Gitiles
Code Review Sign In
gerrit-public.fairphone.software / toolchain / llvm-project / 8bc7d957cf0b29fa8efedd65f5963f9b41ef2348 / . / llvm / docs / LangRef.html
blob: 45352cc4eb56a18182103d11b094fe9bb47b9d38 [file] [log] [blame]
Misha Brukmanc501f552004-03-01 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 Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3<html>
4<head>
5 <title>LLVM Assembly Language Reference Manual</title>
Reid Spencercb84e432004-08-26 20:44:00 +0000[diff] [blame]6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
7 <meta name="author" content="Chris Lattner">
8 <meta name="description"
9 content="LLVM Assembly Language Reference Manual.">
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]10 <link rel="stylesheet" href="llvm.css" type="text/css">
11</head>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]12
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]13<body>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]14
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]15<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]16<ol>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]17 <li><a href="#abstract">Abstract</a></li>
18 <li><a href="#introduction">Introduction</a></li>
19 <li><a href="#identifiers">Identifiers</a></li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]20 <li><a href="#highlevel">High Level Structure</a>
21 <ol>
22 <li><a href="#modulestructure">Module Structure</a></li>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]23 <li><a href="#linkage">Linkage Types</a></li>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]24 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]25 <li><a href="#globalvars">Global Variables</a></li>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]26 <li><a href="#functionstructure">Functions</a></li>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]27 <li><a href="#paramattrs">Parameter Attributes</a></li>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]28 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]29 <li><a href="#datalayout">Data Layout</a></li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]30 </ol>
31 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]32 <li><a href="#typesystem">Type System</a>
33 <ol>
Robert Bocchino820bc75b2006-02-17 21:18:08 +0000[diff] [blame]34 <li><a href="#t_primitive">Primitive Types</a>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]35 <ol>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]36 <li><a href="#t_classifications">Type Classifications</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]37 </ol>
38 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]39 <li><a href="#t_derived">Derived Types</a>
40 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]41 <li><a href="#t_array">Array Type</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]42 <li><a href="#t_function">Function Type</a></li>
43 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]44 <li><a href="#t_struct">Structure Type</a></li>
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]45 <li><a href="#t_pstruct">Packed Structure Type</a></li>
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]46 <li><a href="#t_vector">Vector Type</a></li>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]47 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]48 </ol>
49 </li>
50 </ol>
51 </li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]52 <li><a href="#constants">Constants</a>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]53 <ol>
54 <li><a href="#simpleconstants">Simple Constants</a>
55 <li><a href="#aggregateconstants">Aggregate Constants</a>
56 <li><a href="#globalconstants">Global Variable and Function Addresses</a>
57 <li><a href="#undefvalues">Undefined Values</a>
58 <li><a href="#constantexprs">Constant Expressions</a>
59 </ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]60 </li>
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]61 <li><a href="#othervalues">Other Values</a>
62 <ol>
63 <li><a href="#inlineasm">Inline Assembler Expressions</a>
64 </ol>
65 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]66 <li><a href="#instref">Instruction Reference</a>
67 <ol>
68 <li><a href="#terminators">Terminator Instructions</a>
69 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]70 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
71 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]72 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
73 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]74 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]75 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]76 </ol>
77 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]78 <li><a href="#binaryops">Binary Operations</a>
79 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]80 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
81 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
82 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]83 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
84 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
85 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]86 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
87 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
88 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]89 </ol>
90 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]91 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
92 <ol>
Reid Spencer2ab01932007-02-02 13:57:07 +0000[diff] [blame]93 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
94 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
95 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]96 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]97 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]98 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]99 </ol>
100 </li>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]101 <li><a href="#vectorops">Vector Operations</a>
102 <ol>
103 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
104 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
105 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]106 </ol>
107 </li>
Chris Lattner6ab66722006-08-15 00:45:58 +0000[diff] [blame]108 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]109 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]110 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
111 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
112 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino820bc75b2006-02-17 21:18:08 +0000[diff] [blame]113 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
114 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
115 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]116 </ol>
117 </li>
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]118 <li><a href="#convertops">Conversion Operations</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]119 <ol>
120 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
121 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
122 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
123 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
124 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]125 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
126 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
127 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
128 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]129 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
130 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]131 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]132 </ol>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]133 <li><a href="#otherops">Other Operations</a>
134 <ol>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]135 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
136 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]137 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]138 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]139 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattner33337472006-01-13 23:26:01 +0000[diff] [blame]140 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]141 </ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]142 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]143 </ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]144 </li>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]145 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]146 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]147 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
148 <ol>
149 <li><a href="#i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
150 <li><a href="#i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
151 <li><a href="#i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
152 </ol>
153 </li>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]154 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
155 <ol>
156 <li><a href="#i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
157 <li><a href="#i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
158 <li><a href="#i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
159 </ol>
160 </li>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]161 <li><a href="#int_codegen">Code Generator Intrinsics</a>
162 <ol>
163 <li><a href="#i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
164 <li><a href="#i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]165 <li><a href="#i_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
166 <li><a href="#i_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]167 <li><a href="#i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]168 <li><a href="#i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
Andrew Lenharth01aa5632005-11-11 16:47:30 +0000[diff] [blame]169 <li><a href="#i_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswellaa1c3c12004-04-09 16:43:20 +0000[diff] [blame]170 </ol>
171 </li>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]172 <li><a href="#int_libc">Standard C Library Intrinsics</a>
173 <ol>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]174 <li><a href="#i_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
175 <li><a href="#i_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
176 <li><a href="#i_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
Chris Lattner069b5bd2006-01-16 22:38:59 +0000[diff] [blame]177 <li><a href="#i_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]178 <li><a href="#i_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]179 </ol>
180 </li>
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]181 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]182 <ol>
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]183 <li><a href="#i_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattnerb748c672006-01-16 22:34:14 +0000[diff] [blame]184 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
185 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
186 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]187 </ol>
188 </li>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]189 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskey2211f492007-03-14 19:31:19 +0000[diff] [blame]190 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]191 </ol>
192 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]193</ol>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]194
195<div class="doc_author">
196 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
197 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]198</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]199
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]200<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]201<div class="doc_section"> <a name="abstract">Abstract </a></div>
202<!-- *********************************************************************** -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]203
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]204<div class="doc_text">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]205<p>This document is a reference manual for the LLVM assembly language.
206LLVM is an SSA based representation that provides type safety,
207low-level operations, flexibility, and the capability of representing
208'all' high-level languages cleanly. It is the common code
209representation used throughout all phases of the LLVM compilation
210strategy.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]211</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]212
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]213<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]214<div class="doc_section"> <a name="introduction">Introduction</a> </div>
215<!-- *********************************************************************** -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]216
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]217<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]218
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]219<p>The LLVM code representation is designed to be used in three
220different forms: as an in-memory compiler IR, as an on-disk bytecode
221representation (suitable for fast loading by a Just-In-Time compiler),
222and as a human readable assembly language representation. This allows
223LLVM to provide a powerful intermediate representation for efficient
224compiler transformations and analysis, while providing a natural means
225to debug and visualize the transformations. The three different forms
226of LLVM are all equivalent. This document describes the human readable
227representation and notation.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]228
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]229<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]230while being expressive, typed, and extensible at the same time. It
231aims to be a "universal IR" of sorts, by being at a low enough level
232that high-level ideas may be cleanly mapped to it (similar to how
233microprocessors are "universal IR's", allowing many source languages to
234be mapped to them). By providing type information, LLVM can be used as
235the target of optimizations: for example, through pointer analysis, it
236can be proven that a C automatic variable is never accessed outside of
237the current function... allowing it to be promoted to a simple SSA
238value instead of a memory location.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]239
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]240</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]241
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]242<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]243<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]244
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]245<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]246
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]247<p>It is important to note that this document describes 'well formed'
248LLVM assembly language. There is a difference between what the parser
249accepts and what is considered 'well formed'. For example, the
250following instruction is syntactically okay, but not well formed:</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]251
252<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]253 %x = <a href="#i_add">add</a> i32 1, %x
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]254</pre>
255
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]256<p>...because the definition of <tt>%x</tt> does not dominate all of
257its uses. The LLVM infrastructure provides a verification pass that may
258be used to verify that an LLVM module is well formed. This pass is
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]259automatically run by the parser after parsing input assembly and by
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]260the optimizer before it outputs bytecode. The violations pointed out
261by the verifier pass indicate bugs in transformation passes or input to
262the parser.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]263
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]264<!-- Describe the typesetting conventions here. --> </div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]265
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]266<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]267<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]268<!-- *********************************************************************** -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]269
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]270<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]271
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]272<p>LLVM uses three different forms of identifiers, for different
273purposes:</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]274
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]275<ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]276 <li>Named values are represented as a string of characters with a '%' prefix.
277 For example, %foo, %DivisionByZero, %a.really.long.identifier. The actual
278 regular expression used is '<tt>%[a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
279 Identifiers which require other characters in their names can be surrounded
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]280 with quotes. In this way, anything except a <tt>&quot;</tt> character can be used
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]281 in a name.</li>
282
283 <li>Unnamed values are represented as an unsigned numeric value with a '%'
284 prefix. For example, %12, %2, %44.</li>
285
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]286 <li>Constants, which are described in a <a href="#constants">section about
287 constants</a>, below.</li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]288</ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]289
290<p>LLVM requires that values start with a '%' sign for two reasons: Compilers
291don't need to worry about name clashes with reserved words, and the set of
292reserved words may be expanded in the future without penalty. Additionally,
293unnamed identifiers allow a compiler to quickly come up with a temporary
294variable without having to avoid symbol table conflicts.</p>
295
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]296<p>Reserved words in LLVM are very similar to reserved words in other
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]297languages. There are keywords for different opcodes
298('<tt><a href="#i_add">add</a></tt>',
299 '<tt><a href="#i_bitcast">bitcast</a></tt>',
300 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]301href="#t_void">void</a></tt>', '<tt><a href="#t_primitive">i32</a></tt>', etc...),
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]302and others. These reserved words cannot conflict with variable names, because
303none of them start with a '%' character.</p>
304
305<p>Here is an example of LLVM code to multiply the integer variable
306'<tt>%X</tt>' by 8:</p>
307
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]308<p>The easy way:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]309
310<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]311 %result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]312</pre>
313
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]314<p>After strength reduction:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]315
316<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]317 %result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]318</pre>
319
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]320<p>And the hard way:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]321
322<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]323 <a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
324 <a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
325 %result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]326</pre>
327
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]328<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
329important lexical features of LLVM:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]330
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]331<ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]332
333 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
334 line.</li>
335
336 <li>Unnamed temporaries are created when the result of a computation is not
337 assigned to a named value.</li>
338
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]339 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]340
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]341</ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]342
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]343<p>...and it also shows a convention that we follow in this document. When
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]344demonstrating instructions, we will follow an instruction with a comment that
345defines the type and name of value produced. Comments are shown in italic
346text.</p>
347
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]348</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]349
350<!-- *********************************************************************** -->
351<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
352<!-- *********************************************************************** -->
353
354<!-- ======================================================================= -->
355<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
356</div>
357
358<div class="doc_text">
359
360<p>LLVM programs are composed of "Module"s, each of which is a
361translation unit of the input programs. Each module consists of
362functions, global variables, and symbol table entries. Modules may be
363combined together with the LLVM linker, which merges function (and
364global variable) definitions, resolves forward declarations, and merges
365symbol table entries. Here is an example of the "hello world" module:</p>
366
367<pre><i>; Declare the string constant as a global constant...</i>
368<a href="#identifiers">%.LC0</a> = <a href="#linkage_internal">internal</a> <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]369 href="#globalvars">constant</a> <a href="#t_array">[13 x i8 ]</a> c"hello world\0A\00" <i>; [13 x i8 ]*</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]370
371<i>; External declaration of the puts function</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]372<a href="#functionstructure">declare</a> i32 %puts(i8 *) <i>; i32(i8 *)* </i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]373
374<i>; Definition of main function</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]375define i32 %main() { <i>; i32()* </i>
376 <i>; Convert [13x i8 ]* to i8 *...</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]377 %cast210 = <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]378 href="#i_getelementptr">getelementptr</a> [13 x i8 ]* %.LC0, i64 0, i64 0 <i>; i8 *</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]379
380 <i>; Call puts function to write out the string to stdout...</i>
381 <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]382 href="#i_call">call</a> i32 %puts(i8 * %cast210) <i>; i32</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]383 <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]384 href="#i_ret">ret</a> i32 0<br>}<br></pre>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]385
386<p>This example is made up of a <a href="#globalvars">global variable</a>
387named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
388function, and a <a href="#functionstructure">function definition</a>
389for "<tt>main</tt>".</p>
390
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]391<p>In general, a module is made up of a list of global values,
392where both functions and global variables are global values. Global values are
393represented by a pointer to a memory location (in this case, a pointer to an
394array of char, and a pointer to a function), and have one of the following <a
395href="#linkage">linkage types</a>.</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]396
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]397</div>
398
399<!-- ======================================================================= -->
400<div class="doc_subsection">
401 <a name="linkage">Linkage Types</a>
402</div>
403
404<div class="doc_text">
405
406<p>
407All Global Variables and Functions have one of the following types of linkage:
408</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]409
410<dl>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]411
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]412 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]413
414 <dd>Global values with internal linkage are only directly accessible by
415 objects in the current module. In particular, linking code into a module with
416 an internal global value may cause the internal to be renamed as necessary to
417 avoid collisions. Because the symbol is internal to the module, all
418 references can be updated. This corresponds to the notion of the
Chris Lattnere20b4702007-01-14 06:51:48 +0000[diff] [blame]419 '<tt>static</tt>' keyword in C.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]420 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]421
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]422 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]423
Chris Lattnere20b4702007-01-14 06:51:48 +0000[diff] [blame]424 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
425 the same name when linkage occurs. This is typically used to implement
426 inline functions, templates, or other code which must be generated in each
427 translation unit that uses it. Unreferenced <tt>linkonce</tt> globals are
428 allowed to be discarded.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]429 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]430
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]431 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]432
433 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
434 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
Chris Lattnere20b4702007-01-14 06:51:48 +0000[diff] [blame]435 used for globals that may be emitted in multiple translation units, but that
436 are not guaranteed to be emitted into every translation unit that uses them.
437 One example of this are common globals in C, such as "<tt>int X;</tt>" at
438 global scope.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]439 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]440
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]441 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]442
443 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
444 pointer to array type. When two global variables with appending linkage are
445 linked together, the two global arrays are appended together. This is the
446 LLVM, typesafe, equivalent of having the system linker append together
447 "sections" with identical names when .o files are linked.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]448 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]449
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]450 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt>: </dt>
451 <dd>The semantics of this linkage follow the ELF model: the symbol is weak
452 until linked, if not linked, the symbol becomes null instead of being an
453 undefined reference.
454 </dd>
455</dl>
456
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]457 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]458
459 <dd>If none of the above identifiers are used, the global is externally
460 visible, meaning that it participates in linkage and can be used to resolve
461 external symbol references.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]462 </dd>
Anton Korobeynikovd61d39e2006-09-14 18:23:27 +0000[diff] [blame]463
Anton Korobeynikovd61d39e2006-09-14 18:23:27 +0000[diff] [blame]464 <p>
465 The next two types of linkage are targeted for Microsoft Windows platform
466 only. They are designed to support importing (exporting) symbols from (to)
467 DLLs.
468 </p>
469
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]470 <dl>
Anton Korobeynikovd61d39e2006-09-14 18:23:27 +0000[diff] [blame]471 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt>: </dt>
472
473 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
474 or variable via a global pointer to a pointer that is set up by the DLL
475 exporting the symbol. On Microsoft Windows targets, the pointer name is
476 formed by combining <code>_imp__</code> and the function or variable name.
477 </dd>
478
479 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt>: </dt>
480
481 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
482 pointer to a pointer in a DLL, so that it can be referenced with the
483 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
484 name is formed by combining <code>_imp__</code> and the function or variable
485 name.
486 </dd>
487
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]488</dl>
489
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]490<p><a name="linkage_external"></a>For example, since the "<tt>.LC0</tt>"
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]491variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
492variable and was linked with this one, one of the two would be renamed,
493preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
494external (i.e., lacking any linkage declarations), they are accessible
Reid Spencer92c671e2007-01-05 00:59:10 +0000[diff] [blame]495outside of the current module.</p>
496<p>It is illegal for a function <i>declaration</i>
497to have any linkage type other than "externally visible", <tt>dllimport</tt>,
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]498or <tt>extern_weak</tt>.</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]499
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]500</div>
501
502<!-- ======================================================================= -->
503<div class="doc_subsection">
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]504 <a name="callingconv">Calling Conventions</a>
505</div>
506
507<div class="doc_text">
508
509<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
510and <a href="#i_invoke">invokes</a> can all have an optional calling convention
511specified for the call. The calling convention of any pair of dynamic
512caller/callee must match, or the behavior of the program is undefined. The
513following calling conventions are supported by LLVM, and more may be added in
514the future:</p>
515
516<dl>
517 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
518
519 <dd>This calling convention (the default if no other calling convention is
520 specified) matches the target C calling conventions. This calling convention
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]521 supports varargs function calls and tolerates some mismatch in the declared
Reid Spencer72ba4992006-12-31 21:30:18 +0000[diff] [blame]522 prototype and implemented declaration of the function (as does normal C).
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]523 </dd>
524
525 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
526
527 <dd>This calling convention attempts to make calls as fast as possible
528 (e.g. by passing things in registers). This calling convention allows the
529 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattnerc792eb32005-05-06 23:08:23 +0000[diff] [blame]530 without having to conform to an externally specified ABI. Implementations of
531 this convention should allow arbitrary tail call optimization to be supported.
532 This calling convention does not support varargs and requires the prototype of
533 all callees to exactly match the prototype of the function definition.
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]534 </dd>
535
536 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
537
538 <dd>This calling convention attempts to make code in the caller as efficient
539 as possible under the assumption that the call is not commonly executed. As
540 such, these calls often preserve all registers so that the call does not break
541 any live ranges in the caller side. This calling convention does not support
542 varargs and requires the prototype of all callees to exactly match the
543 prototype of the function definition.
544 </dd>
545
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]546 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]547
548 <dd>Any calling convention may be specified by number, allowing
549 target-specific calling conventions to be used. Target specific calling
550 conventions start at 64.
551 </dd>
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]552</dl>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]553
554<p>More calling conventions can be added/defined on an as-needed basis, to
555support pascal conventions or any other well-known target-independent
556convention.</p>
557
558</div>
559
560<!-- ======================================================================= -->
561<div class="doc_subsection">
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]562 <a name="visibility">Visibility Styles</a>
563</div>
564
565<div class="doc_text">
566
567<p>
568All Global Variables and Functions have one of the following visibility styles:
569</p>
570
571<dl>
572 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
573
574 <dd>On ELF, default visibility means that the declaration is visible to other
575 modules and, in shared libraries, means that the declared entity may be
576 overridden. On Darwin, default visibility means that the declaration is
577 visible to other modules. Default visibility corresponds to "external
578 linkage" in the language.
579 </dd>
580
581 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
582
583 <dd>Two declarations of an object with hidden visibility refer to the same
584 object if they are in the same shared object. Usually, hidden visibility
585 indicates that the symbol will not be placed into the dynamic symbol table,
586 so no other module (executable or shared library) can reference it
587 directly.
