_start">llvm.va_start</a>()
1876
1877 ; Read a single integer argument
1878 %tmp = vaarg sbyte* %ap, int
1879
1880 ; Advance to the next argument
1881 %ap2 = vanext sbyte* %ap, int
1882
1883 ; Demonstrate usage of llvm.va_copy and llvm.va_end
1884 %aq = call sbyte* %<a href="#i_va_copy">llvm.va_copy</a>(sbyte* %ap2)
1885 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte* %aq)
1886
1887 ; Stop processing of arguments.
1888 call void %<a href="#i_va_end">llvm.va_end</a>(sbyte* %ap2)
1889 ret int %tmp
1890}
1891</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1892</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1893
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1894<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1895<div class="doc_subsubsection">
1896 <a name="i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
1897</div>
1898
1899
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1900<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1901<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1902<pre> call va_list ()* %llvm.va_start()<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1903<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1904<p>The '<tt>llvm.va_start</tt>' intrinsic returns a new <tt>&lt;arglist&gt;</tt>
1905for subsequent use by the variable argument intrinsics.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1906<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1907<p>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1908macro available in C. In a target-dependent way, it initializes and
1909returns a <tt>va_list</tt> element, so that the next <tt>vaarg</tt>
1910will produce the first variable argument passed to the function. Unlike
1911the C <tt>va_start</tt> macro, this intrinsic does not need to know the
1912last argument of the function, the compiler can figure that out.</p>
1913<p>Note that this intrinsic function is only legal to be called from
1914within the body of a variable argument function.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1915</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1916
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1917<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1918<div class="doc_subsubsection">
1919 <a name="i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
1920</div>
1921
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1922<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1923<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1924<pre> call void (va_list)* %llvm.va_end(va_list &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1925<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1926<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>&lt;arglist&gt;</tt>
1927which has been initialized previously with <tt><a href="#i_va_start">llvm.va_start</a></tt>
1928or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1929<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1930<p>The argument is a <tt>va_list</tt> to destroy.</p>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1931<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1932<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]1933macro available in C. In a target-dependent way, it destroys the <tt>va_list</tt>.
1934Calls to <a href="#i_va_start"><tt>llvm.va_start</tt></a> and <a
1935 href="#i_va_copy"><tt>llvm.va_copy</tt></a> must be matched exactly
1936with calls to <tt>llvm.va_end</tt>.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1937</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1938
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1939<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1940<div class="doc_subsubsection">
1941 <a name="i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
1942</div>
1943
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1944<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1945
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1946<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1947
1948<pre>
1949 call va_list (va_list)* %llvm.va_copy(va_list &lt;destarglist&gt;)
1950</pre>
1951
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1952<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1953
1954<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
1955from the source argument list to the destination argument list.</p>
1956
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1957<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1958
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1959<p>The argument is the <tt>va_list</tt> to copy.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1960
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]1961<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1962
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1963<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1964macro available in C. In a target-dependent way, it copies the source
1965<tt>va_list</tt> element into the returned list. This intrinsic is necessary
Chris Lattnerfcd37252004-06-21 22:52:48 +0000[diff] [blame]1966because the <tt><a href="#i_va_start">llvm.va_start</a></tt> intrinsic may be
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1967arbitrarily complex and require memory allocation, for example.</p>
1968
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1969</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]1970
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]1971<!-- ======================================================================= -->
1972<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]1973 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
1974</div>
1975
1976<div class="doc_text">
1977
1978<p>
1979LLVM support for <a href="GarbageCollection.html">Accurate Garbage
1980Collection</a> requires the implementation and generation of these intrinsics.
1981These intrinsics allow identification of <a href="#i_gcroot">GC roots on the
1982stack</a>, as well as garbage collector implementations that require <a
1983href="#i_gcread">read</a> and <a href="#i_gcwrite">write</a> barriers.
1984Front-ends for type-safe garbage collected languages should generate these
1985intrinsics to make use of the LLVM garbage collectors. For more details, see <a
1986href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
1987</p>
1988</div>
1989
1990<!-- _______________________________________________________________________ -->
1991<div class="doc_subsubsection">
1992 <a name="i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
1993</div>
1994
1995<div class="doc_text">
1996
1997<h5>Syntax:</h5>
1998
1999<pre>
2000 call void (&lt;ty&gt;**, &lt;ty2&gt;*)* %llvm.gcroot(&lt;ty&gt;** %ptrloc, &lt;ty2&gt;* %metadata)
2001</pre>
2002
2003<h5>Overview:</h5>
2004
2005<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existance of a GC root to
2006the code generator, and allows some metadata to be associated with it.</p>
2007
2008<h5>Arguments:</h5>
2009
2010<p>The first argument specifies the address of a stack object that contains the
2011root pointer. The second pointer (which must be either a constant or a global
2012value address) contains the meta-data to be associated with the root.</p>
2013
2014<h5>Semantics:</h5>
2015
2016<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
2017location. At compile-time, the code generator generates information to allow
2018the runtime to find the pointer at GC safe points.
2019</p>
2020
2021</div>
2022
2023
2024<!-- _______________________________________________________________________ -->
2025<div class="doc_subsubsection">
2026 <a name="i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
2027</div>
2028
2029<div class="doc_text">
2030
2031<h5>Syntax:</h5>
2032
2033<pre>
2034 call sbyte* (sbyte**)* %llvm.gcread(sbyte** %Ptr)
2035</pre>
2036
2037<h5>Overview:</h5>
2038
2039<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
2040locations, allowing garbage collector implementations that require read
2041barriers.</p>
2042
2043<h5>Arguments:</h5>
2044
2045<p>The argument is the address to read from, which should be an address
2046allocated from the garbage collector.</p>
2047
2048<h5>Semantics:</h5>
2049
2050<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
2051instruction, but may be replaced with substantially more complex code by the
2052garbage collector runtime, as needed.</p>
2053
2054</div>
2055
2056
2057<!-- _______________________________________________________________________ -->
2058<div class="doc_subsubsection">
2059 <a name="i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
2060</div>
2061
2062<div class="doc_text">
2063
2064<h5>Syntax:</h5>
2065
2066<pre>
2067 call void (sbyte*, sbyte**)* %llvm.gcwrite(sbyte* %P1, sbyte** %P2)
2068</pre>
2069
2070<h5>Overview:</h5>
2071
2072<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
2073locations, allowing garbage collector implementations that require write
2074barriers (such as generational or reference counting collectors).</p>
2075
2076<h5>Arguments:</h5>
2077
2078<p>The first argument is the reference to store, and the second is the heap
2079location to store to.</p>
2080
2081<h5>Semantics:</h5>
2082
2083<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
2084instruction, but may be replaced with substantially more complex code by the
2085garbage collector runtime, as needed.</p>
2086
2087</div>
2088
2089
2090
2091<!-- ======================================================================= -->
2092<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2093 <a name="int_codegen">Code Generator Intrinsics</a>
2094</div>
2095
2096<div class="doc_text">
2097<p>
2098These intrinsics are provided by LLVM to expose special features that may only
2099be implemented with code generator support.
2100</p>
2101
2102</div>
2103
2104<!-- _______________________________________________________________________ -->
2105<div class="doc_subsubsection">
2106 <a name="i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
2107</div>
2108
2109<div class="doc_text">
2110
2111<h5>Syntax:</h5>
2112<pre>
2113 call void* ()* %llvm.returnaddress(uint &lt;level&gt;)
2114</pre>
2115
2116<h5>Overview:</h5>
2117
2118<p>
2119The '<tt>llvm.returnaddress</tt>' intrinsic returns a target-specific value
2120indicating the return address of the current function or one of its callers.
2121</p>
2122
2123<h5>Arguments:</h5>
2124
2125<p>
2126The argument to this intrinsic indicates which function to return the address
2127for. Zero indicates the calling function, one indicates its caller, etc. The
2128argument is <b>required</b> to be a constant integer value.