588 </dd>
589
590</dl>
591
592</div>
593
594<!-- ======================================================================= -->
595<div class="doc_subsection">
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]596 <a name="globalvars">Global Variables</a>
597</div>
598
599<div class="doc_text">
600
Chris Lattner5d5aede2005-02-12 19:30:21 +0000[diff] [blame]601<p>Global variables define regions of memory allocated at compilation time
Chris Lattner662c8722005-11-12 00:45:07 +0000[diff] [blame]602instead of run-time. Global variables may optionally be initialized, may have
603an explicit section to be placed in, and may
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]604have an optional explicit alignment specified. A
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]605variable may be defined as a global "constant," which indicates that the
Chris Lattner5d5aede2005-02-12 19:30:21 +0000[diff] [blame]606contents of the variable will <b>never</b> be modified (enabling better
607optimization, allowing the global data to be placed in the read-only section of
608an executable, etc). Note that variables that need runtime initialization
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]609cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner5d5aede2005-02-12 19:30:21 +0000[diff] [blame]610
611<p>
612LLVM explicitly allows <em>declarations</em> of global variables to be marked
613constant, even if the final definition of the global is not. This capability
614can be used to enable slightly better optimization of the program, but requires
615the language definition to guarantee that optimizations based on the
616'constantness' are valid for the translation units that do not include the
617definition.
618</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]619
620<p>As SSA values, global variables define pointer values that are in
621scope (i.e. they dominate) all basic blocks in the program. Global
622variables always define a pointer to their "content" type because they
623describe a region of memory, and all memory objects in LLVM are
624accessed through pointers.</p>
625
Chris Lattner662c8722005-11-12 00:45:07 +0000[diff] [blame]626<p>LLVM allows an explicit section to be specified for globals. If the target
627supports it, it will emit globals to the section specified.</p>
628
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]629<p>An explicit alignment may be specified for a global. If not present, or if
630the alignment is set to zero, the alignment of the global is set by the target
631to whatever it feels convenient. If an explicit alignment is specified, the
632global is forced to have at least that much alignment. All alignments must be
633a power of 2.</p>
634
Chris Lattner5760c502007-01-14 00:27:09 +0000[diff] [blame]635<p>For example, the following defines a global with an initializer, section,
636 and alignment:</p>
637
638<pre>
639 %G = constant float 1.0, section "foo", align 4
640</pre>
641
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]642</div>
643
644
645<!-- ======================================================================= -->
646<div class="doc_subsection">
647 <a name="functionstructure">Functions</a>
648</div>
649
650<div class="doc_text">
651
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]652<p>LLVM function definitions consist of the "<tt>define</tt>" keyord,
653an optional <a href="#linkage">linkage type</a>, an optional
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]654<a href="#visibility">visibility style</a>, an optional
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]655<a href="#callingconv">calling convention</a>, a return type, an optional
656<a href="#paramattrs">parameter attribute</a> for the return type, a function
657name, a (possibly empty) argument list (each with optional
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]658<a href="#paramattrs">parameter attributes</a>), an optional section, an
659optional alignment, an opening curly brace, a list of basic blocks, and a
660closing curly brace.
661
662LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
663optional <a href="#linkage">linkage type</a>, an optional
664<a href="#visibility">visibility style</a>, an optional
665<a href="#callingconv">calling convention</a>, a return type, an optional
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]666<a href="#paramattrs">parameter attribute</a> for the return type, a function
667name, a possibly empty list of arguments, and an optional alignment.</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]668
669<p>A function definition contains a list of basic blocks, forming the CFG for
670the function. Each basic block may optionally start with a label (giving the
671basic block a symbol table entry), contains a list of instructions, and ends
672with a <a href="#terminators">terminator</a> instruction (such as a branch or
673function return).</p>
674
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]675<p>The first basic block in a program is special in two ways: it is immediately
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]676executed on entrance to the function, and it is not allowed to have predecessor
677basic blocks (i.e. there can not be any branches to the entry block of a
678function). Because the block can have no predecessors, it also cannot have any
679<a href="#i_phi">PHI nodes</a>.</p>
680
681<p>LLVM functions are identified by their name and type signature. Hence, two
682functions with the same name but different parameter lists or return values are
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]683considered different functions, and LLVM will resolve references to each
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]684appropriately.</p>
685
Chris Lattner662c8722005-11-12 00:45:07 +0000[diff] [blame]686<p>LLVM allows an explicit section to be specified for functions. If the target
687supports it, it will emit functions to the section specified.</p>
688
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]689<p>An explicit alignment may be specified for a function. If not present, or if
690the alignment is set to zero, the alignment of the function is set by the target
691to whatever it feels convenient. If an explicit alignment is specified, the
692function is forced to have at least that much alignment. All alignments must be
693a power of 2.</p>
694
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]695</div>
696
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]697<!-- ======================================================================= -->
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]698<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
699<div class="doc_text">
700 <p>The return type and each parameter of a function type may have a set of
701 <i>parameter attributes</i> associated with them. Parameter attributes are
702 used to communicate additional information about the result or parameters of
703 a function. Parameter attributes are considered to be part of the function
704 type so two functions types that differ only by the parameter attributes
705 are different function types.</p>
706
Reid Spencercf7ebf52007-01-15 18:27:39 +0000[diff] [blame]707 <p>Parameter attributes are simple keywords that follow the type specified. If
708 multiple parameter attributes are needed, they are space separated. For
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]709 example:</p><pre>
Reid Spencercf7ebf52007-01-15 18:27:39 +0000[diff] [blame]710 %someFunc = i16 (i8 sext %someParam) zext
711 %someFunc = i16 (i8 zext %someParam) zext</pre>
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]712 <p>Note that the two function types above are unique because the parameter has
Reid Spencercf7ebf52007-01-15 18:27:39 +0000[diff] [blame]713 a different attribute (sext in the first one, zext in the second). Also note
714 that the attribute for the function result (zext) comes immediately after the
715 argument list.</p>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]716
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]717 <p>Currently, only the following parameter attributes are defined:</p>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]718 <dl>
Reid Spencercf7ebf52007-01-15 18:27:39 +0000[diff] [blame]719 <dt><tt>zext</tt></dt>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]720 <dd>This indicates that the parameter should be zero extended just before
721 a call to this function.</dd>
Reid Spencercf7ebf52007-01-15 18:27:39 +0000[diff] [blame]722 <dt><tt>sext</tt></dt>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]723 <dd>This indicates that the parameter should be sign extended just before
724 a call to this function.</dd>
Anton Korobeynikove8166852007-01-28 14:30:45 +0000[diff] [blame]725 <dt><tt>inreg</tt></dt>
726 <dd>This indicates that the parameter should be placed in register (if
Anton Korobeynikove93c6e82007-01-28 15:27:21 +0000[diff] [blame]727 possible) during assembling function call. Support for this attribute is
728 target-specific</dd>
Anton Korobeynikove8166852007-01-28 14:30:45 +0000[diff] [blame]729 <dt><tt>sret</tt></dt>
Anton Korobeynikove93c6e82007-01-28 15:27:21 +0000[diff] [blame]730 <dd>This indicates that the parameter specifies the address of a structure
Reid Spencer05dbb9d2007-03-22 02:02:11 +0000[diff] [blame]731 that is the return value of the function in the source program.</dd>
Reid Spencer9d1700e2007-03-22 02:18:56 +0000[diff] [blame]732 <dt><tt>noreturn</tt></dt>
733 <dd>This function attribute indicates that the function never returns. This
734 indicates to LLVM that every call to this function should be treated as if
735 an <tt>unreachable</tt> instruction immediately followed the call.</dd>
Reid Spencer05dbb9d2007-03-22 02:02:11 +0000[diff] [blame]736 <dt><tt>nounwind</tt></dt>
737 <dd>This function attribute indicates that the function type does not use
738 the unwind instruction and does not allow stack unwinding to propagate
739 through it.</dd>
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]740 </dl>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]741
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]742</div>
743
744<!-- ======================================================================= -->
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]745<div class="doc_subsection">
Chris Lattner93564892006-04-08 04:40:53 +0000[diff] [blame]746 <a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]747</div>
748
749<div class="doc_text">
750<p>
751Modules may contain "module-level inline asm" blocks, which corresponds to the
752GCC "file scope inline asm" blocks. These blocks are internally concatenated by
753LLVM and treated as a single unit, but may be separated in the .ll file if
754desired. The syntax is very simple:
755</p>
756
757<div class="doc_code"><pre>
Chris Lattnera1280ad2006-01-24 00:37:20 +0000[diff] [blame]758 module asm "inline asm code goes here"
759 module asm "more can go here"
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]760</pre></div>
761
762<p>The strings can contain any character by escaping non-printable characters.
763 The escape sequence used is simply "\xx" where "xx" is the two digit hex code
764 for the number.
765</p>
766
767<p>
768 The inline asm code is simply printed to the machine code .s file when
769 assembly code is generated.
770</p>
771</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]772
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]773<!-- ======================================================================= -->
774<div class="doc_subsection">
775 <a name="datalayout">Data Layout</a>
776</div>
777
778<div class="doc_text">
779<p>A module may specify a target specific data layout string that specifies how
780data is to be laid out in memory. The syntax for the data layout is simply:<br/>
781<pre> target datalayout = "<i>layout specification</i>"
782</pre>
783The <i>layout specification</i> consists of a list of specifications separated
784by the minus sign character ('-'). Each specification starts with a letter
785and may include other information after the letter to define some aspect of the
786data layout. The specifications accepted are as follows: </p>
787<dl>
788 <dt><tt>E</tt></dt>
789 <dd>Specifies that the target lays out data in big-endian form. That is, the
790 bits with the most significance have the lowest address location.</dd>
791 <dt><tt>e</tt></dt>
792 <dd>Specifies that hte target lays out data in little-endian form. That is,
793 the bits with the least significance have the lowest address location.</dd>
794 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
795 <dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
796 <i>preferred</i> alignments. All sizes are in bits. Specifying the <i>pref</i>
797 alignment is optional. If omitted, the preceding <tt>:</tt> should be omitted
798 too.</dd>
799 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
800 <dd>This specifies the alignment for an integer type of a given bit
801 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
802 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
803 <dd>This specifies the alignment for a vector type of a given bit
804 <i>size</i>.</dd>
805 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
806 <dd>This specifies the alignment for a floating point type of a given bit
807 <i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
808 (double).</dd>
809 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
810 <dd>This specifies the alignment for an aggregate type of a given bit
811 <i>size</i>.</dd>
812</dl>
813<p>When constructing the data layout for a given target, LLVM starts with a
814default set of specifications which are then (possibly) overriden by the
815specifications in the <tt>datalayout</tt> keyword. The default specifications
816are given in this list:</p>
817<ul>
818 <li><tt>E</tt> - big endian</li>
819 <li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
820 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
821 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
822 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
823 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
824 <li><tt>i64:32:64</tt> - i64 has abi alignment of 32-bits but preferred
825 alignment of 64-bits</li>
826 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
827 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
828 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
829 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
830 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
831</ul>
832<p>When llvm is determining the alignment for a given type, it uses the
833following rules:
834<ol>
835 <li>If the type sought is an exact match for one of the specifications, that
836 specification is used.</li>
837 <li>If no match is found, and the type sought is an integer type, then the
838 smallest integer type that is larger than the bitwidth of the sought type is
839 used. If none of the specifications are larger than the bitwidth then the the
840 largest integer type is used. For example, given the default specifications
841 above, the i7 type will use the alignment of i8 (next largest) while both
842 i65 and i256 will use the alignment of i64 (largest specified).</li>
843 <li>If no match is found, and the type sought is a vector type, then the
844 largest vector type that is smaller than the sought vector type will be used
845 as a fall back. This happens because <128 x double> can be implemented in
846 terms of 64 <2 x double>, for example.</li>
847</ol>
848</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]849
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]850<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]851<div class="doc_section"> <a name="typesystem">Type System</a> </div>
852<!-- *********************************************************************** -->
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]853
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]854<div class="doc_text">
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]855
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]856<p>The LLVM type system is one of the most important features of the
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]857intermediate representation. Being typed enables a number of
858optimizations to be performed on the IR directly, without having to do
859extra analyses on the side before the transformation. A strong type
860system makes it easier to read the generated code and enables novel
861analyses and transformations that are not feasible to perform on normal
862three address code representations.</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]863
864</div>
865
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]866<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]867<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]868<div class="doc_text">
John Criswell417228d2004-04-09 16:48:45 +0000[diff] [blame]869<p>The primitive types are the fundamental building blocks of the LLVM
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]870system. The current set of primitive types is as follows:</p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]871
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]872<table class="layout">
873 <tr class="layout">
874 <td class="left">
875 <table>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]876 <tbody>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]877 <tr><th>Type</th><th>Description</th></tr>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]878 <tr><td><tt><a name="t_void">void</a></tt></td><td>No value</td></tr>
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]879 <tr><td><tt>i8</tt></td><td>8-bit value</td></tr>
880 <tr><td><tt>i32</tt></td><td>32-bit value</td></tr>
Misha Brukman36c6bc12005-04-22 18:02:52 +0000[diff] [blame]881 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]882 <tr><td><tt>label</tt></td><td>Branch destination</td></tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]883 </tbody>
884 </table>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]885 </td>
886 <td class="right">
887 <table>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]888 <tbody>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]889 <tr><th>Type</th><th>Description</th></tr>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]890 <tr><td><tt>i1</tt></td><td>True or False value</td></tr>
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]891 <tr><td><tt>i16</tt></td><td>16-bit value</td></tr>
892 <tr><td><tt>i64</tt></td><td>64-bit value</td></tr>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]893 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]894 </tbody>
895 </table>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]896 </td>
897 </tr>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]898</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]899</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]900
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]901<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]902<div class="doc_subsubsection"> <a name="t_classifications">Type
903Classifications</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]904<div class="doc_text">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]905<p>These different primitive types fall into a few useful
906classifications:</p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]907
908<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]909 <tbody>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]910 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]911 <tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]912 <td><a name="t_integer">integer</a></td>
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]913 <td><tt>i1, i8, i16, i32, i64</tt></td>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]914 </tr>
915 <tr>
916 <td><a name="t_floating">floating point</a></td>
917 <td><tt>float, double</tt></td>
918 </tr>
919 <tr>
920 <td><a name="t_firstclass">first class</a></td>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]921 <td><tt>i1, i8, i16, i32, i64, float, double, <br/>
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]922 <a href="#t_pointer">pointer</a>,<a href="#t_vector">vector</a></tt>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]923 </td>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]924 </tr>
925 </tbody>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]926</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]927
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]928<p>The <a href="#t_firstclass">first class</a> types are perhaps the
929most important. Values of these types are the only ones which can be
930produced by instructions, passed as arguments, or used as operands to
931instructions. This means that all structures and arrays must be
932manipulated either by pointer or by component.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]933</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]934
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]935<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]936<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]937
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]938<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]939
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]940<p>The real power in LLVM comes from the derived types in the system.
941This is what allows a programmer to represent arrays, functions,
942pointers, and other useful types. Note that these derived types may be
943recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]944
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]945</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]946
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]947<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]948<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]949
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]950<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]951
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]952<h5>Overview:</h5>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]953
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]954<p>The array type is a very simple derived type that arranges elements
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]955sequentially in memory. The array type requires a size (number of
956elements) and an underlying data type.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]957
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]958<h5>Syntax:</h5>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]959
960<pre>
961 [&lt;# elements&gt; x &lt;elementtype&gt;]
962</pre>
963
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]964<p>The number of elements is a constant integer value; elementtype may
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]965be any type with a size.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]966
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]967<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]968<table class="layout">
969 <tr class="layout">
970 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]971 <tt>[40 x i32 ]</tt><br/>
972 <tt>[41 x i32 ]</tt><br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]973 <tt>[40 x i8]</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]974 </td>
975 <td class="left">
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]976 Array of 40 32-bit integer values.<br/>
977 Array of 41 32-bit integer values.<br/>
978 Array of 40 8-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]979 </td>
980 </tr>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]981</table>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]982<p>Here are some examples of multidimensional arrays:</p>
983<table class="layout">
984 <tr class="layout">
985 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]986 <tt>[3 x [4 x i32]]</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]987 <tt>[12 x [10 x float]]</tt><br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]988 <tt>[2 x [3 x [4 x i16]]]</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]989 </td>
990 <td class="left">
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]991 3x4 array of 32-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]992 12x10 array of single precision floating point values.<br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]993 2x3x4 array of 16-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]994 </td>
995 </tr>
996</table>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]997
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]998<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
999length array. Normally, accesses past the end of an array are undefined in
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]1000LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
1001As a special case, however, zero length arrays are recognized to be variable
1002length. This allows implementation of 'pascal style arrays' with the LLVM
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1003type "{ i32, [0 x float]}", for example.</p>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]1004
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1005</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1006
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1007<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1008<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1009<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1010<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1011<p>The function type can be thought of as a function signature. It
1012consists of a return type and a list of formal parameter types.
John Criswella0d50d22003-11-25 21:45:46 +0000[diff] [blame]1013Function types are usually used to build virtual function tables
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1014(which are structures of pointers to functions), for indirect function
1015calls, and when defining a function.</p>
John Criswella0d50d22003-11-25 21:45:46 +0000[diff] [blame]1016<p>
1017The return type of a function type cannot be an aggregate type.
1018</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1019<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1020<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]1021<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
Misha Brukman20f9a622004-08-12 20:16:08 +0000[diff] [blame]1022specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1023which indicates that the function takes a variable number of arguments.
1024Variable argument functions can access their arguments with the <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1025 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1026<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1027<table class="layout">
1028 <tr class="layout">
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1029 <td class="left"><tt>i32 (i32)</tt></td>
1030 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1031 </td>
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1032 </tr><tr class="layout">
Reid Spencere6a338d2007-01-15 18:28:34 +0000[diff] [blame]1033 <td class="left"><tt>float&nbsp;(i16&nbsp;sext,&nbsp;i32&nbsp;*)&nbsp;*
Reid Spencer655dcc62006-12-31 07:20:23 +0000[diff] [blame]1034 </tt></td>
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1035 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
1036 an <tt>i16</tt> that should be sign extended and a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1037 <a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1038 <tt>float</tt>.
1039 </td>
1040 </tr><tr class="layout">
1041 <td class="left"><tt>i32 (i8*, ...)</tt></td>
1042 <td class="left">A vararg function that takes at least one
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1043 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1044 which returns an integer. This is the signature for <tt>printf</tt> in
1045 LLVM.
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1046 </td>
1047 </tr>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1048</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]1049
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1050</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1051<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1052<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1053<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1054<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1055<p>The structure type is used to represent a collection of data members
1056together in memory. The packing of the field types is defined to match
1057the ABI of the underlying processor. The elements of a structure may
1058be any type that has a size.</p>
1059<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1060and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1061field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1062instruction.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1063<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1064<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1065<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1066<table class="layout">
1067 <tr class="layout">
1068 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1069 <tt>{ i32, i32, i32 }</tt><br/>
1070 <tt>{ float, i32 (i32) * }</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1071 </td>
1072 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1073 a triple of three <tt>i32</tt> values<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1074 A pair, where the first element is a <tt>float</tt> and the second element
1075 is a <a href="#t_pointer">pointer</a> to a <a href="#t_function">function</a>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1076 that takes an <tt>i32</tt>, returning an <tt>i32</tt>.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1077 </td>
1078 </tr>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1079</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1080</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1081
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1082<!-- _______________________________________________________________________ -->
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]1083<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1084</div>
1085<div class="doc_text">
1086<h5>Overview:</h5>
1087<p>The packed structure type is used to represent a collection of data members
1088together in memory. There is no padding between fields. Further, the alignment
1089of a packed structure is 1 byte. The elements of a packed structure may
1090be any type that has a size.</p>
1091<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1092and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1093field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1094instruction.</p>
1095<h5>Syntax:</h5>
1096<pre> &lt; { &lt;type list&gt; } &gt; <br></pre>
1097<h5>Examples:</h5>
1098<table class="layout">
1099 <tr class="layout">
1100 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1101 <tt> &lt; { i32, i32, i32 } &gt; </tt><br/>
1102 <tt> &lt; { float, i32 (i32) * } &gt; </tt><br/>
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]1103 </td>
1104 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1105 a triple of three <tt>i32</tt> values<br/>
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]1106 A pair, where the first element is a <tt>float</tt> and the second element
1107 is a <a href="#t_pointer">pointer</a> to a <a href="#t_function">function</a>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1108 that takes an <tt>i32</tt>, returning an <tt>i32</tt>.<br/>
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]1109 </td>
1110 </tr>
1111</table>
1112</div>
1113
1114<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1115<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1116<div class="doc_text">
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1117<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1118<p>As in many languages, the pointer type represents a pointer or
1119reference to another object, which must live in memory.</p>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1120<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1121<pre> &lt;type&gt; *<br></pre>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1122<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1123<table class="layout">
1124 <tr class="layout">
1125 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1126 <tt>[4x i32]*</tt><br/>
1127 <tt>i32 (i32 *) *</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1128 </td>
1129 <td class="left">
1130 A <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a> of
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1131 four <tt>i32</tt> values<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1132 A <a href="#t_pointer">pointer</a> to a <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1133 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
1134 <tt>i32</tt>.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1135 </td>
1136 </tr>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1137</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1138</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1139
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1140<!-- _______________________________________________________________________ -->
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1141<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1142<div class="doc_text">
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1143
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1144<h5>Overview:</h5>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1145
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1146<p>A vector type is a simple derived type that represents a vector
1147of elements. Vector types are used when multiple primitive data
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1148are operated in parallel using a single instruction (SIMD).