2129</p>
2130
2131<h5>Semantics:</h5>
2132
2133<p>
2134The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
2135the return address of the specified call frame, or zero if it cannot be
2136identified. The value returned by this intrinsic is likely to be incorrect or 0
2137for arguments other than zero, so it should only be used for debugging purposes.
2138</p>
2139
2140<p>
2141Note that calling this intrinsic does not prevent function inlining or other
2142aggressive transformations, so the value returned may not that of the obvious
2143source-language caller.
2144</p>
2145</div>
2146
2147
2148<!-- _______________________________________________________________________ -->
2149<div class="doc_subsubsection">
2150 <a name="i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
2151</div>
2152
2153<div class="doc_text">
2154
2155<h5>Syntax:</h5>
2156<pre>
2157 call void* ()* %llvm.frameaddress(uint &lt;level&gt;)
2158</pre>
2159
2160<h5>Overview:</h5>
2161
2162<p>
2163The '<tt>llvm.frameaddress</tt>' intrinsic returns the target-specific frame
2164pointer value for the specified stack frame.
2165</p>
2166
2167<h5>Arguments:</h5>
2168
2169<p>
2170The argument to this intrinsic indicates which function to return the frame
2171pointer for. Zero indicates the calling function, one indicates its caller,
2172etc. The argument is <b>required</b> to be a constant integer value.
2173</p>
2174
2175<h5>Semantics:</h5>
2176
2177<p>
2178The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
2179the frame address of the specified call frame, or zero if it cannot be
2180identified. The value returned by this intrinsic is likely to be incorrect or 0
2181for arguments other than zero, so it should only be used for debugging purposes.
2182</p>
2183
2184<p>
2185Note that calling this intrinsic does not prevent function inlining or other
2186aggressive transformations, so the value returned may not that of the obvious
2187source-language caller.
2188</p>
2189</div>
2190
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2191<!-- ======================================================================= -->
2192<div class="doc_subsection">
2193 <a name="int_os">Operating System Intrinsics</a>
2194</div>
2195
2196<div class="doc_text">
2197<p>
2198These intrinsics are provided by LLVM to support the implementation of
2199operating system level code.
2200</p>
2201
2202</div>
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2203
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2204<!-- _______________________________________________________________________ -->
2205<div class="doc_subsubsection">
2206 <a name="i_readport">'<tt>llvm.readport</tt>' Intrinsic</a>
2207</div>
2208
2209<div class="doc_text">
2210
2211<h5>Syntax:</h5>
2212<pre>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2213 call &lt;integer type&gt; (&lt;integer type&gt;)* %llvm.readport (&lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2214</pre>
2215
2216<h5>Overview:</h5>
2217
2218<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2219The '<tt>llvm.readport</tt>' intrinsic reads data from the specified hardware
2220I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2221</p>
2222
2223<h5>Arguments:</h5>
2224
2225<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2226The argument to this intrinsic indicates the hardware I/O address from which
2227to read the data. The address is in the hardware I/O address namespace (as
2228opposed to being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2229</p>
2230
2231<h5>Semantics:</h5>
2232
2233<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2234The '<tt>llvm.readport</tt>' intrinsic reads data from the hardware I/O port
2235specified by <i>address</i> and returns the value. The address and return
2236value must be integers, but the size is dependent upon the platform upon which
2237the program is code generated. For example, on x86, the address must be an
2238unsigned 16 bit value, and the return value must be 8, 16, or 32 bits.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2239</p>
2240
2241</div>
2242
2243<!-- _______________________________________________________________________ -->
2244<div class="doc_subsubsection">
2245 <a name="i_writeport">'<tt>llvm.writeport</tt>' Intrinsic</a>
2246</div>
2247
2248<div class="doc_text">
2249
2250<h5>Syntax:</h5>
2251<pre>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2252 call void (&lt;integer type&gt;, &lt;integer type&gt;)* %llvm.writeport (&lt;integer type&gt; &lt;value&gt;, &lt;integer type&gt; &lt;address&gt;)
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2253</pre>
2254
2255<h5>Overview:</h5>
2256
2257<p>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2258The '<tt>llvm.writeport</tt>' intrinsic writes data to the specified hardware
2259I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2260</p>
2261
2262<h5>Arguments:</h5>
2263
2264<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2265The first argument is the value to write to the I/O port.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2266</p>
2267
2268<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2269The second argument indicates the hardware I/O address to which data should be
2270written. The address is in the hardware I/O address namespace (as opposed to
2271being a memory location for memory mapped I/O).
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2272</p>
2273
2274<h5>Semantics:</h5>
2275
2276<p>
2277The '<tt>llvm.writeport</tt>' intrinsic writes <i>value</i> to the I/O port
2278specified by <i>address</i>. The address and value must be integers, but the
2279size is dependent upon the platform upon which the program is code generated.
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]2280For example, on x86, the address must be an unsigned 16 bit value, and the
2281value written must be 8, 16, or 32 bits in length.
John Criswellcfd3bac2004-04-09 15:23:37 +0000[diff] [blame]2282</p>
2283
2284</div>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2285
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2286<!-- _______________________________________________________________________ -->
2287<div class="doc_subsubsection">
2288 <a name="i_readio">'<tt>llvm.readio</tt>' Intrinsic</a>
2289</div>
2290
2291<div class="doc_text">
2292
2293<h5>Syntax:</h5>
2294<pre>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2295 call &lt;result&gt; (&lt;ty&gt;*)* %llvm.readio (&lt;ty&gt; * &lt;pointer&gt;)
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2296</pre>
2297
2298<h5>Overview:</h5>
2299
2300<p>
2301The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
2302address.
2303</p>
2304
2305<h5>Arguments:</h5>
2306
2307<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2308The argument to this intrinsic is a pointer indicating the memory address from
2309which to read the data. The data must be a
2310<a href="#t_firstclass">first class</a> type.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2311</p>
2312
2313<h5>Semantics:</h5>
2314
2315<p>
2316The '<tt>llvm.readio</tt>' intrinsic reads data from a memory mapped I/O
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2317location specified by <i>pointer</i> and returns the value. The argument must
2318be a pointer, and the return value must be a
2319<a href="#t_firstclass">first class</a> type. However, certain architectures
2320may not support I/O on all first class types. For example, 32 bit processors
2321may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2322</p>
2323
2324<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2325This intrinsic enforces an in-order memory model for llvm.readio and
2326llvm.writeio calls on machines that use dynamic scheduling. Dynamically
2327scheduled processors may execute loads and stores out of order, re-ordering at
2328run time accesses to memory mapped I/O registers. Using these intrinsics
2329ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2330</p>
2331
2332</div>
2333
2334<!-- _______________________________________________________________________ -->
2335<div class="doc_subsubsection">
2336 <a name="i_writeio">'<tt>llvm.writeio</tt>' Intrinsic</a>
2337</div>
2338
2339<div class="doc_text">
2340
2341<h5>Syntax:</h5>
2342<pre>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2343 call void (&lt;ty1&gt;, &lt;ty2&gt;*)* %llvm.writeio (&lt;ty1&gt; &lt;value&gt;, &lt;ty2&gt; * &lt;pointer&gt;)
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2344</pre>
2345
2346<h5>Overview:</h5>
2347
2348<p>
2349The '<tt>llvm.writeio</tt>' intrinsic writes data to the specified memory
2350mapped I/O address.
2351</p>
2352
2353<h5>Arguments:</h5>
2354
2355<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2356The first argument is the value to write to the memory mapped I/O location.
2357The second argument is a pointer indicating the memory address to which the
2358data should be written.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2359</p>
2360
2361<h5>Semantics:</h5>
2362
2363<p>
2364The '<tt>llvm.writeio</tt>' intrinsic writes <i>value</i> to the memory mapped
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2365I/O address specified by <i>pointer</i>. The value must be a
2366<a href="#t_firstclass">first class</a> type. However, certain architectures
2367may not support I/O on all first class types. For example, 32 bit processors
2368may only support I/O on data types that are 32 bits or less.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2369</p>
2370
2371<p>
John Criswell96db6fc2004-04-12 16:33:19 +0000[diff] [blame]2372This intrinsic enforces an in-order memory model for llvm.readio and
2373llvm.writeio calls on machines that use dynamic scheduling. Dynamically
2374scheduled processors may execute loads and stores out of order, re-ordering at
2375run time accesses to memory mapped I/O registers. Using these intrinsics
2376ensures that accesses to memory mapped I/O registers occur in program order.