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1149A vector type requires a size (number of
Chris Lattner330ce692005-11-10 01:44:22 +0000[diff] [blame]1150elements) and an underlying primitive data type. Vectors must have a power
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1151of two length (1, 2, 4, 8, 16 ...). Vector types are
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1152considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1153
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1154<h5>Syntax:</h5>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1155
1156<pre>
1157 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1158</pre>
1159
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1160<p>The number of elements is a constant integer value; elementtype may
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1161be any integer or floating point type.</p>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1162
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1163<h5>Examples:</h5>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1164
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1165<table class="layout">
1166 <tr class="layout">
1167 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1168 <tt>&lt;4 x i32&gt;</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1169 <tt>&lt;8 x float&gt;</tt><br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1170 <tt>&lt;2 x i64&gt;</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1171 </td>
1172 <td class="left">
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1173 Vector of 4 32-bit integer values.<br/>
1174 Vector of 8 floating-point values.<br/>
1175 Vector of 2 64-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1176 </td>
1177 </tr>
1178</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1179</div>
1180
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1181<!-- _______________________________________________________________________ -->
1182<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1183<div class="doc_text">
1184
1185<h5>Overview:</h5>
1186
1187<p>Opaque types are used to represent unknown types in the system. This
1188corresponds (for example) to the C notion of a foward declared structure type.
1189In LLVM, opaque types can eventually be resolved to any type (not just a
1190structure type).</p>
1191
1192<h5>Syntax:</h5>
1193
1194<pre>
1195 opaque
1196</pre>
1197
1198<h5>Examples:</h5>
1199
1200<table class="layout">
1201 <tr class="layout">
1202 <td class="left">
1203 <tt>opaque</tt>
1204 </td>
1205 <td class="left">
1206 An opaque type.<br/>
1207 </td>
1208 </tr>
1209</table>
1210</div>
1211
1212
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1213<!-- *********************************************************************** -->
1214<div class="doc_section"> <a name="constants">Constants</a> </div>
1215<!-- *********************************************************************** -->
1216
1217<div class="doc_text">
1218
1219<p>LLVM has several different basic types of constants. This section describes
1220them all and their syntax.</p>
1221
1222</div>
1223
1224<!-- ======================================================================= -->
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]1225<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1226
1227<div class="doc_text">
1228
1229<dl>
1230 <dt><b>Boolean constants</b></dt>
1231
1232 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1233 constants of the <tt><a href="#t_primitive">i1</a></tt> type.
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1234 </dd>
1235
1236 <dt><b>Integer constants</b></dt>
1237
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]1238 <dd>Standard integers (such as '4') are constants of the <a
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1239 href="#t_integer">integer</a> type. Negative numbers may be used with
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1240 integer types.
1241 </dd>
1242
1243 <dt><b>Floating point constants</b></dt>
1244
1245 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
1246 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1247 notation (see below). Floating point constants must have a <a
1248 href="#t_floating">floating point</a> type. </dd>
1249
1250 <dt><b>Null pointer constants</b></dt>
1251
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1252 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1253 and must be of <a href="#t_pointer">pointer type</a>.</dd>
1254
1255</dl>
1256
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1257<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1258of floating point constants. For example, the form '<tt>double
12590x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
12604.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]1261(and the only time that they are generated by the disassembler) is when a
1262floating point constant must be emitted but it cannot be represented as a
1263decimal floating point number. For example, NaN's, infinities, and other
1264special values are represented in their IEEE hexadecimal format so that
1265assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1266
1267</div>
1268
1269<!-- ======================================================================= -->
1270<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
1271</div>
1272
1273<div class="doc_text">
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]1274<p>Aggregate constants arise from aggregation of simple constants
1275and smaller aggregate constants.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1276
1277<dl>
1278 <dt><b>Structure constants</b></dt>
1279
1280 <dd>Structure constants are represented with notation similar to structure
1281 type definitions (a comma separated list of elements, surrounded by braces
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1282 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* %G }</tt>",
1283 where "<tt>%G</tt>" is declared as "<tt>%G = external global i32</tt>". Structure constants
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]1284 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1285 types of elements must match those specified by the type.
1286 </dd>
1287
1288 <dt><b>Array constants</b></dt>
1289
1290 <dd>Array constants are represented with notation similar to array type
1291 definitions (a comma separated list of elements, surrounded by square brackets
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1292 (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74 ]</tt>". Array
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1293 constants must have <a href="#t_array">array type</a>, and the number and
1294 types of elements must match those specified by the type.
1295 </dd>
1296
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1297 <dt><b>Vector constants</b></dt>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1298
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1299 <dd>Vector constants are represented with notation similar to vector type
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1300 definitions (a comma separated list of elements, surrounded by
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1301 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32 42,
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1302 i32 11, i32 74, i32 100 &gt;</tt>". VEctor constants must have <a
1303 href="#t_vector">vector type</a>, and the number and types of elements must
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1304 match those specified by the type.
1305 </dd>
1306
1307 <dt><b>Zero initialization</b></dt>
1308
1309 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
1310 value to zero of <em>any</em> type, including scalar and aggregate types.
1311 This is often used to avoid having to print large zero initializers (e.g. for
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]1312 large arrays) and is always exactly equivalent to using explicit zero
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1313 initializers.
1314 </dd>
1315</dl>
1316
1317</div>
1318
1319<!-- ======================================================================= -->
1320<div class="doc_subsection">
1321 <a name="globalconstants">Global Variable and Function Addresses</a>
1322</div>
1323
1324<div class="doc_text">
1325
1326<p>The addresses of <a href="#globalvars">global variables</a> and <a
1327href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1328constants. These constants are explicitly referenced when the <a
1329href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1330href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1331file:</p>
1332
1333<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1334 %X = global i32 17
1335 %Y = global i32 42
1336 %Z = global [2 x i32*] [ i32* %X, i32* %Y ]
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1337</pre>
1338
1339</div>
1340
1341<!-- ======================================================================= -->
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1342<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1343<div class="doc_text">
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1344 <p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1345 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1346 a constant is permitted.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1347
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1348 <p>Undefined values indicate to the compiler that the program is well defined
1349 no matter what value is used, giving the compiler more freedom to optimize.
1350 </p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1351</div>
1352
1353<!-- ======================================================================= -->
1354<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1355</div>
1356
1357<div class="doc_text">
1358
1359<p>Constant expressions are used to allow expressions involving other constants
1360to be used as constants. Constant expressions may be of any <a
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1361href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1362that does not have side effects (e.g. load and call are not supported). The
1363following is the syntax for constant expressions:</p>
1364
1365<dl>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1366 <dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
1367 <dd>Truncate a constant to another type. The bit size of CST must be larger
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1368 than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1369
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1370 <dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
1371 <dd>Zero extend a constant to another type. The bit size of CST must be
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1372 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1373
1374 <dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
1375 <dd>Sign extend a constant to another type. The bit size of CST must be
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1376 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1377
1378 <dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
1379 <dd>Truncate a floating point constant to another floating point type. The
1380 size of CST must be larger than the size of TYPE. Both types must be
1381 floating point.</dd>
1382
1383 <dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
1384 <dd>Floating point extend a constant to another type. The size of CST must be
1385 smaller or equal to the size of TYPE. Both types must be floating point.</dd>
1386
1387 <dt><b><tt>fp2uint ( CST to TYPE )</tt></b></dt>
1388 <dd>Convert a floating point constant to the corresponding unsigned integer
1389 constant. TYPE must be an integer type. CST must be floating point. If the
1390 value won't fit in the integer type, the results are undefined.</dd>
1391
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]1392 <dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1393 <dd>Convert a floating point constant to the corresponding signed integer
1394 constant. TYPE must be an integer type. CST must be floating point. If the
1395 value won't fit in the integer type, the results are undefined.</dd>
1396
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]1397 <dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1398 <dd>Convert an unsigned integer constant to the corresponding floating point
1399 constant. TYPE must be floating point. CST must be of integer type. If the
1400 value won't fit in the floating point type, the results are undefined.</dd>
1401
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]1402 <dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1403 <dd>Convert a signed integer constant to the corresponding floating point
1404 constant. TYPE must be floating point. CST must be of integer type. If the
1405 value won't fit in the floating point type, the results are undefined.</dd>
1406
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]1407 <dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
1408 <dd>Convert a pointer typed constant to the corresponding integer constant
1409 TYPE must be an integer type. CST must be of pointer type. The CST value is
1410 zero extended, truncated, or unchanged to make it fit in TYPE.</dd>
1411
1412 <dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
1413 <dd>Convert a integer constant to a pointer constant. TYPE must be a
1414 pointer type. CST must be of integer type. The CST value is zero extended,
1415 truncated, or unchanged to make it fit in a pointer size. This one is
1416 <i>really</i> dangerous!</dd>
1417
1418 <dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1419 <dd>Convert a constant, CST, to another TYPE. The size of CST and TYPE must be
1420 identical (same number of bits). The conversion is done as if the CST value
1421 was stored to memory and read back as TYPE. In other words, no bits change
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]1422 with this operator, just the type. This can be used for conversion of
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1423 vector types to any other type, as long as they have the same bit width. For
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]1424 pointers it is only valid to cast to another pointer type.
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1425 </dd>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1426
1427 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1428
1429 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1430 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1431 instruction, the index list may have zero or more indexes, which are required
1432 to make sense for the type of "CSTPTR".</dd>
1433
Robert Bocchino7e97a6d2006-01-10 19:31:34 +0000[diff] [blame]1434 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
1435
1436 <dd>Perform the <a href="#i_select">select operation</a> on
Reid Spencer9965ee72006-12-04 19:23:19 +0000[diff] [blame]1437 constants.</dd>
1438
1439 <dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
1440 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
1441
1442 <dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
1443 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino7e97a6d2006-01-10 19:31:34 +0000[diff] [blame]1444
1445 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
1446
1447 <dd>Perform the <a href="#i_extractelement">extractelement
1448 operation</a> on constants.
1449
Robert Bocchinof72fdfe2006-01-15 20:48:27 +0000[diff] [blame]1450 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
1451
1452 <dd>Perform the <a href="#i_insertelement">insertelement
Reid Spencer9965ee72006-12-04 19:23:19 +0000[diff] [blame]1453 operation</a> on constants.</dd>
Robert Bocchinof72fdfe2006-01-15 20:48:27 +0000[diff] [blame]1454
Chris Lattner016a0e52006-04-08 00:13:41 +0000[diff] [blame]1455
1456 <dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
1457
1458 <dd>Perform the <a href="#i_shufflevector">shufflevector
Reid Spencer9965ee72006-12-04 19:23:19 +0000[diff] [blame]1459 operation</a> on constants.</dd>
Chris Lattner016a0e52006-04-08 00:13:41 +0000[diff] [blame]1460
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1461 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1462
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1463 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1464 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1465 binary</a> operations. The constraints on operands are the same as those for
1466 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1467 values are allowed).</dd>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1468</dl>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1469</div>
Chris Lattnerb1652612004-03-08 16:49:10 +0000[diff] [blame]1470
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1471<!-- *********************************************************************** -->
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1472<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
1473<!-- *********************************************************************** -->
1474
1475<!-- ======================================================================= -->
1476<div class="doc_subsection">
1477<a name="inlineasm">Inline Assembler Expressions</a>
1478</div>
1479
1480<div class="doc_text">
1481
1482<p>
1483LLVM supports inline assembler expressions (as opposed to <a href="#moduleasm">
1484Module-Level Inline Assembly</a>) through the use of a special value. This
1485value represents the inline assembler as a string (containing the instructions
1486to emit), a list of operand constraints (stored as a string), and a flag that
1487indicates whether or not the inline asm expression has side effects. An example
1488inline assembler expression is:
1489</p>
1490
1491<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1492 i32 (i32) asm "bswap $0", "=r,r"
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1493</pre>
1494
1495<p>
1496Inline assembler expressions may <b>only</b> be used as the callee operand of
1497a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we have:
1498</p>
1499
1500<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1501 %X = call i32 asm "<a href="#i_bswap">bswap</a> $0", "=r,r"(i32 %Y)
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1502</pre>
1503
1504<p>
1505Inline asms with side effects not visible in the constraint list must be marked
1506as having side effects. This is done through the use of the
1507'<tt>sideeffect</tt>' keyword, like so:
1508</p>
1509
1510<pre>
1511 call void asm sideeffect "eieio", ""()
1512</pre>
1513
1514<p>TODO: The format of the asm and constraints string still need to be
1515documented here. Constraints on what can be done (e.g. duplication, moving, etc
1516need to be documented).
1517</p>
1518
1519</div>
1520
1521<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1522<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1523<!-- *********************************************************************** -->
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1524
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1525<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1526
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1527<p>The LLVM instruction set consists of several different
1528classifications of instructions: <a href="#terminators">terminator
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1529instructions</a>, <a href="#binaryops">binary instructions</a>,
1530<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1531 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1532instructions</a>.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1533
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1534</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1535
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1536<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1537<div class="doc_subsection"> <a name="terminators">Terminator
1538Instructions</a> </div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1539
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1540<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1541
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1542<p>As mentioned <a href="#functionstructure">previously</a>, every
1543basic block in a program ends with a "Terminator" instruction, which
1544indicates which block should be executed after the current block is
1545finished. These terminator instructions typically yield a '<tt>void</tt>'
1546value: they produce control flow, not values (the one exception being
1547the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1548<p>There are six different terminator instructions: the '<a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1549 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1550instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1551the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1552 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1553 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1554
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1555</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1556
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1557<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1558<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1559Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1560<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1561<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1562<pre> ret &lt;type&gt; &lt;value&gt; <i>; Return a value from a non-void function</i>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1563 ret void <i>; Return from void function</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1564</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1565<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1566<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1567value) from a function back to the caller.</p>
John Criswell417228d2004-04-09 16:48:45 +0000[diff] [blame]1568<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1569returns a value and then causes control flow, and one that just causes
1570control flow to occur.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1571<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1572<p>The '<tt>ret</tt>' instruction may return any '<a
1573 href="#t_firstclass">first class</a>' type. Notice that a function is
1574not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1575instruction inside of the function that returns a value that does not
1576match the return type of the function.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1577<h5>Semantics:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1578<p>When the '<tt>ret</tt>' instruction is executed, control flow
1579returns back to the calling function's context. If the caller is a "<a
John Criswell40db33f2004-06-25 15:16:57 +0000[diff] [blame]1580 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1581the instruction after the call. If the caller was an "<a
1582 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1583at the beginning of the "normal" destination block. If the instruction
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1584returns a value, that value shall set the call or invoke instruction's
1585return value.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1586<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1587<pre> ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1588 ret void <i>; Return from a void function</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1589</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1590</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1591<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1592<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1593<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1594<h5>Syntax:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1595<pre> br i1 &lt;cond&gt;, label &lt;iftrue&gt;, label &lt;iffalse&gt;<br> br label &lt;dest&gt; <i>; Unconditional branch</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1596</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1597<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1598<p>The '<tt>br</tt>' instruction is used to cause control flow to
1599transfer to a different basic block in the current function. There are
1600two forms of this instruction, corresponding to a conditional branch
1601and an unconditional branch.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1602<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1603<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1604single '<tt>i1</tt>' value and two '<tt>label</tt>' values. The
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]1605unconditional form of the '<tt>br</tt>' instruction takes a single
1606'<tt>label</tt>' value as a target.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1607<h5>Semantics:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1608<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1609argument is evaluated. If the value is <tt>true</tt>, control flows
1610to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1611control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1612<h5>Example:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1613<pre>Test:<br> %cond = <a href="#i_icmp">icmp</a> eq, i32 %a, %b<br> br i1 %cond, label %IfEqual, label %IfUnequal<br>IfEqual:<br> <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1614 href="#i_ret">ret</a> i32 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> i32 0<br></pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1615</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1616<!-- _______________________________________________________________________ -->
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1617<div class="doc_subsubsection">
1618 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1619</div>
1620
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1621<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1622<h5>Syntax:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1623
1624<pre>
1625 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1626</pre>
1627
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1628<h5>Overview:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1629
1630<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1631several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1632instruction, allowing a branch to occur to one of many possible
1633destinations.</p>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1634
1635
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1636<h5>Arguments:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1637
1638<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1639comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1640an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1641table is not allowed to contain duplicate constant entries.</p>
1642
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1643<h5>Semantics:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1644
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1645<p>The <tt>switch</tt> instruction specifies a table of values and
1646destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswellbcbb18c2004-06-25 16:05:06 +0000[diff] [blame]1647table is searched for the given value. If the value is found, control flow is
1648transfered to the corresponding destination; otherwise, control flow is
1649transfered to the default destination.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1650
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1651<h5>Implementation:</h5>
1652
1653<p>Depending on properties of the target machine and the particular
1654<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswellbcbb18c2004-06-25 16:05:06 +0000[diff] [blame]1655ways. For example, it could be generated as a series of chained conditional
1656branches or with a lookup table.</p>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1657
1658<h5>Example:</h5>
1659
1660<pre>
1661 <i>; Emulate a conditional br instruction</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1662 %Val = <a href="#i_zext">zext</a> i1 %value to i32
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1663 switch i32 %Val, label %truedest [i32 0, label %falsedest ]
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1664
1665 <i>; Emulate an unconditional br instruction</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1666 switch i32 0, label %dest [ ]
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1667
1668 <i>; Implement a jump table:</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1669 switch i32 %val, label %otherwise [ i32 0, label %onzero
1670 i32 1, label %onone
1671 i32 2, label %ontwo ]
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1672</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1673</div>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1674
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1675<!-- _______________________________________________________________________ -->
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1676<div class="doc_subsubsection">
1677 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1678</div>
1679
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1680<div class="doc_text">
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1681
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1682<h5>Syntax:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1683
1684<pre>
1685 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)
Chris Lattner6b7a0082006-05-14 18:23:06 +0000[diff] [blame]1686 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1687</pre>
1688
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]1689<h5>Overview:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1690
1691<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1692function, with the possibility of control flow transfer to either the
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1693'<tt>normal</tt>' label or the
1694'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1695"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1696"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1697href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1698continued at the dynamically nearest "exception" label.</p>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1699
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1700<h5>Arguments:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1701
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1702<p>This instruction requires several arguments:</p>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1703
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1704<ol>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1705 <li>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]1706 The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1707 convention</a> the call should use. If none is specified, the call defaults
1708 to using C calling conventions.