John Criswell183402a2004-04-12 15:02:16 +0000[diff] [blame]2377</p>
2378
2379</div>
2380
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2381<!-- ======================================================================= -->
2382<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2383 <a name="int_libc">Standard C Library Intrinsics</a>
2384</div>
2385
2386<div class="doc_text">
2387<p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2388LLVM provides intrinsics for a few important standard C library functions.
2389These intrinsics allow source-language front-ends to pass information about the
2390alignment of the pointer arguments to the code generator, providing opportunity
2391for more efficient code generation.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2392</p>
2393
2394</div>
2395
2396<!-- _______________________________________________________________________ -->
2397<div class="doc_subsubsection">
2398 <a name="i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
2399</div>
2400
2401<div class="doc_text">
2402
2403<h5>Syntax:</h5>
2404<pre>
2405 call void (sbyte*, sbyte*, uint, uint)* %llvm.memcpy(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
2406 uint &lt;len&gt;, uint &lt;align&gt;)
2407</pre>
2408
2409<h5>Overview:</h5>
2410
2411<p>
2412The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
2413location to the destination location.
2414</p>
2415
2416<p>
2417Note that, unlike the standard libc function, the <tt>llvm.memcpy</tt> intrinsic
2418does not return a value, and takes an extra alignment argument.
2419</p>
2420
2421<h5>Arguments:</h5>
2422
2423<p>
2424The first argument is a pointer to the destination, the second is a pointer to
2425the source. The third argument is an (arbitrarily sized) integer argument
2426specifying the number of bytes to copy, and the fourth argument is the alignment
2427of the source and destination locations.
2428</p>
2429
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]2430<p>
2431If the call to this intrinisic has an alignment value that is not 0 or 1, then
2432the caller guarantees that the size of the copy is a multiple of the alignment
2433and that both the source and destination pointers are aligned to that boundary.
2434</p>
2435
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]2436<h5>Semantics:</h5>
2437
2438<p>
2439The '<tt>llvm.memcpy</tt>' intrinsic copies a block of memory from the source
2440location to the destination location, which are not allowed to overlap. It
2441copies "len" bytes of memory over. If the argument is known to be aligned to
2442some boundary, this can be specified as the fourth argument, otherwise it should
2443be set to 0 or 1.
2444</p>
2445</div>
2446
2447
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]2448<!-- _______________________________________________________________________ -->
2449<div class="doc_subsubsection">
2450 <a name="i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
2451</div>
2452
2453<div class="doc_text">
2454
2455<h5>Syntax:</h5>
2456<pre>
2457 call void (sbyte*, sbyte*, uint, uint)* %llvm.memmove(sbyte* &lt;dest&gt;, sbyte* &lt;src&gt;,
2458 uint &lt;len&gt;, uint &lt;align&gt;)
2459</pre>
2460
2461<h5>Overview:</h5>
2462
2463<p>
2464The '<tt>llvm.memmove</tt>' intrinsic moves a block of memory from the source
2465location to the destination location. It is similar to the '<tt>llvm.memcpy</tt>'
2466intrinsic but allows the two memory locations to overlap.
2467</p>
2468
2469<p>
2470Note that, unlike the standard libc function, the <tt>llvm.memmove</tt> intrinsic
2471does not return a value, and takes an extra alignment argument.
2472</p>
2473
2474<h5>Arguments:</h5>
2475
2476<p>
2477The first argument is a pointer to the destination, the second is a pointer to
2478the source. The third argument is an (arbitrarily sized) integer argument
2479specifying the number of bytes to copy, and the fourth argument is the alignment
2480of the source and destination locations.
2481</p>
2482
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]2483<p>
2484If the call to this intrinisic has an alignment value that is not 0 or 1, then
2485the caller guarantees that the size of the copy is a multiple of the alignment
2486and that both the source and destination pointers are aligned to that boundary.