1709 </li>
1710 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1711 function value being invoked. In most cases, this is a direct function
1712 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1713 an arbitrary pointer to function value.
1714 </li>
1715
1716 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1717 function to be invoked. </li>
1718
1719 <li>'<tt>function args</tt>': argument list whose types match the function
1720 signature argument types. If the function signature indicates the function
1721 accepts a variable number of arguments, the extra arguments can be
1722 specified. </li>
1723
1724 <li>'<tt>normal label</tt>': the label reached when the called function
1725 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1726
1727 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1728 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1729
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1730</ol>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1731
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1732<h5>Semantics:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1733
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1734<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1735href="#i_call">call</a></tt>' instruction in most regards. The primary
1736difference is that it establishes an association with a label, which is used by
1737the runtime library to unwind the stack.</p>
1738
1739<p>This instruction is used in languages with destructors to ensure that proper
1740cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1741exception. Additionally, this is important for implementation of
1742'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1743
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1744<h5>Example:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1745<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1746 %retval = invoke i32 %Test(i32 15) to label %Continue
1747 unwind label %TestCleanup <i>; {i32}:retval set</i>
1748 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Test(i32 15) to label %Continue
1749 unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1750</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1751</div>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1752
1753
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1754<!-- _______________________________________________________________________ -->
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1755
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1756<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1757Instruction</a> </div>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1758
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1759<div class="doc_text">
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1760
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1761<h5>Syntax:</h5>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1762<pre>
1763 unwind
1764</pre>
1765
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1766<h5>Overview:</h5>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1767
1768<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1769at the first callee in the dynamic call stack which used an <a
1770href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1771primarily used to implement exception handling.</p>
1772
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1773<h5>Semantics:</h5>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1774
1775<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
1776immediately halt. The dynamic call stack is then searched for the first <a
1777href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
1778execution continues at the "exceptional" destination block specified by the
1779<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
1780dynamic call chain, undefined behavior results.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1781</div>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1782
1783<!-- _______________________________________________________________________ -->
1784
1785<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
1786Instruction</a> </div>
1787
1788<div class="doc_text">
1789
1790<h5>Syntax:</h5>
1791<pre>
1792 unreachable
1793</pre>
1794
1795<h5>Overview:</h5>
1796
1797<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
1798instruction is used to inform the optimizer that a particular portion of the
1799code is not reachable. This can be used to indicate that the code after a
1800no-return function cannot be reached, and other facts.</p>
1801
1802<h5>Semantics:</h5>
1803
1804<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
1805</div>
1806
1807
1808
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1809<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1810<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1811<div class="doc_text">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1812<p>Binary operators are used to do most of the computation in a
1813program. They require two operands, execute an operation on them, and
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1814produce a single value. The operands might represent
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1815multiple data, as is the case with the <a href="#t_vector">vector</a> data type.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1816The result value of a binary operator is not
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1817necessarily the same type as its operands.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1818<p>There are several different binary operators:</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1819</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1820<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1821<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
1822Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1823<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1824<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1825<pre> &lt;result&gt; = add &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1826</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1827<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1828<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1829<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1830<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1831 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1832 This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1833Both arguments must have identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1834<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1835<p>The value produced is the integer or floating point sum of the two
1836operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1837<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1838<pre> &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1839</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1840</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1841<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1842<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
1843Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1844<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1845<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1846<pre> &lt;result&gt; = sub &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1847</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1848<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1849<p>The '<tt>sub</tt>' instruction returns the difference of its two
1850operands.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1851<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
1852instruction present in most other intermediate representations.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1853<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1854<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1855 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1856values.
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1857This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1858Both arguments must have identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1859<h5>Semantics:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1860<p>The value produced is the integer or floating point difference of
1861the two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1862<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1863<pre> &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
1864 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1865</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1866</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1867<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1868<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
1869Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1870<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1871<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1872<pre> &lt;result&gt; = mul &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1873</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1874<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1875<p>The '<tt>mul</tt>' instruction returns the product of its two
1876operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1877<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1878<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1879 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1880values.
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1881This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1882Both arguments must have identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1883<h5>Semantics:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1884<p>The value produced is the integer or floating point product of the
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1885two operands.</p>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1886<p>Because the operands are the same width, the result of an integer
1887multiplication is the same whether the operands should be deemed unsigned or
1888signed.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1889<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1890<pre> &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1891</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1892</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1893<!-- _______________________________________________________________________ -->
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1894<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
1895</a></div>
1896<div class="doc_text">
1897<h5>Syntax:</h5>
1898<pre> &lt;result&gt; = udiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1899</pre>
1900<h5>Overview:</h5>
1901<p>The '<tt>udiv</tt>' instruction returns the quotient of its two
1902operands.</p>
1903<h5>Arguments:</h5>
1904<p>The two arguments to the '<tt>udiv</tt>' instruction must be
1905<a href="#t_integer">integer</a> values. Both arguments must have identical
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1906types. This instruction can also take <a href="#t_vector">vector</a> versions
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1907of the values in which case the elements must be integers.</p>
1908<h5>Semantics:</h5>
1909<p>The value produced is the unsigned integer quotient of the two operands. This
1910instruction always performs an unsigned division operation, regardless of
1911whether the arguments are unsigned or not.</p>
1912<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1913<pre> &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1914</pre>
1915</div>
1916<!-- _______________________________________________________________________ -->
1917<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
1918</a> </div>
1919<div class="doc_text">
1920<h5>Syntax:</h5>
1921<pre> &lt;result&gt; = sdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1922</pre>
1923<h5>Overview:</h5>
1924<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two
1925operands.</p>
1926<h5>Arguments:</h5>
1927<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
1928<a href="#t_integer">integer</a> values. Both arguments must have identical
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1929types. This instruction can also take <a href="#t_vector">vector</a> versions
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1930of the values in which case the elements must be integers.</p>
1931<h5>Semantics:</h5>
1932<p>The value produced is the signed integer quotient of the two operands. This
1933instruction always performs a signed division operation, regardless of whether
1934the arguments are signed or not.</p>
1935<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1936<pre> &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1937</pre>
1938</div>
1939<!-- _______________________________________________________________________ -->
1940<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1941Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1942<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1943<h5>Syntax:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1944<pre> &lt;result&gt; = fdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1945</pre>
1946<h5>Overview:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1947<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1948operands.</p>
1949<h5>Arguments:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1950<p>The two arguments to the '<tt>div</tt>' instruction must be
1951<a href="#t_floating">floating point</a> values. Both arguments must have
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1952identical types. This instruction can also take <a href="#t_vector">vector</a>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1953versions of the values in which case the elements must be floating point.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1954<h5>Semantics:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1955<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1956<h5>Example:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]1957<pre> &lt;result&gt; = fdiv float 4.0, %var <i>; yields {float}:result = 4.0 / %var</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1958</pre>
1959</div>
1960<!-- _______________________________________________________________________ -->
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]1961<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
1962</div>
1963<div class="doc_text">
1964<h5>Syntax:</h5>
1965<pre> &lt;result&gt; = urem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1966</pre>
1967<h5>Overview:</h5>
1968<p>The '<tt>urem</tt>' instruction returns the remainder from the
1969unsigned division of its two arguments.</p>
1970<h5>Arguments:</h5>
1971<p>The two arguments to the '<tt>urem</tt>' instruction must be
1972<a href="#t_integer">integer</a> values. Both arguments must have identical
1973types.</p>
1974<h5>Semantics:</h5>
1975<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
1976This instruction always performs an unsigned division to get the remainder,
1977regardless of whether the arguments are unsigned or not.</p>
1978<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1979<pre> &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]1980</pre>
1981
1982</div>
1983<!-- _______________________________________________________________________ -->
1984<div class="doc_subsubsection"> <a name="i_srem">'<tt>srem</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1985Instruction</a> </div>
1986<div class="doc_text">
1987<h5>Syntax:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]1988<pre> &lt;result&gt; = srem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1989</pre>
1990<h5>Overview:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]1991<p>The '<tt>srem</tt>' instruction returns the remainder from the
1992signed division of its two operands.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1993<h5>Arguments:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]1994<p>The two arguments to the '<tt>srem</tt>' instruction must be
1995<a href="#t_integer">integer</a> values. Both arguments must have identical
1996types.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1997<h5>Semantics:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]1998<p>This instruction returns the <i>remainder</i> of a division (where the result
Reid Spencer806ad6a2007-03-24 22:23:39 +0000[diff] [blame]1999has the same sign as the dividend, <tt>var1</tt>), not the <i>modulo</i>
2000operator (where the result has the same sign as the divisor, <tt>var2</tt>) of
2001a value. For more information about the difference, see <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2002 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
Reid Spencer806ad6a2007-03-24 22:23:39 +0000[diff] [blame]2003Math Forum</a>. For a table of how this is implemented in various languages,
Reid Spencerdb3b93b2007-03-24 22:40:44 +0000[diff] [blame]2004please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
Reid Spencer806ad6a2007-03-24 22:23:39 +0000[diff] [blame]2005Wikipedia: modulo operation</a>.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2006<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2007<pre> &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2008</pre>
2009
2010</div>
2011<!-- _______________________________________________________________________ -->
2012<div class="doc_subsubsection"> <a name="i_frem">'<tt>frem</tt>'
2013Instruction</a> </div>
2014<div class="doc_text">
2015<h5>Syntax:</h5>
2016<pre> &lt;result&gt; = frem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2017</pre>
2018<h5>Overview:</h5>
2019<p>The '<tt>frem</tt>' instruction returns the remainder from the
2020division of its two operands.</p>
2021<h5>Arguments:</h5>
2022<p>The two arguments to the '<tt>frem</tt>' instruction must be
2023<a href="#t_floating">floating point</a> values. Both arguments must have
2024identical types.</p>
2025<h5>Semantics:</h5>
2026<p>This instruction returns the <i>remainder</i> of a division.</p>
2027<h5>Example:</h5>
2028<pre> &lt;result&gt; = frem float 4.0, %var <i>; yields {float}:result = 4.0 % %var</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2029</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2030</div>
Robert Bocchino820bc75b2006-02-17 21:18:08 +0000[diff] [blame]2031
Reid Spencer2ab01932007-02-02 13:57:07 +0000[diff] [blame]2032<!-- ======================================================================= -->
2033<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
2034Operations</a> </div>
2035<div class="doc_text">
2036<p>Bitwise binary operators are used to do various forms of
2037bit-twiddling in a program. They are generally very efficient
2038instructions and can commonly be strength reduced from other
2039instructions. They require two operands, execute an operation on them,
2040and produce a single value. The resulting value of the bitwise binary
2041operators is always the same type as its first operand.</p>
2042</div>
2043
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2044<!-- _______________________________________________________________________ -->
2045<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
2046Instruction</a> </div>
2047<div class="doc_text">
2048<h5>Syntax:</h5>
2049<pre> &lt;result&gt; = shl &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2050</pre>
2051<h5>Overview:</h5>
2052<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
2053the left a specified number of bits.</p>
2054<h5>Arguments:</h5>
2055<p>Both arguments to the '<tt>shl</tt>' instruction must be the same <a
2056 href="#t_integer">integer</a> type.</p>
2057<h5>Semantics:</h5>
2058<p>The value produced is <tt>var1</tt> * 2<sup><tt>var2</tt></sup>.</p>
2059<h5>Example:</h5><pre>
2060 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
2061 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
2062 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
2063</pre>
2064</div>
2065<!-- _______________________________________________________________________ -->
2066<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
2067Instruction</a> </div>
2068<div class="doc_text">
2069<h5>Syntax:</h5>
2070<pre> &lt;result&gt; = lshr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2071</pre>
2072
2073<h5>Overview:</h5>
2074<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
2075operand shifted to the right a specified number of bits.</p>
2076
2077<h5>Arguments:</h5>
2078<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
2079<a href="#t_integer">integer</a> type.</p>
2080
2081<h5>Semantics:</h5>
2082<p>This instruction always performs a logical shift right operation. The most
2083significant bits of the result will be filled with zero bits after the
2084shift.</p>
2085
2086<h5>Example:</h5>
2087<pre>
2088 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
2089 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
2090 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
2091 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
2092</pre>
2093</div>
2094
Reid Spencer2ab01932007-02-02 13:57:07 +0000[diff] [blame]2095<!-- _______________________________________________________________________ -->
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2096<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
2097Instruction</a> </div>
2098<div class="doc_text">
2099
2100<h5>Syntax:</h5>
2101<pre> &lt;result&gt; = ashr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2102</pre>
2103
2104<h5>Overview:</h5>
2105<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
2106operand shifted to the right a specified number of bits.</p>
2107
2108<h5>Arguments:</h5>
2109<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
2110<a href="#t_integer">integer</a> type.</p>
2111
2112<h5>Semantics:</h5>
2113<p>This instruction always performs an arithmetic shift right operation,
2114The most significant bits of the result will be filled with the sign bit
2115of <tt>var1</tt>.</p>
2116
2117<h5>Example:</h5>
2118<pre>
2119 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
2120 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
2121 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
2122 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
2123</pre>
2124</div>
2125
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2126<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2127<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
2128Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2129<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2130<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2131<pre> &lt;result&gt; = and &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2132</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2133<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2134<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
2135its two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2136<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2137<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2138 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2139identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2140<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2141<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2142<p> </p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2143<div style="align: center">
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2144<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2145 <tbody>
2146 <tr>
2147 <td>In0</td>
2148 <td>In1</td>
2149 <td>Out</td>
2150 </tr>
2151 <tr>
2152 <td>0</td>
2153 <td>0</td>
2154 <td>0</td>
2155 </tr>
2156 <tr>
2157 <td>0</td>
2158 <td>1</td>
2159 <td>0</td>
2160 </tr>
2161 <tr>
2162 <td>1</td>
2163 <td>0</td>
2164 <td>0</td>
2165 </tr>
2166 <tr>
2167 <td>1</td>
2168 <td>1</td>
2169 <td>1</td>
2170 </tr>
2171 </tbody>
2172</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2173</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2174<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2175<pre> &lt;result&gt; = and i32 4, %var <i>; yields {i32}:result = 4 &amp; %var</i>
2176 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
2177 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2178</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2179</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2180<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2181<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2182<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2183<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2184<pre> &lt;result&gt; = or &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2185</pre>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2186<h5>Overview:</h5>
2187<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
2188or of its two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2189<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2190<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2191 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2192identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2193<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2194<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2195<p> </p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2196<div style="align: center">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2197<table border="1" cellspacing="0" cellpadding="4">
2198 <tbody>
2199 <tr>
2200 <td>In0</td>
2201 <td>In1</td>
2202 <td>Out</td>
2203 </tr>
2204 <tr>
2205 <td>0</td>
2206 <td>0</td>
2207 <td>0</td>
2208 </tr>
2209 <tr>
2210 <td>0</td>
2211 <td>1</td>
2212 <td>1</td>
2213 </tr>
2214 <tr>
2215 <td>1</td>
2216 <td>0</td>
2217 <td>1</td>
2218 </tr>
2219 <tr>
2220 <td>1</td>
2221 <td>1</td>
2222 <td>1</td>
2223 </tr>
2224 </tbody>
2225</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2226</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2227<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2228<pre> &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
2229 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
2230 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2231</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2232</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2233<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2234<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
2235Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2236<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2237<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2238<pre> &lt;result&gt; = xor &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2239</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2240<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2241<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
2242or of its two operands. The <tt>xor</tt> is used to implement the
2243"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2244<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2245<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2246 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2247identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2248<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2249<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2250<p> </p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2251<div style="align: center">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2252<table border="1" cellspacing="0" cellpadding="4">
2253 <tbody>
2254 <tr>
2255 <td>In0</td>
2256 <td>In1</td>
2257 <td>Out</td>
2258 </tr>
2259 <tr>
2260 <td>0</td>
2261 <td>0</td>
2262 <td>0</td>
2263 </tr>
2264 <tr>
2265 <td>0</td>
2266 <td>1</td>
2267 <td>1</td>
2268 </tr>
2269 <tr>
2270 <td>1</td>
2271 <td>0</td>
2272 <td>1</td>
2273 </tr>
2274 <tr>
2275 <td>1</td>
2276 <td>1</td>
2277 <td>0</td>
2278 </tr>
2279 </tbody>
2280</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2281</div>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2282<p> </p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2283<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2284<pre> &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
2285 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
2286 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
2287 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2288</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2289</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2290
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2291<!-- ======================================================================= -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2292<div class="doc_subsection">
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2293 <a name="vectorops">Vector Operations</a>
2294</div>
2295
2296<div class="doc_text">
2297
2298<p>LLVM supports several instructions to represent vector operations in a
2299target-independent manner. This instructions cover the element-access and
2300vector-specific operations needed to process vectors effectively. While LLVM
2301does directly support these vector operations, many sophisticated algorithms
2302will want to use target-specific intrinsics to take full advantage of a specific
2303target.</p>
2304
2305</div>
2306
2307<!-- _______________________________________________________________________ -->
2308<div class="doc_subsubsection">
2309 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
2310</div>
2311
2312<div class="doc_text">
2313
2314<h5>Syntax:</h5>
2315
2316<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2317 &lt;result&gt; = extractelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, i32 &lt;idx&gt; <i>; yields &lt;ty&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2318</pre>
2319
2320<h5>Overview:</h5>
2321
2322<p>
2323The '<tt>extractelement</tt>' instruction extracts a single scalar
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2324element from a vector at a specified index.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2325</p>
2326
2327
2328<h5>Arguments:</h5>
2329
2330<p>
2331The first operand of an '<tt>extractelement</tt>' instruction is a
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2332value of <a href="#t_vector">vector</a> type. The second operand is
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2333an index indicating the position from which to extract the element.
2334The index may be a variable.</p>
2335
2336<h5>Semantics:</h5>
2337
2338<p>
2339The result is a scalar of the same type as the element type of
2340<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
2341<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
2342results are undefined.
2343</p>
2344
2345<h5>Example:</h5>
2346
2347<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2348 %result = extractelement &lt;4 x i32&gt; %vec, i32 0 <i>; yields i32</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2349</pre>
2350</div>
2351
2352
2353<!-- _______________________________________________________________________ -->
2354<div class="doc_subsubsection">
2355 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
2356</div>
2357
2358<div class="doc_text">
2359
2360<h5>Syntax:</h5>
2361
2362<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2363 &lt;result&gt; = insertelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, &lt;ty&gt; &lt;elt&gt, i32 &lt;idx&gt; <i>; yields &lt;n x &lt;ty&gt;&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2364</pre>
2365
2366<h5>Overview:</h5>
2367
2368<p>
2369The '<tt>insertelement</tt>' instruction inserts a scalar
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2370element into a vector at a specified index.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2371</p>
2372
2373
2374<h5>Arguments:</h5>
2375
2376<p>
2377The first operand of an '<tt>insertelement</tt>' instruction is a
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2378value of <a href="#t_vector">vector</a> type. The second operand is a
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2379scalar value whose type must equal the element type of the first
2380operand. The third operand is an index indicating the position at
2381which to insert the value. The index may be a variable.</p>
2382
2383<h5>Semantics:</h5>
2384
2385<p>
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2386The result is a vector of the same type as <tt>val</tt>. Its
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2387element values are those of <tt>val</tt> except at position
2388<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
2389exceeds the length of <tt>val</tt>, the results are undefined.
2390</p>
2391
2392<h5>Example:</h5>
2393
2394<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2395 %result = insertelement &lt;4 x i32&gt; %vec, i32 1, i32 0 <i>; yields &lt;4 x i32&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2396</pre>
2397</div>
2398
2399<!-- _______________________________________________________________________ -->
2400<div class="doc_subsubsection">
2401 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
2402</div>
2403
2404<div class="doc_text">
2405
2406<h5>Syntax:</h5>
2407
2408<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2409 &lt;result&gt; = shufflevector &lt;n x &lt;ty&gt;&gt; &lt;v1&gt;, &lt;n x &lt;ty&gt;&gt; &lt;v2&gt;, &lt;n x i32&gt; &lt;mask&gt; <i>; yields &lt;n x &lt;ty&gt;&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2410</pre>
2411
2412<h5>Overview:</h5>
2413
2414<p>
2415The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
2416from two input vectors, returning a vector of the same type.
2417</p>
2418
2419<h5>Arguments:</h5>
2420
2421<p>
2422The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
2423with types that match each other and types that match the result of the
2424instruction. The third argument is a shuffle mask, which has the same number
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2425of elements as the other vector type, but whose element type is always 'i32'.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2426</p>
2427
2428<p>
2429The shuffle mask operand is required to be a constant vector with either
2430constant integer or undef values.
2431</p>
2432
2433<h5>Semantics:</h5>
2434
2435<p>
2436The elements of the two input vectors are numbered from left to right across
2437both of the vectors. The shuffle mask operand specifies, for each element of
2438the result vector, which element of the two input registers the result element
2439gets. The element selector may be undef (meaning "don't care") and the second
2440operand may be undef if performing a shuffle from only one vector.
2441</p>
2442
2443<h5>Example:</h5>
2444
2445<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2446 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
2447 &lt;4 x i32&gt; &lt;i32 0, i32 4, i32 1, i32 5&gt; <i>; yields &lt;4 x i32&gt;</i>
2448 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; undef,
2449 &lt;4 x i32&gt; &lt;i32 0, i32 1, i32 2, i32 3&gt; <i>; yields &lt;4 x i32&gt;</i> - Identity shuffle.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2450</pre>
2451</div>
2452
Tanya Lattnerb138bbe2006-04-14 19:24:33 +0000[diff] [blame]2453
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2454<!-- ======================================================================= -->
2455<div class="doc_subsection">
Chris Lattner6ab66722006-08-15 00:45:58 +0000[diff] [blame]2456 <a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2457</div>
2458
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2459<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2460
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2461<p>A key design point of an SSA-based representation is how it
2462represents memory. In LLVM, no memory locations are in SSA form, which
2463makes things very simple. This section describes how to read, write,
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]2464allocate, and free memory in LLVM.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2465
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2466</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2467
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2468<!-- _______________________________________________________________________ -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2469<div class="doc_subsubsection">
2470 <a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
2471</div>
2472
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2473<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2474
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2475<h5>Syntax:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2476
2477<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2478 &lt;result&gt; = malloc &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2479</pre>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2480
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2481<h5>Overview:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2482
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2483<p>The '<tt>malloc</tt>' instruction allocates memory from the system
2484heap and returns a pointer to it.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2485
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2486<h5>Arguments:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2487
2488<p>The '<tt>malloc</tt>' instruction allocates
2489<tt>sizeof(&lt;type&gt;)*NumElements</tt>
John Criswella92e5862004-02-24 16:13:56 +0000[diff] [blame]2490bytes of memory from the operating system and returns a pointer of the
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2491appropriate type to the program. If "NumElements" is specified, it is the
2492number of elements allocated. If an alignment is specified, the value result
2493of the allocation is guaranteed to be aligned to at least that boundary. If
2494not specified, or if zero, the target can choose to align the allocation on any
2495convenient boundary.</p>
2496
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2497<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2498
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2499<h5>Semantics:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2500
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2501<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
2502a pointer is returned.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2503
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2504<h5>Example:</h5>
2505
2506<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2507 %array = malloc [4 x i8 ] <i>; yields {[%4 x i8]*}:array</i>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2508
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2509 %size = <a href="#i_add">add</a> i32 2, 2 <i>; yields {i32}:size = i32 4</i>
2510 %array1 = malloc i8, i32 4 <i>; yields {i8*}:array1</i>
2511 %array2 = malloc [12 x i8], i32 %size <i>; yields {[12 x i8]*}:array2</i>
2512 %array3 = malloc i32, i32 4, align 1024 <i>; yields {i32*}:array3</i>
2513 %array4 = malloc i32, align 1024 <i>; yields {i32*}:array4</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2514</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2515</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2516
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2517<!-- _______________________________________________________________________ -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2518<div class="doc_subsubsection">
2519 <a name="i_free">'<tt>free</tt>' Instruction</a>
2520</div>
2521
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2522<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2523
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2524<h5>Syntax:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2525
2526<pre>
2527 free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2528</pre>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2529
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2530<h5>Overview:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2531
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2532<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2533memory heap to be reallocated in the future.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2534
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2535<h5>Arguments:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2536
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2537<p>'<tt>value</tt>' shall be a pointer value that points to a value
2538that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
2539instruction.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2540
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2541<h5>Semantics:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2542
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]2543<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2544after this instruction executes.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2545
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2546<h5>Example:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2547
2548<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2549 %array = <a href="#i_malloc">malloc</a> [4 x i8] <i>; yields {[4 x i8]*}:array</i>
2550 free [4 x i8]* %array
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2551</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2552</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2553
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2554<!-- _______________________________________________________________________ -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2555<div class="doc_subsubsection">
2556 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
2557</div>
2558
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2559<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2560
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2561<h5>Syntax:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2562
2563<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2564 &lt;result&gt; = alloca &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2565</pre>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2566
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2567<h5>Overview:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2568
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2569<p>The '<tt>alloca</tt>' instruction allocates memory on the current
2570stack frame of the procedure that is live until the current function
2571returns to its caller.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2572
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2573<h5>Arguments:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2574
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]2575<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2576bytes of memory on the runtime stack, returning a pointer of the
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2577appropriate type to the program. If "NumElements" is specified, it is the
2578number of elements allocated. If an alignment is specified, the value result
2579of the allocation is guaranteed to be aligned to at least that boundary. If
2580not specified, or if zero, the target can choose to align the allocation on any
2581convenient boundary.</p>
2582
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2583<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2584
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2585<h5>Semantics:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2586
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2587<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2588memory is automatically released when the function returns. The '<tt>alloca</tt>'
2589instruction is commonly used to represent automatic variables that must
2590have an address available. When the function returns (either with the <tt><a
John Criswellc932bef2005-05-12 16:55:34 +0000[diff] [blame]2591 href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2592instructions), the memory is reclaimed.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2593
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2594<h5>Example:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2595
2596<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2597 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
2598 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
2599 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
2600 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2601</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2602</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2603
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2604<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2605<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
2606Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2607<div class="doc_text">
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2608<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2609<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 Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2610<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2611<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2612<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2613<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]2614address from which to load. The pointer must point to a <a
Chris Lattner10ee9652004-06-03 22:57:15 +0000[diff] [blame]2615 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]2616marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2617the number or order of execution of this <tt>load</tt> with other
2618volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
2619instructions. </p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2620<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2621<p>The location of memory pointed to is loaded.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2622<h5>Examples:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2623<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2624 <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2625 href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
2626 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2627</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2628</div>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2629<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2630<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
2631Instruction</a> </div>
Reid Spencera89fb182006-11-09 21:18:01 +0000[diff] [blame]2632<div class="doc_text">
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2633<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2634<pre> store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt; <i>; yields {void}</i>
Chris Lattner12d456c2003-09-08 18:27:49 +0000[diff] [blame]2635 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt; <i>; yields {void}</i>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2636</pre>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2637<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2638<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2639<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2640<p>There are two arguments to the '<tt>store</tt>' instruction: a value
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]2641to store and an address in which to store it. The type of the '<tt>&lt;pointer&gt;</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2642operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2643operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2644optimizer is not allowed to modify the number or order of execution of
2645this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
2646 href="#i_store">store</a></tt> instructions.</p>
2647<h5>Semantics:</h5>
2648<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
2649at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2650<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2651<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2652 <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2653 href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
2654 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2655</pre>
Reid Spencer443460a2006-11-09 21:15:49 +0000[diff] [blame]2656</div>
2657
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2658<!-- _______________________________________________________________________ -->
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2659<div class="doc_subsubsection">
2660 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
2661</div>
2662
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2663<div class="doc_text">
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2664<h5>Syntax:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2665<pre>
2666 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
2667</pre>
2668
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2669<h5>Overview:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2670
2671<p>
2672The '<tt>getelementptr</tt>' instruction is used to get the address of a
2673subelement of an aggregate data structure.</p>
2674
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2675<h5>Arguments:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2676
Reid Spencercee005c2006-12-04 21:29:24 +0000[diff] [blame]2677<p>This instruction takes a list of integer operands that indicate what
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2678elements of the aggregate object to index to. The actual types of the arguments
2679provided depend on the type of the first pointer argument. The
2680'<tt>getelementptr</tt>' instruction is used to index down through the type
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]2681levels of a structure or to a specific index in an array. When indexing into a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2682structure, only <tt>i32</tt> integer constants are allowed. When indexing
Reid Spencercee005c2006-12-04 21:29:24 +0000[diff] [blame]2683into an array or pointer, only integers of 32 or 64 bits are allowed, and will
2684be sign extended to 64-bit values.</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2685
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2686<p>For example, let's consider a C code fragment and how it gets
2687compiled to LLVM:</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2688
2689<pre>
2690 struct RT {
2691 char A;
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2692 i32 B[10][20];
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2693 char C;
2694 };
2695 struct ST {
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2696 i32 X;
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2697 double Y;
2698 struct RT Z;
2699 };
2700
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2701 define i32 *foo(struct ST *s) {
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2702 return &amp;s[1].Z.B[5][13];
2703 }
2704</pre>
2705
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2706<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2707
2708<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2709 %RT = type { i8 , [10 x [20 x i32]], i8 }
2710 %ST = type { i32, double, %RT }
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2711
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2712 define i32* %foo(%ST* %s) {
Brian Gaeke317ef962004-07-02 21:08:14 +0000[diff] [blame]2713 entry:
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2714 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
2715 ret i32* %reg
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2716 }
2717</pre>
2718
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2719<h5>Semantics:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2720
2721<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2722on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
Reid Spencercee005c2006-12-04 21:29:24 +0000[diff] [blame]2723and <a href="#t_array">array</a> types can use a 32-bit or 64-bit
Reid Spencerc0312692006-12-03 16:53:48 +0000[diff] [blame]2724<a href="#t_integer">integer</a> type but the value will always be sign extended
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2725to 64-bits. <a href="#t_struct">Structure</a> types, require <tt>i32</tt>
Reid Spencerc0312692006-12-03 16:53:48 +0000[diff] [blame]2726<b>constants</b>.</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2727
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2728<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2729type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2730}</tt>' type, a structure. The second index indexes into the third element of
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2731the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
2732i8 }</tt>' type, another structure. The third index indexes into the second
2733element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2734array. The two dimensions of the array are subscripted into, yielding an
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2735'<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
2736to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2737
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2738<p>Note that it is perfectly legal to index partially through a
2739structure, returning a pointer to an inner element. Because of this,
2740the LLVM code for the given testcase is equivalent to:</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2741
2742<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2743 define i32* %foo(%ST* %s) {
2744 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
2745 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
2746 %t3 = getelementptr %RT* %t2, i32 0, i32 1 <i>; yields [10 x [20 x i32]]*:%t3</i>
2747 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
2748 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
2749 ret i32* %t5
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2750 }
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]2751</pre>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]2752
2753<p>Note that it is undefined to access an array out of bounds: array and
2754pointer indexes must always be within the defined bounds of the array type.
2755The one exception for this rules is zero length arrays. These arrays are
2756defined to be accessible as variable length arrays, which requires access
2757beyond the zero'th element.</p>
2758
Chris Lattner6ab66722006-08-15 00:45:58 +0000[diff] [blame]2759<p>The getelementptr instruction is often confusing. For some more insight
2760into how it works, see <a href="GetElementPtr.html">the getelementptr
2761FAQ</a>.</p>
2762
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2763<h5>Example:</h5>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]2764
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2765<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2766 <i>; yields [12 x i8]*:aptr</i>
2767 %aptr = getelementptr {i32, [12 x i8]}* %sptr, i64 0, i32 1
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2768</pre>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2769</div>
Reid Spencer443460a2006-11-09 21:15:49 +0000[diff] [blame]2770
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2771<!-- ======================================================================= -->
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]2772<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2773</div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2774<div class="doc_text">
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]2775<p>The instructions in this category are the conversion instructions (casting)
2776which all take a single operand and a type. They perform various bit conversions
2777on the operand.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2778</div>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]2779
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]2780<!-- _______________________________________________________________________ -->
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]2781<div class="doc_subsubsection">
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2782 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
2783</div>
2784<div class="doc_text">
2785
2786<h5>Syntax:</h5>
2787<pre>
2788 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2789</pre>
2790
2791<h5>Overview:</h5>
2792<p>
2793The '<tt>trunc</tt>' instruction truncates its operand to the type <tt>ty2</tt>.
2794</p>
2795
2796<h5>Arguments:</h5>
2797<p>
2798The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
2799be an <a href="#t_integer">integer</a> type, and a type that specifies the size
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2800and type of the result, which must be an <a href="#t_integer">integer</a>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2801type. The bit size of <tt>value</tt> must be larger than the bit size of
2802<tt>ty2</tt>. Equal sized types are not allowed.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2803
2804<h5>Semantics:</h5>
2805<p>
2806The '<tt>trunc</tt>' instruction truncates the high order bits in <tt>value</tt>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2807and converts the remaining bits to <tt>ty2</tt>. Since the source size must be
2808larger than the destination size, <tt>trunc</tt> cannot be a <i>no-op cast</i>.
2809It will always truncate bits.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2810
2811<h5>Example:</h5>
2812<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2813 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]2814 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
2815 %Y = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2816</pre>
2817</div>
2818
2819<!-- _______________________________________________________________________ -->
2820<div class="doc_subsubsection">
2821 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
2822</div>
2823<div class="doc_text">
2824
2825<h5>Syntax:</h5>
2826<pre>
2827 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2828</pre>
2829
2830<h5>Overview:</h5>
2831<p>The '<tt>zext</tt>' instruction zero extends its operand to type
2832<tt>ty2</tt>.</p>
2833
2834
2835<h5>Arguments:</h5>
2836<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2837<a href="#t_integer">integer</a> type, and a type to cast it to, which must
2838also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2839<tt>value</tt> must be smaller than the bit size of the destination type,
2840<tt>ty2</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2841
2842<h5>Semantics:</h5>
2843<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
2844bits until it reaches the size of the destination type, <tt>ty2</tt>. When the
2845the operand and the type are the same size, no bit filling is done and the
2846cast is considered a <i>no-op cast</i> because no bits change (only the type
2847changes).</p>
2848
Reid Spencer07c9c682007-01-12 15:46:11 +0000[diff] [blame]2849<p>When zero extending from i1, the result will always be either 0 or 1.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2850
2851<h5>Example:</h5>
2852<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2853 %X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]2854 %Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2855</pre>
2856</div>
2857
2858<!-- _______________________________________________________________________ -->
2859<div class="doc_subsubsection">
2860 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
2861</div>
2862<div class="doc_text">
2863
2864<h5>Syntax:</h5>
2865<pre>
2866 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2867</pre>
2868
2869<h5>Overview:</h5>
2870<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
2871
2872<h5>Arguments:</h5>
2873<p>
2874The '<tt>sext</tt>' instruction takes a value to cast, which must be of
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2875<a href="#t_integer">integer</a> type, and a type to cast it to, which must
2876also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2877<tt>value</tt> must be smaller than the bit size of the destination type,
2878<tt>ty2</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2879
2880<h5>Semantics:</h5>
2881<p>
2882The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
2883bit (highest order bit) of the <tt>value</tt> until it reaches the bit size of
2884the type <tt>ty2</tt>. When the the operand and the type are the same size,
2885no bit filling is done and the cast is considered a <i>no-op cast</i> because
2886no bits change (only the type changes).</p>
2887
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]2888<p>When sign extending from i1, the extension always results in -1 or 0.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2889
2890<h5>Example:</h5>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2891<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2892 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]2893 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2894</pre>
2895</div>
2896
2897<!-- _______________________________________________________________________ -->
2898<div class="doc_subsubsection">
Reid Spencer2e2740d2006-11-09 21:48:10 +0000[diff] [blame]2899 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
2900</div>
2901
2902<div class="doc_text">
2903
2904<h5>Syntax:</h5>
2905
2906<pre>
2907 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2908</pre>
2909
2910<h5>Overview:</h5>
2911<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
2912<tt>ty2</tt>.</p>
2913
2914
2915<h5>Arguments:</h5>
2916<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
2917 point</a> value to cast and a <a href="#t_floating">floating point</a> type to
2918cast it to. The size of <tt>value</tt> must be larger than the size of
2919<tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
2920<i>no-op cast</i>.</p>
2921
2922<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2923<p> The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
2924<a href="#t_floating">floating point</a> type to a smaller
2925<a href="#t_floating">floating point</a> type. If the value cannot fit within
2926the destination type, <tt>ty2</tt>, then the results are undefined.</p>
Reid Spencer2e2740d2006-11-09 21:48:10 +0000[diff] [blame]2927
2928<h5>Example:</h5>
2929<pre>
2930 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
2931 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
2932</pre>
2933</div>
2934
2935<!-- _______________________________________________________________________ -->
2936<div class="doc_subsubsection">
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2937 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
2938</div>
2939<div class="doc_text">
2940
2941<h5>Syntax:</h5>
2942<pre>
2943 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2944</pre>
2945
2946<h5>Overview:</h5>
2947<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
2948floating point value.</p>
2949
2950<h5>Arguments:</h5>
2951<p>The '<tt>fpext</tt>' instruction takes a
2952<a href="#t_floating">floating point</a> <tt>value</tt> to cast,
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2953and a <a href="#t_floating">floating point</a> type to cast it to. The source
2954type must be smaller than the destination type.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2955
2956<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2957<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]2958<a href="#t_floating">floating point</a> type to a larger
2959<a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2960used to make a <i>no-op cast</i> because it always changes bits. Use
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]2961<tt>bitcast</tt> to make a <i>no-op cast</i> for a floating point cast.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2962
2963<h5>Example:</h5>
2964<pre>
2965 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
2966 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
2967</pre>
2968</div>
2969
2970<!-- _______________________________________________________________________ -->
2971<div class="doc_subsubsection">
Reid Spencer2eadb532007-01-21 00:29:26 +0000[diff] [blame]2972 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2973</div>
2974<div class="doc_text">
2975
2976<h5>Syntax:</h5>
2977<pre>
2978 &lt;result&gt; = fp2uint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2979</pre>
2980
2981<h5>Overview:</h5>
2982<p>The '<tt>fp2uint</tt>' converts a floating point <tt>value</tt> to its
2983unsigned integer equivalent of type <tt>ty2</tt>.
2984</p>
2985
2986<h5>Arguments:</h5>
2987<p>The '<tt>fp2uint</tt>' instruction takes a value to cast, which must be a
2988<a href="#t_floating">floating point</a> value, and a type to cast it to, which
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2989must be an <a href="#t_integer">integer</a> type.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2990
2991<h5>Semantics:</h5>
2992<p> The '<tt>fp2uint</tt>' instruction converts its
2993<a href="#t_floating">floating point</a> operand into the nearest (rounding
2994towards zero) unsigned integer value. If the value cannot fit in <tt>ty2</tt>,
2995the results are undefined.</p>
2996
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]2997<p>When converting to i1, the conversion is done as a comparison against
2998zero. If the <tt>value</tt> was zero, the i1 result will be <tt>false</tt>.