2487</p>
2488
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]2489<h5>Semantics:</h5>
2490
2491<p>
2492The '<tt>llvm.memmove</tt>' intrinsic copies a block of memory from the source
2493location to the destination location, which may overlap. It
2494copies "len" bytes of memory over. If the argument is known to be aligned to
2495some boundary, this can be specified as the fourth argument, otherwise it should
2496be set to 0 or 1.
2497</p>
2498</div>
2499
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2500
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]2501<!-- _______________________________________________________________________ -->
2502<div class="doc_subsubsection">
2503 <a name="i_memset">'<tt>llvm.memset</tt>' Intrinsic</a>
2504</div>
2505
2506<div class="doc_text">
2507
2508<h5>Syntax:</h5>
2509<pre>
2510 call void (sbyte*, ubyte, uint, uint)* %llvm.memset(sbyte* &lt;dest&gt;, ubyte &lt;val&gt;,
2511 uint &lt;len&gt;, uint &lt;align&gt;)
2512</pre>
2513
2514<h5>Overview:</h5>
2515
2516<p>
2517The '<tt>llvm.memset</tt>' intrinsic fills a block of memory with a particular
2518byte value.
2519</p>
2520
2521<p>
2522Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
2523does not return a value, and takes an extra alignment argument.
2524</p>
2525
2526<h5>Arguments:</h5>
2527
2528<p>
2529The first argument is a pointer to the destination to fill, the second is the
2530byte value to fill it with, the third argument is an (arbitrarily sized) integer
2531argument specifying the number of bytes to fill, and the fourth argument is the
2532known alignment of destination location.
2533</p>
2534
2535<p>
2536If the call to this intrinisic has an alignment value that is not 0 or 1, then
2537the caller guarantees that the size of the copy is a multiple of the alignment
2538and that the destination pointer is aligned to that boundary.
2539</p>
2540
2541<h5>Semantics:</h5>
2542
2543<p>
2544The '<tt>llvm.memset</tt>' intrinsic fills "len" bytes of memory starting at the
2545destination location. If the argument is known to be aligned to some boundary,
2546this can be specified as the fourth argument, otherwise it should be set to 0 or
25471.
2548</p>
2549</div>
2550
2551
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]2552<!-- _______________________________________________________________________ -->
2553<div class="doc_subsubsection">
Alkis Evlogimenos26bbe932004-06-13 01:16:15 +0000[diff] [blame]2554 <a name="i_isunordered">'<tt>llvm.isunordered</tt>' Intrinsic</a>
2555</div>
2556
2557<div class="doc_text">
2558
2559<h5>Syntax:</h5>
2560<pre>
2561 call bool (&lt;float or double&gt;, &lt;float or double&gt;)* %llvm.isunordered(&lt;float or double&gt; Val1,
2562 &lt;float or double&gt; Val2)
2563</pre>
2564
2565<h5>Overview:</h5>
2566
2567<p>
2568The '<tt>llvm.isunordered</tt>' intrinsic returns true if either or both of the
2569specified floating point values is a NAN.
2570</p>
2571
2572<h5>Arguments:</h5>
2573
2574<p>
2575The arguments are floating point numbers of the same type.
2576</p>
2577
2578<h5>Semantics:</h5>
2579
2580<p>
2581If either or both of the arguments is a SNAN or QNAN, it returns true, otherwise
2582false.
2583</p>
2584</div>
2585
2586
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]2587
2588
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]2589<!-- ======================================================================= -->
2590<div class="doc_subsection">
2591 <a name="int_debugger">Debugger Intrinsics</a>
2592</div>
2593
2594<div class="doc_text">
2595<p>
2596The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
2597are described in the <a
2598href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
2599Debugging</a> document.
2600</p>
2601</div>
2602
2603
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2604<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2605<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2606<address>
2607 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
2608 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
2609 <a href="http://validator.w3.org/check/referer"><img
2610 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
2611
2612 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
2613 <a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br>
2614 Last modified: $Date$
2615</address>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2616</body>
2617</html>