2999If the <tt>value</tt> was non-zero, the i1 result will be <tt>true</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3000
3001<h5>Example:</h5>
3002<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3003 %X = fp2uint double 123.0 to i32 <i>; yields i32:123</i>
3004 %Y = fp2uint float 1.0E+300 to i1 <i>; yields i1:true</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3005 %X = fp2uint float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3006</pre>
3007</div>
3008
3009<!-- _______________________________________________________________________ -->
3010<div class="doc_subsubsection">
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3011 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3012</div>
3013<div class="doc_text">
3014
3015<h5>Syntax:</h5>
3016<pre>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3017 &lt;result&gt; = fptosi &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3018</pre>
3019
3020<h5>Overview:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3021<p>The '<tt>fptosi</tt>' instruction converts
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3022<a href="#t_floating">floating point</a> <tt>value</tt> to type <tt>ty2</tt>.
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3023</p>
3024
3025
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3026<h5>Arguments:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3027<p> The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3028<a href="#t_floating">floating point</a> value, and a type to cast it to, which
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3029must also be an <a href="#t_integer">integer</a> type.</p>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3030
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3031<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3032<p>The '<tt>fptosi</tt>' instruction converts its
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3033<a href="#t_floating">floating point</a> operand into the nearest (rounding
3034towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
3035the results are undefined.</p>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3036
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3037<p>When converting to i1, the conversion is done as a comparison against
3038zero. If the <tt>value</tt> was zero, the i1 result will be <tt>false</tt>.
3039If the <tt>value</tt> was non-zero, the i1 result will be <tt>true</tt>.</p>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3040
Chris Lattner70de6632001-07-09 00:26:23 +0000[diff] [blame]3041<h5>Example:</h5>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3042<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3043 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
3044 %Y = fptosi float 1.0E-247 to i1 <i>; yields i1:true</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3045 %X = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3046</pre>
3047</div>
3048
3049<!-- _______________________________________________________________________ -->
3050<div class="doc_subsubsection">
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3051 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3052</div>
3053<div class="doc_text">
3054
3055<h5>Syntax:</h5>
3056<pre>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3057 &lt;result&gt; = uitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3058</pre>
3059
3060<h5>Overview:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3061<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3062integer and converts that value to the <tt>ty2</tt> type.</p>
3063
3064
3065<h5>Arguments:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3066<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be an
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3067<a href="#t_integer">integer</a> value, and a type to cast it to, which must
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3068be a <a href="#t_floating">floating point</a> type.</p>
3069
3070<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3071<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3072integer quantity and converts it to the corresponding floating point value. If
3073the value cannot fit in the floating point value, the results are undefined.</p>
3074
3075
3076<h5>Example:</h5>
3077<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3078 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
3079 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3080</pre>
3081</div>
3082
3083<!-- _______________________________________________________________________ -->
3084<div class="doc_subsubsection">
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3085 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3086</div>
3087<div class="doc_text">
3088
3089<h5>Syntax:</h5>
3090<pre>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3091 &lt;result&gt; = sitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3092</pre>
3093
3094<h5>Overview:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3095<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3096integer and converts that value to the <tt>ty2</tt> type.</p>
3097
3098<h5>Arguments:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3099<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be an
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3100<a href="#t_integer">integer</a> value, and a type to cast it to, which must be
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3101a <a href="#t_floating">floating point</a> type.</p>
3102
3103<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3104<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3105integer quantity and converts it to the corresponding floating point value. If
3106the value cannot fit in the floating point value, the results are undefined.</p>
3107
3108<h5>Example:</h5>
3109<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3110 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
3111 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3112</pre>
3113</div>
3114
3115<!-- _______________________________________________________________________ -->
3116<div class="doc_subsubsection">
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3117 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
3118</div>
3119<div class="doc_text">
3120
3121<h5>Syntax:</h5>
3122<pre>
3123 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3124</pre>
3125
3126<h5>Overview:</h5>
3127<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
3128the integer type <tt>ty2</tt>.</p>
3129
3130<h5>Arguments:</h5>
3131<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3132must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3133<tt>ty2</tt>, which must be an <a href="#t_integer">integer</a> type.
3134
3135<h5>Semantics:</h5>
3136<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
3137<tt>ty2</tt> by interpreting the pointer value as an integer and either
3138truncating or zero extending that value to the size of the integer type. If
3139<tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
3140<tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
3141are the same size, then nothing is done (<i>no-op cast</i>).</p>
3142
3143<h5>Example:</h5>
3144<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3145 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit</i>
3146 %Y = ptrtoint i32* %x to i64 <i>; yields zero extend on 32-bit</i>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3147</pre>
3148</div>
3149
3150<!-- _______________________________________________________________________ -->
3151<div class="doc_subsubsection">
3152 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
3153</div>
3154<div class="doc_text">
3155
3156<h5>Syntax:</h5>
3157<pre>
3158 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3159</pre>
3160
3161<h5>Overview:</h5>
3162<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to
3163a pointer type, <tt>ty2</tt>.</p>
3164
3165<h5>Arguments:</h5>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3166<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3167value to cast, and a type to cast it to, which must be a
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]3168<a href="#t_pointer">pointer</a> type.
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3169
3170<h5>Semantics:</h5>
3171<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
3172<tt>ty2</tt> by applying either a zero extension or a truncation depending on
3173the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
3174size of a pointer then a truncation is done. If <tt>value</tt> is smaller than
3175the size of a pointer then a zero extension is done. If they are the same size,
3176nothing is done (<i>no-op cast</i>).</p>
3177
3178<h5>Example:</h5>
3179<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3180 %X = inttoptr i32 255 to i32* <i>; yields zero extend on 64-bit</i>
3181 %X = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit </i>
3182 %Y = inttoptr i16 0 to i32* <i>; yields zero extend on 32-bit</i>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3183</pre>
3184</div>
3185
3186<!-- _______________________________________________________________________ -->
3187<div class="doc_subsubsection">
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3188 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3189</div>
3190<div class="doc_text">
3191
3192<h5>Syntax:</h5>
3193<pre>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3194 &lt;result&gt; = bitcast &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3195</pre>
3196
3197<h5>Overview:</h5>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3198<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3199<tt>ty2</tt> without changing any bits.</p>
3200
3201<h5>Arguments:</h5>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3202<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3203a first class value, and a type to cast it to, which must also be a <a
3204 href="#t_firstclass">first class</a> type. The bit sizes of <tt>value</tt>
Reid Spencere3db84c2007-01-09 20:08:58 +0000[diff] [blame]3205and the destination type, <tt>ty2</tt>, must be identical. If the source
3206type is a pointer, the destination type must also be a pointer.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3207
3208<h5>Semantics:</h5>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3209<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3210<tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
3211this conversion. The conversion is done as if the <tt>value</tt> had been
3212stored to memory and read back as type <tt>ty2</tt>. Pointer types may only be
3213converted to other pointer types with this instruction. To convert pointers to
3214other types, use the <a href="#i_inttoptr">inttoptr</a> or
3215<a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3216
3217<h5>Example:</h5>
3218<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3219 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
3220 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
3221 %Z = bitcast <2xint> %V to i64; <i>; yields i64: %V</i>
Chris Lattner70de6632001-07-09 00:26:23 +0000[diff] [blame]3222</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3223</div>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3224
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3225<!-- ======================================================================= -->
3226<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
3227<div class="doc_text">
3228<p>The instructions in this category are the "miscellaneous"
3229instructions, which defy better classification.</p>
3230</div>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3231
3232<!-- _______________________________________________________________________ -->
3233<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
3234</div>
3235<div class="doc_text">
3236<h5>Syntax:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3237<pre> &lt;result&gt; = icmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt;
3238<i>; yields {i1}:result</i>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3239</pre>
3240<h5>Overview:</h5>
3241<p>The '<tt>icmp</tt>' instruction returns a boolean value based on comparison
3242of its two integer operands.</p>
3243<h5>Arguments:</h5>
3244<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
3245the condition code which indicates the kind of comparison to perform. It is not
3246a value, just a keyword. The possibilities for the condition code are:
3247<ol>
3248 <li><tt>eq</tt>: equal</li>
3249 <li><tt>ne</tt>: not equal </li>
3250 <li><tt>ugt</tt>: unsigned greater than</li>
3251 <li><tt>uge</tt>: unsigned greater or equal</li>
3252 <li><tt>ult</tt>: unsigned less than</li>
3253 <li><tt>ule</tt>: unsigned less or equal</li>
3254 <li><tt>sgt</tt>: signed greater than</li>
3255 <li><tt>sge</tt>: signed greater or equal</li>
3256 <li><tt>slt</tt>: signed less than</li>
3257 <li><tt>sle</tt>: signed less or equal</li>
3258</ol>
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3259<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Reid Spencer784ef792007-01-04 05:19:58 +0000[diff] [blame]3260<a href="#t_pointer">pointer</a> typed. They must also be identical types.</p>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3261<h5>Semantics:</h5>
3262<p>The '<tt>icmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3263the condition code given as <tt>cond</tt>. The comparison performed always
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3264yields a <a href="#t_primitive">i1</a> result, as follows:
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3265<ol>
3266 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
3267 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3268 </li>
3269 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
3270 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3271 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
3272 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3273 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
3274 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3275 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
3276 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3277 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
3278 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3279 <li><tt>sgt</tt>: interprets the operands as signed values and yields
3280 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3281 <li><tt>sge</tt>: interprets the operands as signed values and yields
3282 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3283 <li><tt>slt</tt>: interprets the operands as signed values and yields
3284 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3285 <li><tt>sle</tt>: interprets the operands as signed values and yields
3286 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3287</ol>
3288<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
3289values are treated as integers and then compared.</p>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3290
3291<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3292<pre> &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
3293 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
3294 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
3295 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
3296 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
3297 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3298</pre>
3299</div>
3300
3301<!-- _______________________________________________________________________ -->
3302<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
3303</div>
3304<div class="doc_text">
3305<h5>Syntax:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3306<pre> &lt;result&gt; = fcmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt;
3307<i>; yields {i1}:result</i>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3308</pre>
3309<h5>Overview:</h5>
3310<p>The '<tt>fcmp</tt>' instruction returns a boolean value based on comparison
3311of its floating point operands.</p>
3312<h5>Arguments:</h5>
3313<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
3314the condition code which indicates the kind of comparison to perform. It is not
3315a value, just a keyword. The possibilities for the condition code are:
3316<ol>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3317 <li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3318 <li><tt>oeq</tt>: ordered and equal</li>
3319 <li><tt>ogt</tt>: ordered and greater than </li>
3320 <li><tt>oge</tt>: ordered and greater than or equal</li>
3321 <li><tt>olt</tt>: ordered and less than </li>
3322 <li><tt>ole</tt>: ordered and less than or equal</li>
3323 <li><tt>one</tt>: ordered and not equal</li>
3324 <li><tt>ord</tt>: ordered (no nans)</li>
3325 <li><tt>ueq</tt>: unordered or equal</li>
3326 <li><tt>ugt</tt>: unordered or greater than </li>
3327 <li><tt>uge</tt>: unordered or greater than or equal</li>
3328 <li><tt>ult</tt>: unordered or less than </li>
3329 <li><tt>ule</tt>: unordered or less than or equal</li>
3330 <li><tt>une</tt>: unordered or not equal</li>
3331 <li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3332 <li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3333</ol>
Reid Spencer02e0d1d2006-12-06 07:08:07 +0000[diff] [blame]3334<p>In the preceding, <i>ordered</i> means that neither operand is a QNAN while
3335<i>unordered</i> means that either operand may be a QNAN.</p>
Reid Spencer784ef792007-01-04 05:19:58 +0000[diff] [blame]3336<p>The <tt>val1</tt> and <tt>val2</tt> arguments must be
3337<a href="#t_floating">floating point</a> typed. They must have identical
3338types.</p>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3339<p>In the foregoing, <i>ordered</i> means that neither operand is a QNAN and
3340<i>unordered</i> means that either operand is a QNAN.</p>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3341<h5>Semantics:</h5>
3342<p>The '<tt>fcmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3343the condition code given as <tt>cond</tt>. The comparison performed always
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3344yields a <a href="#t_primitive">i1</a> result, as follows:
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3345<ol>
3346 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3347 <li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3348 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3349 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3350 <tt>var1</tt> is greather than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3351 <li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3352 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3353 <li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3354 <tt>var1</tt> is less than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3355 <li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3356 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3357 <li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3358 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3359 <li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
3360 <li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3361 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3362 <li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3363 <tt>var1</tt> is greater than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3364 <li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3365 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3366 <li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3367 <tt>var1</tt> is less than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3368 <li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3369 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3370 <li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3371 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3372 <li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3373 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
3374</ol>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3375
3376<h5>Example:</h5>
3377<pre> &lt;result&gt; = fcmp oeq float 4.0, 5.0 <i>; yields: result=false</i>
3378 &lt;result&gt; = icmp one float 4.0, 5.0 <i>; yields: result=true</i>
3379 &lt;result&gt; = icmp olt float 4.0, 5.0 <i>; yields: result=true</i>
3380 &lt;result&gt; = icmp ueq double 1.0, 2.0 <i>; yields: result=false</i>
3381</pre>
3382</div>
3383
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3384<!-- _______________________________________________________________________ -->
3385<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
3386Instruction</a> </div>
3387<div class="doc_text">
3388<h5>Syntax:</h5>
3389<pre> &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...<br></pre>
3390<h5>Overview:</h5>
3391<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
3392the SSA graph representing the function.</p>
3393<h5>Arguments:</h5>
3394<p>The type of the incoming values are specified with the first type
3395field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
3396as arguments, with one pair for each predecessor basic block of the
3397current block. Only values of <a href="#t_firstclass">first class</a>
3398type may be used as the value arguments to the PHI node. Only labels
3399may be used as the label arguments.</p>
3400<p>There must be no non-phi instructions between the start of a basic
3401block and the PHI instructions: i.e. PHI instructions must be first in
3402a basic block.</p>
3403<h5>Semantics:</h5>
3404<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the
3405value specified by the parameter, depending on which basic block we
3406came from in the last <a href="#terminators">terminator</a> instruction.</p>
3407<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3408<pre>Loop: ; Infinite loop that counts from 0 on up...<br> %indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]<br> %nextindvar = add i32 %indvar, 1<br> br label %Loop<br></pre>
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3409</div>
3410
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3411<!-- _______________________________________________________________________ -->
3412<div class="doc_subsubsection">
3413 <a name="i_select">'<tt>select</tt>' Instruction</a>
3414</div>
3415
3416<div class="doc_text">
3417
3418<h5>Syntax:</h5>
3419
3420<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3421 &lt;result&gt; = select i1 &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3422</pre>
3423
3424<h5>Overview:</h5>
3425
3426<p>
3427The '<tt>select</tt>' instruction is used to choose one value based on a
3428condition, without branching.
3429</p>
3430
3431
3432<h5>Arguments:</h5>
3433
3434<p>
3435The '<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.
3436</p>
3437
3438<h5>Semantics:</h5>
3439
3440<p>
3441If the boolean condition evaluates to true, the instruction returns the first
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3442value argument; otherwise, it returns the second value argument.
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3443</p>
3444
3445<h5>Example:</h5>
3446
3447<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3448 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3449</pre>
3450</div>
3451
Robert Bocchinof72fdfe2006-01-15 20:48:27 +0000[diff] [blame]3452
3453<!-- _______________________________________________________________________ -->
3454<div class="doc_subsubsection">
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3455 <a name="i_call">'<tt>call</tt>' Instruction</a>
3456</div>
3457
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3458<div class="doc_text">
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3459
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3460<h5>Syntax:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3461<pre>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]3462 &lt;result&gt; = [tail] call [<a href="#callingconv">cconv</a>] &lt;ty&gt;* &lt;fnptrval&gt;(&lt;param list&gt;)
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3463</pre>
3464
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3465<h5>Overview:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3466
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3467<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3468
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3469<h5>Arguments:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3470
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3471<p>This instruction requires several arguments:</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3472
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3473<ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3474 <li>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]3475 <p>The optional "tail" marker indicates whether the callee function accesses
3476 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3477 function call is eligible for tail call optimization. Note that calls may
3478 be marked "tail" even if they do not occur before a <a
3479 href="#i_ret"><tt>ret</tt></a> instruction.
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3480 </li>
3481 <li>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3482 <p>The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]3483 convention</a> the call should use. If none is specified, the call defaults
3484 to using C calling conventions.
3485 </li>
3486 <li>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3487 <p>'<tt>ty</tt>': shall be the signature of the pointer to function value
3488 being invoked. The argument types must match the types implied by this
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3489 signature. This type can be omitted if the function is not varargs and
3490 if the function type does not return a pointer to a function.</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3491 </li>
3492 <li>
3493 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
3494 be invoked. In most cases, this is a direct function invocation, but
3495 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3496 to function value.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3497 </li>
3498 <li>
3499 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencerd845d162005-05-01 22:22:57 +0000[diff] [blame]3500 function signature argument types. All arguments must be of
3501 <a href="#t_firstclass">first class</a> type. If the function signature
3502 indicates the function accepts a variable number of arguments, the extra
3503 arguments can be specified.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3504 </li>
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3505</ol>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3506
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3507<h5>Semantics:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3508
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3509<p>The '<tt>call</tt>' instruction is used to cause control flow to
3510transfer to a specified function, with its incoming arguments bound to
3511the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
3512instruction in the called function, control flow continues with the
3513instruction after the function call, and the return value of the
3514function is bound to the result argument. This is a simpler case of
3515the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3516
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3517<h5>Example:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3518
3519<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3520 %retval = call i32 %test(i32 %argc)
3521 call i32(i8 *, ...) *%printf(i8 * %msg, i32 12, i8 42);
3522 %X = tail call i32 %foo()
3523 %Y = tail call <a href="#callingconv">fastcc</a> i32 %foo()
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3524</pre>
3525
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3526</div>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3527
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3528<!-- _______________________________________________________________________ -->
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3529<div class="doc_subsubsection">
Chris Lattner33337472006-01-13 23:26:01 +0000[diff] [blame]3530 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3531</div>
3532
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3533<div class="doc_text">
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3534
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3535<h5>Syntax:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3536
3537<pre>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3538 &lt;resultval&gt; = va_arg &lt;va_list*&gt; &lt;arglist&gt;, &lt;argty&gt;
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3539</pre>
3540
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3541<h5>Overview:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3542
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3543<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3544the "variable argument" area of a function call. It is used to implement the
3545<tt>va_arg</tt> macro in C.</p>
3546
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3547<h5>Arguments:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3548
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3549<p>This instruction takes a <tt>va_list*</tt> value and the type of
3550the argument. It returns a value of the specified argument type and
Jeff Cohendc6bfea2005-11-11 02:15:27 +0000[diff] [blame]3551increments the <tt>va_list</tt> to point to the next argument. Again, the
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3552actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3553
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3554<h5>Semantics:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3555
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3556<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
3557type from the specified <tt>va_list</tt> and causes the
3558<tt>va_list</tt> to point to the next argument. For more information,
3559see the variable argument handling <a href="#int_varargs">Intrinsic
3560Functions</a>.</p>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3561
3562<p>It is legal for this instruction to be called in a function which does not
3563take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3564function.</p>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3565
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3566<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3567href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3568argument.</p>
3569
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3570<h5>Example:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3571
3572<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
3573
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3574</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3575
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3576<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3577<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
3578<!-- *********************************************************************** -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3579
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3580<div class="doc_text">
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3581
3582<p>LLVM supports the notion of an "intrinsic function". These functions have
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]3583well known names and semantics and are required to follow certain restrictions.
3584Overall, these intrinsics represent an extension mechanism for the LLVM
3585language that does not require changing all of the transformations in LLVM to
3586add to the language (or the bytecode reader/writer, the parser,
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3587etc...).</p>
3588
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3589<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
3590prefix is reserved in LLVM for intrinsic names; thus, functions may not be named
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3591this. Intrinsic functions must always be external functions: you cannot define
3592the body of intrinsic functions. Intrinsic functions may only be used in call
3593or invoke instructions: it is illegal to take the address of an intrinsic
3594function. Additionally, because intrinsic functions are part of the LLVM
3595language, it is required that they all be documented here if any are added.</p>
3596
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]3597<p>Some intrinsic functions can be overloaded. That is, the intrinsic represents
3598a family of functions that perform the same operation but on different data
3599types. This is most frequent with the integer types. Since LLVM can represent
3600over 8 million different integer types, there is a way to declare an intrinsic
3601that can be overloaded based on its arguments. Such intrinsics will have the
3602names of the arbitrary types encoded into the intrinsic function name, each
3603preceded by a period. For example, the <tt>llvm.ctpop</tt> function can take an
3604integer of any width. This leads to a family of functions such as
3605<tt>i32 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i32 @llvm.ctpop.i29(i29 %val)</tt>.
3606</p>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3607
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]3608
3609<p>To learn how to add an intrinsic function, please see the
3610<a href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3611</p>
3612
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3613</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3614
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3615<!-- ======================================================================= -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3616<div class="doc_subsection">
3617 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
3618</div>
3619
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3620<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3621
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3622<p>Variable argument support is defined in LLVM with the <a
Chris Lattner33337472006-01-13 23:26:01 +0000[diff] [blame]3623 href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3624intrinsic functions. These functions are related to the similarly
3625named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3626
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3627<p>All of these functions operate on arguments that use a
3628target-specific value type "<tt>va_list</tt>". The LLVM assembly
3629language reference manual does not define what this type is, so all
3630transformations should be prepared to handle intrinsics with any type
3631used.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3632
Chris Lattner30b868d2006-05-15 17:26:46 +0000[diff] [blame]3633<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3634instruction and the variable argument handling intrinsic functions are
3635used.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3636
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3637<pre>
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3638define i32 @test(i32 %X, ...) {
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3639 ; Initialize variable argument processing
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3640 %ap = alloca i8 *
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3641 %ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3642 call void @llvm.va_start(i8* %ap2)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3643
3644 ; Read a single integer argument
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3645 %tmp = va_arg i8 ** %ap, i32
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3646
3647 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3648 %aq = alloca i8 *
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3649 %aq2 = bitcast i8** %aq to i8*
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3650 call void @llvm.va_copy(i8 *%aq2, i8* %ap2)
3651 call void @llvm.va_end(i8* %aq2)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3652
3653 ; Stop processing of arguments.
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3654 call void @llvm.va_end(i8* %ap2)
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3655 ret i32 %tmp
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3656}
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3657
3658declare void @llvm.va_start(i8*)
3659declare void @llvm.va_copy(i8*, i8*)
3660declare void @llvm.va_end(i8*)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3661</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3662</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3663
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3664<!-- _______________________________________________________________________ -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3665<div class="doc_subsubsection">
3666 <a name="i_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
3667</div>
3668
3669
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3670<div class="doc_text">
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3671<h5>Syntax:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3672<pre> declare void %llvm.va_start(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3673<h5>Overview:</h5>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3674<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
3675<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
3676href="#i_va_arg">va_arg</a></tt>.</p>
3677
3678<h5>Arguments:</h5>
3679
3680<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
3681
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3682<h5>Semantics:</h5>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3683
3684<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
3685macro available in C. In a target-dependent way, it initializes the
3686<tt>va_list</tt> element the argument points to, so that the next call to
3687<tt>va_arg</tt> will produce the first variable argument passed to the function.
3688Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
3689last argument of the function, the compiler can figure that out.</p>
3690
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3691</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3692
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3693<!-- _______________________________________________________________________ -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3694<div class="doc_subsubsection">
3695 <a name="i_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
3696</div>
3697
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3698<div class="doc_text">
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3699<h5>Syntax:</h5>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3700<pre> declare void @llvm.va_end(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3701<h5>Overview:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3702
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3703<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>&lt;arglist&gt;</tt>
3704which has been initialized previously with <tt><a href="#i_va_start">llvm.va_start</a></tt>
3705or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3706
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3707<h5>Arguments:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3708
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3709<p>The argument is a <tt>va_list</tt> to destroy.</p>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3710
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3711<h5>Semantics:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3712
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3713<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3714macro available in C. In a target-dependent way, it destroys the <tt>va_list</tt>.
3715Calls to <a href="#i_va_start"><tt>llvm.va_start</tt></a> and <a
3716 href="#i_va_copy"><tt>llvm.va_copy</tt></a> must be matched exactly
3717with calls to <tt>llvm.va_end</tt>.</p>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3718
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3719</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3720
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3721<!-- _______________________________________________________________________ -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3722<div class="doc_subsubsection">
3723 <a name="i_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
3724</div>
3725
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3726<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3727
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3728<h5>Syntax:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3729
3730<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3731 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3732</pre>
3733
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3734<h5>Overview:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3735
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3736<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position from
3737the source argument list to the destination argument list.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3738
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3739<h5>Arguments:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3740
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3741<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Andrew Lenharth5305ea52005-06-22 20:38:11 +0000[diff] [blame]3742The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3743
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3744
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3745<h5>Semantics:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3746
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3747<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt> macro
3748available in C. In a target-dependent way, it copies the source
3749<tt>va_list</tt> element into the destination list. This intrinsic is necessary
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3750because the <tt><a href="#i_va_start">llvm.va_start</a></tt> intrinsic may be
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3751arbitrarily complex and require memory allocation, for example.</p>
3752
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3753</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3754
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3755<!-- ======================================================================= -->
3756<div class="doc_subsection">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3757 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
3758</div>
3759
3760<div class="doc_text">
3761
3762<p>
3763LLVM support for <a href="GarbageCollection.html">Accurate Garbage
3764Collection</a> requires the implementation and generation of these intrinsics.
3765These intrinsics allow identification of <a href="#i_gcroot">GC roots on the
3766stack</a>, as well as garbage collector implementations that require <a
3767href="#i_gcread">read</a> and <a href="#i_gcwrite">write</a> barriers.
3768Front-ends for type-safe garbage collected languages should generate these
3769intrinsics to make use of the LLVM garbage collectors. For more details, see <a
3770href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
3771</p>
3772</div>
3773
3774<!-- _______________________________________________________________________ -->
3775<div class="doc_subsubsection">
3776 <a name="i_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
3777</div>
3778
3779<div class="doc_text">
3780
3781<h5>Syntax:</h5>
3782
3783<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3784 declare void @llvm.gcroot(&lt;ty&gt;** %ptrloc, &lt;ty2&gt;* %metadata)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3785</pre>
3786
3787<h5>Overview:</h5>
3788
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]3789<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3790the code generator, and allows some metadata to be associated with it.</p>
3791
3792<h5>Arguments:</h5>
3793
3794<p>The first argument specifies the address of a stack object that contains the
3795root pointer. The second pointer (which must be either a constant or a global
3796value address) contains the meta-data to be associated with the root.</p>
3797
3798<h5>Semantics:</h5>
3799
3800<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
3801location. At compile-time, the code generator generates information to allow
3802the runtime to find the pointer at GC safe points.
3803</p>
3804
3805</div>
3806
3807
3808<!-- _______________________________________________________________________ -->
3809<div class="doc_subsubsection">
3810 <a name="i_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
3811</div>
3812
3813<div class="doc_text">
3814
3815<h5>Syntax:</h5>
3816
3817<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3818 declare i8 * @llvm.gcread(i8 * %ObjPtr, i8 ** %Ptr)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3819</pre>
3820
3821<h5>Overview:</h5>
3822
3823<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
3824locations, allowing garbage collector implementations that require read
3825barriers.</p>
3826
3827<h5>Arguments:</h5>
3828
Chris Lattnerf9228072006-03-14 20:02:51 +0000[diff] [blame]3829<p>The second argument is the address to read from, which should be an address
3830allocated from the garbage collector. The first object is a pointer to the
3831start of the referenced object, if needed by the language runtime (otherwise
3832null).</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3833
3834<h5>Semantics:</h5>
3835
3836<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
3837instruction, but may be replaced with substantially more complex code by the
3838garbage collector runtime, as needed.</p>
3839
3840</div>
3841
3842
3843<!-- _______________________________________________________________________ -->
3844<div class="doc_subsubsection">
3845 <a name="i_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
3846</div>
3847
3848<div class="doc_text">
3849
3850<h5>Syntax:</h5>
3851
3852<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3853 declare void @llvm.gcwrite(i8 * %P1, i8 * %Obj, i8 ** %P2)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3854</pre>
3855
3856<h5>Overview:</h5>
3857
3858<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
3859locations, allowing garbage collector implementations that require write
3860barriers (such as generational or reference counting collectors).</p>
3861
3862<h5>Arguments:</h5>
3863
Chris Lattnerf9228072006-03-14 20:02:51 +0000[diff] [blame]3864<p>The first argument is the reference to store, the second is the start of the
3865object to store it to, and the third is the address of the field of Obj to
3866store to. If the runtime does not require a pointer to the object, Obj may be
3867null.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3868
3869<h5>Semantics:</h5>
3870
3871<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
3872instruction, but may be replaced with substantially more complex code by the
3873garbage collector runtime, as needed.</p>
3874
3875</div>
3876
3877
3878
3879<!-- ======================================================================= -->
3880<div class="doc_subsection">
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3881 <a name="int_codegen">Code Generator Intrinsics</a>
3882</div>
3883
3884<div class="doc_text">
3885<p>
3886These intrinsics are provided by LLVM to expose special features that may only
3887be implemented with code generator support.
3888</p>
3889
3890</div>
3891
3892<!-- _______________________________________________________________________ -->
3893<div class="doc_subsubsection">
3894 <a name="i_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
3895</div>
3896
3897<div class="doc_text">
3898
3899<h5>Syntax:</h5>
3900<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3901 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3902</pre>
3903
3904<h5>Overview:</h5>
3905
3906<p>
Chris Lattnerc1fb4262006-10-15 20:05:59 +0000[diff] [blame]3907The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
3908target-specific value indicating the return address of the current function
3909or one of its callers.
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3910</p>
3911
3912<h5>Arguments:</h5>
3913
3914<p>
3915The argument to this intrinsic indicates which function to return the address
3916for. Zero indicates the calling function, one indicates its caller, etc. The
3917argument is <b>required</b> to be a constant integer value.
3918</p>
3919
3920<h5>Semantics:</h5>
3921
3922<p>
3923The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
3924the return address of the specified call frame, or zero if it cannot be
3925identified. The value returned by this intrinsic is likely to be incorrect or 0
3926for arguments other than zero, so it should only be used for debugging purposes.
3927</p>
3928
3929<p>
3930Note that calling this intrinsic does not prevent function inlining or other
Chris Lattner2e6eb5f2005-03-07 20:30:51 +0000[diff] [blame]3931aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3932source-language caller.
3933</p>
3934</div>
3935
3936
3937<!-- _______________________________________________________________________ -->
3938<div class="doc_subsubsection">
3939 <a name="i_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
3940</div>
3941
3942<div class="doc_text">
3943
3944<h5>Syntax:</h5>
3945<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3946 declare i8 *@llvm.frameaddress(i32 &lt;level&gt;)
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3947</pre>
3948
3949<h5>Overview:</h5>
3950
3951<p>
Chris Lattnerc1fb4262006-10-15 20:05:59 +0000[diff] [blame]3952The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
3953target-specific frame pointer value for the specified stack frame.
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3954</p>
3955
3956<h5>Arguments:</h5>
3957
3958<p>
3959The argument to this intrinsic indicates which function to return the frame
3960pointer for. Zero indicates the calling function, one indicates its caller,
3961etc. The argument is <b>required</b> to be a constant integer value.
3962</p>
3963
3964<h5>Semantics:</h5>
3965
3966<p>
3967The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
3968the frame address of the specified call frame, or zero if it cannot be
3969identified. The value returned by this intrinsic is likely to be incorrect or 0
3970for arguments other than zero, so it should only be used for debugging purposes.
3971</p>
3972
3973<p>
3974Note that calling this intrinsic does not prevent function inlining or other
Chris Lattner2e6eb5f2005-03-07 20:30:51 +0000[diff] [blame]3975aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]3976source-language caller.
3977</p>
3978</div>
3979
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]3980<!-- _______________________________________________________________________ -->
3981<div class="doc_subsubsection">
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]3982 <a name="i_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
3983</div>
3984
3985<div class="doc_text">
3986
3987<h5>Syntax:</h5>
3988<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3989 declare i8 *@llvm.stacksave()
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]3990</pre>
3991
3992<h5>Overview:</h5>
3993
3994<p>
3995The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
3996the function stack, for use with <a href="#i_stackrestore">
3997<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
3998features like scoped automatic variable sized arrays in C99.
3999</p>
4000
4001<h5>Semantics:</h5>
4002
4003<p>
4004This intrinsic returns a opaque pointer value that can be passed to <a
4005href="#i_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
4006<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
4007<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
4008state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
4009practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
4010that were allocated after the <tt>llvm.stacksave</tt> was executed.
4011</p>
4012
4013</div>
4014
4015<!-- _______________________________________________________________________ -->
4016<div class="doc_subsubsection">
4017 <a name="i_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
4018</div>
4019
4020<div class="doc_text">
4021
4022<h5>Syntax:</h5>
4023<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4024 declare void @llvm.stackrestore(i8 * %ptr)
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4025</pre>
4026
4027<h5>Overview:</h5>
4028
4029<p>
4030The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
4031the function stack to the state it was in when the corresponding <a
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]4032href="#i_stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4033useful for implementing language features like scoped automatic variable sized
4034arrays in C99.
4035</p>
4036
4037<h5>Semantics:</h5>
4038
4039<p>
4040See the description for <a href="#i_stacksave"><tt>llvm.stacksave</tt></a>.
4041</p>
4042
4043</div>
4044
4045
4046<!-- _______________________________________________________________________ -->
4047<div class="doc_subsubsection">
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4048 <a name="i_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
4049</div>
4050
4051<div class="doc_text">
4052
4053<h5>Syntax:</h5>
4054<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4055 declare void @llvm.prefetch(i8 * &lt;address&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4056 i32 &lt;rw&gt;, i32 &lt;locality&gt;)
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4057</pre>
4058
4059<h5>Overview:</h5>
4060
4061
4062<p>
4063The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]4064a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
4065no
4066effect on the behavior of the program but can change its performance
Chris Lattnerff851072005-02-28 19:47:14 +0000[diff] [blame]4067characteristics.
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4068</p>
4069
4070<h5>Arguments:</h5>
4071
4072<p>
4073<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
4074determining if the fetch should be for a read (0) or write (1), and
4075<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattnerd3e641c2005-03-07 20:31:38 +0000[diff] [blame]4076locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4077<tt>locality</tt> arguments must be constant integers.
4078</p>
4079
4080<h5>Semantics:</h5>
4081
4082<p>
4083This intrinsic does not modify the behavior of the program. In particular,
4084prefetches cannot trap and do not produce a value. On targets that support this
4085intrinsic, the prefetch can provide hints to the processor cache for better
4086performance.
4087</p>
4088
4089</div>
4090
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4091<!-- _______________________________________________________________________ -->
4092<div class="doc_subsubsection">
4093 <a name="i_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
4094</div>
4095
4096<div class="doc_text">
4097
4098<h5>Syntax:</h5>
4099<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4100 declare void @llvm.pcmarker( i32 &lt;id&gt; )
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4101</pre>
4102
4103<h5>Overview:</h5>
4104
4105
4106<p>
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]4107The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
4108(PC) in a region of
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4109code to simulators and other tools. The method is target specific, but it is
4110expected that the marker will use exported symbols to transmit the PC of the marker.
Jeff Cohendc6bfea2005-11-11 02:15:27 +0000[diff] [blame]4111The marker makes no guarantees that it will remain with any specific instruction
Chris Lattnere64d41d2005-11-15 06:07:55 +0000[diff] [blame]4112after optimizations. It is possible that the presence of a marker will inhibit
Chris Lattnerb40261e2006-03-24 07:16:10 +0000[diff] [blame]4113optimizations. The intended use is to be inserted after optimizations to allow
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]4114correlations of simulation runs.
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4115</p>
4116
4117<h5>Arguments:</h5>
4118
4119<p>
4120<tt>id</tt> is a numerical id identifying the marker.
4121</p>
4122
4123<h5>Semantics:</h5>
4124
4125<p>
4126This intrinsic does not modify the behavior of the program. Backends that do not
4127support this intrinisic may ignore it.
4128</p>
4129
4130</div>
4131
Andrew Lenharth01aa5632005-11-11 16:47:30 +0000[diff] [blame]4132<!-- _______________________________________________________________________ -->
4133<div class="doc_subsubsection">
4134 <a name="i_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
4135</div>
4136
4137<div class="doc_text">
4138
4139<h5>Syntax:</h5>
4140<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4141 declare i64 @llvm.readcyclecounter( )
Andrew Lenharth01aa5632005-11-11 16:47:30 +0000[diff] [blame]4142</pre>
4143
4144<h5>Overview:</h5>
4145
4146
4147<p>
4148The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
4149counter register (or similar low latency, high accuracy clocks) on those targets
4150that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
4151As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
4152should only be used for small timings.
4153</p>
4154
4155<h5>Semantics:</h5>
4156
4157<p>
4158When directly supported, reading the cycle counter should not modify any memory.
4159Implementations are allowed to either return a application specific value or a
4160system wide value. On backends without support, this is lowered to a constant 0.
4161</p>
4162
4163</div>
4164
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4165<!-- ======================================================================= -->
4166<div class="doc_subsection">
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4167 <a name="int_libc">Standard C Library Intrinsics</a>
4168</div>
4169
4170<div class="doc_text">
4171<p>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4172LLVM provides intrinsics for a few important standard C library functions.
4173These intrinsics allow source-language front-ends to pass information about the
4174alignment of the pointer arguments to the code generator, providing opportunity
4175for more efficient code generation.
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4176</p>
4177
4178</div>
4179
4180<!-- _______________________________________________________________________ -->
4181<div class="doc_subsubsection">
4182 <a name="i_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
4183</div>
4184
4185<div class="doc_text">
4186
4187<h5>Syntax:</h5>
4188<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4189 declare void @llvm.memcpy.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4190 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4191 declare void @llvm.memcpy.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4192 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4193</pre>
4194
4195<h5>Overview:</h5>
4196
4197<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4198The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4199location to the destination location.
4200</p>
4201
4202<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4203Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
4204intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4205</p>
4206
4207<h5>Arguments:</h5>
4208
4209<p>
4210The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4211the source. The third argument is an integer argument
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4212specifying the number of bytes to copy, and the fourth argument is the alignment
4213of the source and destination locations.
4214</p>
4215
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4216<p>
4217If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner5316e5d2006-03-04 00:02:10 +0000[diff] [blame]4218the caller guarantees that both the source and destination pointers are aligned
4219to that boundary.
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4220</p>
4221
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4222<h5>Semantics:</h5>
4223
4224<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4225The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4226location to the destination location, which are not allowed to overlap. It
4227copies "len" bytes of memory over. If the argument is known to be aligned to
4228some boundary, this can be specified as the fourth argument, otherwise it should
4229be set to 0 or 1.
4230</p>
4231</div>
4232
4233
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4234<!-- _______________________________________________________________________ -->
4235<div class="doc_subsubsection">
4236 <a name="i_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
4237</div>
4238
4239<div class="doc_text">
4240
4241<h5>Syntax:</h5>
4242<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4243 declare void @llvm.memmove.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4244 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4245 declare void @llvm.memmove.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4246 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4247</pre>
4248
4249<h5>Overview:</h5>
4250
4251<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4252The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the source
4253location to the destination location. It is similar to the
4254'<tt>llvm.memcmp</tt>' intrinsic but allows the two memory locations to overlap.
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4255</p>
4256
4257<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4258Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
4259intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4260</p>
4261
4262<h5>Arguments:</h5>
4263
4264<p>
4265The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4266the source. The third argument is an integer argument
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4267specifying the number of bytes to copy, and the fourth argument is the alignment
4268of the source and destination locations.
4269</p>
4270
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4271<p>
4272If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner5316e5d2006-03-04 00:02:10 +0000[diff] [blame]4273the caller guarantees that the source and destination pointers are aligned to
4274that boundary.
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4275</p>
4276
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4277<h5>Semantics:</h5>
4278
4279<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4280The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the source
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4281location to the destination location, which may overlap. It
4282copies "len" bytes of memory over. If the argument is known to be aligned to
4283some boundary, this can be specified as the fourth argument, otherwise it should
4284be set to 0 or 1.
4285</p>
4286</div>
4287
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]4288
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4289<!-- _______________________________________________________________________ -->
4290<div class="doc_subsubsection">
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4291 <a name="i_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4292</div>
4293
4294<div class="doc_text">
4295
4296<h5>Syntax:</h5>
4297<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4298 declare void @llvm.memset.i32(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4299 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4300 declare void @llvm.memset.i64(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4301 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4302</pre>
4303
4304<h5>Overview:</h5>
4305
4306<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4307The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a particular
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4308byte value.
4309</p>
4310
4311<p>
4312Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
4313does not return a value, and takes an extra alignment argument.
4314</p>
4315
4316<h5>Arguments:</h5>
4317
4318<p>
4319The first argument is a pointer to the destination to fill, the second is the
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4320byte value to fill it with, the third argument is an integer
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4321argument specifying the number of bytes to fill, and the fourth argument is the
4322known alignment of destination location.
4323</p>
4324
4325<p>
4326If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner5316e5d2006-03-04 00:02:10 +0000[diff] [blame]4327the caller guarantees that the destination pointer is aligned to that boundary.
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4328</p>
4329
4330<h5>Semantics:</h5>
4331
4332<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4333The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting at
4334the
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4335destination location. If the argument is known to be aligned to some boundary,
4336this can be specified as the fourth argument, otherwise it should be set to 0 or
43371.
4338</p>
4339</div>
4340
4341
Chris Lattner3b4f4372004-06-11 02:28:03 +0000[diff] [blame]4342<!-- _______________________________________________________________________ -->
4343<div class="doc_subsubsection">
Chris Lattner069b5bd2006-01-16 22:38:59 +0000[diff] [blame]4344 <a name="i_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4345</div>
4346
4347<div class="doc_text">
4348
4349<h5>Syntax:</h5>
4350<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4351 declare float @llvm.sqrt.f32(float %Val)
4352 declare double @llvm.sqrt.f64(double %Val)
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4353</pre>
4354
4355<h5>Overview:</h5>
4356
4357<p>
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4358The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4359returning the same value as the libm '<tt>sqrt</tt>' function would. Unlike
4360<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
4361negative numbers (which allows for better optimization).
4362</p>
4363
4364<h5>Arguments:</h5>
4365
4366<p>
4367The argument and return value are floating point numbers of the same type.
4368</p>
4369
4370<h5>Semantics:</h5>
4371
4372<p>
4373This function returns the sqrt of the specified operand if it is a positive
4374floating point number.
4375</p>
4376</div>
4377
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4378<!-- _______________________________________________________________________ -->
4379<div class="doc_subsubsection">
4380 <a name="i_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
4381</div>
4382
4383<div class="doc_text">
4384
4385<h5>Syntax:</h5>
4386<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4387 declare float @llvm.powi.f32(float %Val, i32 %power)
4388 declare double @llvm.powi.f64(double %Val, i32 %power)
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4389</pre>
4390
4391<h5>Overview:</h5>
4392
4393<p>
4394The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
4395specified (positive or negative) power. The order of evaluation of
4396multiplications is not defined.
4397</p>
4398
4399<h5>Arguments:</h5>
4400
4401<p>
4402The second argument is an integer power, and the first is a value to raise to
4403that power.
4404</p>
4405
4406<h5>Semantics:</h5>
4407
4408<p>
4409This function returns the first value raised to the second power with an
4410unspecified sequence of rounding operations.</p>
4411</div>
4412
4413
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4414<!-- ======================================================================= -->
4415<div class="doc_subsection">
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4416 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4417</div>
4418
4419<div class="doc_text">
4420<p>
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4421LLVM provides intrinsics for a few important bit manipulation operations.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4422These allow efficient code generation for some algorithms.
4423</p>
4424
4425</div>
4426
4427<!-- _______________________________________________________________________ -->
4428<div class="doc_subsubsection">
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4429 <a name="i_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
4430</div>
4431
4432<div class="doc_text">
4433
4434<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4435<p>This is an overloaded intrinsic function. You can use bswap on any integer
4436type that is an even number of bytes (i.e. BitWidth % 16 == 0). Note the suffix
4437that includes the type for the result and the operand.
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4438<pre>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4439 declare i16 @llvm.bswap.i16.i16(i16 &lt;id&gt;)
4440 declare i32 @llvm.bswap.i32.i32(i32 &lt;id&gt;)
4441 declare i64 @llvm.bswap.i64.i32(i64 &lt;id&gt;)
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4442</pre>
4443
4444<h5>Overview:</h5>
4445
4446<p>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4447The '<tt>llvm.bwsap</tt>' family of intrinsics is used to byteswap integer
4448values with an even number of bytes (positive multiple of 16 bits). These are
4449useful for performing operations on data that is not in the target's native
4450byte order.
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4451</p>
4452
4453<h5>Semantics:</h5>
4454
4455<p>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4456The <tt>llvm.bswap.16.i16</tt> intrinsic returns an i16 value that has the high
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4457and low byte of the input i16 swapped. Similarly, the <tt>llvm.bswap.i32</tt>
4458intrinsic returns an i32 value that has the four bytes of the input i32
4459swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the returned
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4460i32 will have its bytes in 3, 2, 1, 0 order. The <tt>llvm.bswap.i48.i48</tt>,
4461<tt>llvm.bswap.i64.i64</tt> and other intrinsics extend this concept to
4462additional even-byte lengths (6 bytes, 8 bytes and more, respectively).
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4463</p>
4464
4465</div>
4466
4467<!-- _______________________________________________________________________ -->
4468<div class="doc_subsubsection">
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4469 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4470</div>
4471
4472<div class="doc_text">
4473
4474<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4475<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
4476width. Not all targets support all bit widths however.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4477<pre>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4478 declare i32 @llvm.ctpop.i8 (i8 &lt;src&gt;)
4479 declare i32 @llvm.ctpop.i16(i16 &lt;src&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4480 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4481 declare i32 @llvm.ctpop.i64(i64 &lt;src&gt;)
4482 declare i32 @llvm.ctpop.i256(i256 &lt;src&gt;)
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4483</pre>
4484
4485<h5>Overview:</h5>
4486
4487<p>
Chris Lattner069b5bd2006-01-16 22:38:59 +0000[diff] [blame]4488The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set in a
4489value.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4490</p>
4491
4492<h5>Arguments:</h5>
4493
4494<p>
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]4495The only argument is the value to be counted. The argument may be of any
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]4496integer type. The return type must match the argument type.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4497</p>
4498
4499<h5>Semantics:</h5>
4500
4501<p>
4502The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
4503</p>
4504</div>
4505
4506<!-- _______________________________________________________________________ -->
4507<div class="doc_subsubsection">
Chris Lattnerb748c672006-01-16 22:34:14 +0000[diff] [blame]4508 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4509</div>
4510
4511<div class="doc_text">
4512
4513<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4514<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
4515integer bit width. Not all targets support all bit widths however.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4516<pre>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4517 declare i32 @llvm.ctlz.i8 (i8 &lt;src&gt;)
4518 declare i32 @llvm.ctlz.i16(i16 &lt;src&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4519 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4520 declare i32 @llvm.ctlz.i64(i64 &lt;src&gt;)
4521 declare i32 @llvm.ctlz.i256(i256 &lt;src&gt;)
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4522</pre>
4523
4524<h5>Overview:</h5>
4525
4526<p>
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4527The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
4528leading zeros in a variable.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4529</p>
4530
4531<h5>Arguments:</h5>
4532
4533<p>
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]4534The only argument is the value to be counted. The argument may be of any
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]4535integer type. The return type must match the argument type.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4536</p>
4537
4538<h5>Semantics:</h5>
4539
4540<p>
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4541The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
4542in a variable. If the src == 0 then the result is the size in bits of the type
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4543of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4544</p>
4545</div>
Chris Lattner3b4f4372004-06-11 02:28:03 +0000[diff] [blame]4546
4547
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4548
4549<!-- _______________________________________________________________________ -->
4550<div class="doc_subsubsection">
Chris Lattnerb748c672006-01-16 22:34:14 +0000[diff] [blame]4551 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4552</div>
4553
4554<div class="doc_text">
4555
4556<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4557<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
4558integer bit width. Not all targets support all bit widths however.
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4559<pre>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4560 declare i32 @llvm.cttz.i8 (i8 &lt;src&gt;)
4561 declare i32 @llvm.cttz.i16(i16 &lt;src&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4562 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4563 declare i32 @llvm.cttz.i64(i64 &lt;src&gt;)
4564 declare i32 @llvm.cttz.i256(i256 &lt;src&gt;)
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4565</pre>
4566
4567<h5>Overview:</h5>
4568
4569<p>
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4570The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
4571trailing zeros.
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4572</p>
4573
4574<h5>Arguments:</h5>
4575
4576<p>
4577The only argument is the value to be counted. The argument may be of any
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]4578integer type. The return type must match the argument type.
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4579</p>
4580
4581<h5>Semantics:</h5>
4582
4583<p>
4584The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
4585in a variable. If the src == 0 then the result is the size in bits of the type
4586of src. For example, <tt>llvm.cttz(2) = 1</tt>.
4587</p>
4588</div>
4589
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4590<!-- _______________________________________________________________________ -->
4591<div class="doc_subsubsection">
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame^]4592 <a name="int_bit_and_reduce">'<tt>llvm.bit.and.reduce.*</tt>' Intrinsic</a>
4593</div>
4594
4595<div class="doc_text">
4596<h5>Syntax:</h5>
4597<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.and.reduce</tt> on
4598any integer bit width.
4599<pre>
4600 declare i1 @llvm.bit.and.reduce.i32(i32 %val)
4601 declare i1 @llvm.bit.and.reduce.i97(i97 %val)
4602</pre>
4603<h5>Overview:</h5>
4604<p>The '<tt>llvm.bit.and.reduce</tt>' family of intrinsic functions applies the
4605AND operator bitwise to each bit in <tt>%val</tt> until it yields the result.
4606</p>
4607
4608<h5>Arguments:</h5>
4609<p>The argument may be any bit width. The result is always a 1-bit integer.</p>
4610
4611<h5>Semantics:</h5>
4612<p>The '<tt>llvm.bit.and.reduce</tt>' intrinsic is the equivalent of a test
4613against <tt>-1</tt>. Only if all bits in <tt>%val</tt> are set will the result
4614be 1, otherwise 0.</p>
4615</div>
4616
4617<!-- _______________________________________________________________________ -->
4618<div class="doc_subsubsection">
4619 <a name="int_bit_or_reduce">'<tt>llvm.bit.or.reduce.*</tt>' Intrinsic</a>
4620</div>
4621
4622<div class="doc_text">
4623<h5>Syntax:</h5>
4624<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.or.reduce</tt> on
4625any integer bit width.
4626<pre>
4627 declare i1 @llvm.bit.or.reduce.i32(i32 %val)
4628 declare i1 @llvm.bit.or.reduce.i97(i97 %val)
4629</pre>
4630<h5>Overview:</h5>
4631<p>The '<tt>llvm.bit.or.reduce</tt>' family of intrinsic functions applies the
4632OR operator bitwise to each bit in <tt>%val</tt> until it yields the result.
4633</p>
4634
4635<h5>Arguments:</h5>
4636<p>The argument may be any bit width. The result is always a 1-bit integer.</p>
4637
4638<h5>Semantics:</h5>
4639<p>The '<tt>llvm.bit.or.reduce</tt>' intrinsic is the equivalent of a test
4640against <tt>0</tt>. Only if all bits in <tt>%val</tt> are clear will the result
4641be 0, otherwise 1.</p>
4642</div>
4643
4644<!-- _______________________________________________________________________ -->
4645<div class="doc_subsubsection">
4646 <a name="int_bit_xor_reduce">'<tt>llvm.bit.xor.reduce.*</tt>' Intrinsic</a>
4647</div>
4648
4649<div class="doc_text">
4650<h5>Syntax:</h5>
4651<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.xor.reduce</tt> on
4652any integer bit width.
4653<pre>
4654 declare i1 @llvm.bit.xor.reduce.i32(i32 %val)
4655 declare i1 @llvm.bit.xor.reduce.i97(i97 %val)
4656</pre>
4657<h5>Overview:</h5>
4658<p>The '<tt>llvm.bit.xor.reduce</tt>' family of intrinsic functions applies the
4659XOR operator bitwise to each bit in <tt>%val</tt> until it yields the result.
4660</p>
4661
4662<h5>Arguments:</h5>
4663<p>The argument may be any bit width. The result is always a 1-bit integer.</p>
4664
4665<h5>Semantics:</h5>
4666<p>The '<tt>llvm.bit.xor.reduce</tt>' computes its result by performing an XOR
4667operation on the two lowest order bits in <tt>%val</tt>. That result is then
4668XOR'd with the next bit in <tt>%val</tt> and this process continues until all
4669bits in <tt>%val</tt> have been XOR'd with the result of the previous XORs. The
4670resulting bit is returned.</p>
4671</div>
4672
4673<!-- _______________________________________________________________________ -->
4674<div class="doc_subsubsection">
4675 <a name="int_bit_nand_reduce">'<tt>llvm.bit.nand.reduce.*</tt>' Intrinsic</a>
4676</div>
4677
4678<div class="doc_text">
4679<h5>Syntax:</h5>
4680<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.nand.reduce</tt> on
4681any integer bit width.
4682<pre>
4683 declare i1 @llvm.bit.nand.reduce.i32(i32 %val)
4684 declare i1 @llvm.bit.nand.reduce.i97(i97 %val)
4685</pre>
4686<h5>Overview:</h5>
4687<p>The '<tt>llvm.bit.nand.reduce</tt>' family of intrinsic functions applies the
4688NAND operator bitwise to each bit in <tt>%val</tt> until it yields the result.
4689</p>
4690
4691<h5>Arguments:</h5>
4692<p>The argument may be any bit width. The result is always a 1-bit integer.</p>
4693
4694<h5>Semantics:</h5>
4695<p>The '<tt>llvm.bit.nand.reduce</tt>' intrinsic is the equivalent of taking the
4696complement of the <tt>llvm.bit.and.reduce</tt> intrinsic. That is, it returns 0
4697if <tt>%val</tt> is all ones (-1) and 1 otherwise.</p>
4698</div>
4699
4700<!-- _______________________________________________________________________ -->
4701<div class="doc_subsubsection">
4702 <a name="int_bit_nor_reduce">'<tt>llvm.bit.nor.reduce.*</tt>' Intrinsic</a>
4703</div>
4704
4705<div class="doc_text">
4706<h5>Syntax:</h5>
4707<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.nor.reduce</tt> on
4708any integer bit width.
4709<pre>
4710 declare i1 @llvm.bit.nor.reduce.i32(i32 %val)
4711 declare i1 @llvm.bit.nor.reduce.i97(i97 %val)
4712</pre>
4713<h5>Overview:</h5>
4714<p>The '<tt>llvm.bit.nor.reduce</tt>' family of intrinsic functions applies the
4715NOR operator bitwise to each bit in <tt>%val</tt> until it yields the result.
4716</p>
4717
4718<h5>Arguments:</h5>
4719<p>The argument may be any bit width. The result is always a 1-bit integer.</p>
4720
4721<h5>Semantics:</h5>
4722<p>The '<tt>llvm.bit.nor.reduce</tt>' intrinsic is equivalent to the complement
4723of the <tt>llvm.bit.or.reduce</tt> intrinsic. That is, it returns 1 if all bits
4724in <tt>%val</tt> are 0, and 1 otherwise.</p>
4725</div>
4726
4727<!-- _______________________________________________________________________ -->
4728<div class="doc_subsubsection">
4729 <a name="int_bit_nxor_reduce">'<tt>llvm.bit.nxor.reduce.*</tt>' Intrinsic</a>
4730</div>
4731
4732<div class="doc_text">
4733<h5>Syntax:</h5>
4734<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.nxor.reduce</tt> on
4735any integer bit width.
4736<pre>
4737 declare i1 @llvm.bit.nxor.reduce.i32(i32 %val)
4738 declare i1 @llvm.bit.nxor.reduce.i97(i97 %val)
4739</pre>
4740<h5>Overview:</h5>
4741<p>The '<tt>llvm.bit.nxor.reduce</tt>' family of intrinsic functions applies the
4742AND operator bitwise to each bit in <tt>%val</tt> until it yields the result.
4743</p>
4744
4745<h5>Arguments:</h5>
4746<p>The argument may be any bit width. The result is always a 1-bit integer.</p>
4747
4748<h5>Semantics:</h5>
4749<p>The '<tt>llvm.bit.nxor.reduce</tt>' intrinsic is the equivalent of the
4750complement of the <tt>llvm.bit.xor.reduce</tt> intrinsic.</p>
4751</div>
4752
4753<!-- _______________________________________________________________________ -->
4754<div class="doc_subsubsection">
4755 <a name="int_bit_select">'<tt>llvm.bit.select.*</tt>' Intrinsic</a>
4756</div>
4757
4758<div class="doc_text">
4759<h5>Syntax:</h5>
4760<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.select</tt> on any
4761integer bit width.
4762<pre>
4763 declare i1 @llvm.bit.select.i17 (i17 %val, i32 %bit)
4764 declare i1 @llvm.bit.select.i29 (i29 %val, i32 %bit)
4765</pre>
4766
4767<h5>Overview:</h5>
4768<p>The '<tt>llvm.bit.select</tt>' family of intrinsic functions selects a
4769specific bit from an integer value and returns it.</p>
4770
4771<h5>Arguments:</h5>
4772<p>The two arguments may be any bit width. The result is always a 1-bit
4773integer. The first argument, <tt>%val</tt> may be any bit width and is the
4774value from which the bit is selected. The second argument, <tt>%bit</tt> must
4775be an <tt>i32</tt> and is the bit index of the bit to be selected. Bits are
4776numbered starting with 0 as the lowest ordered bit.</p>
4777
4778<h5>Semantics:</h5>
4779<p>The '<tt>llvm.bit.select</tt>' intrinsic is the equivalent of shift and a
4780truncate operation. The <tt>%val</tt> is shifted right by <tt>%bit</tt> bits and
4781then truncated to a 1-bit integer.</p>
4782</div>
4783
4784<!-- _______________________________________________________________________ -->
4785<div class="doc_subsubsection">
4786 <a name="int_bit_set">'<tt>llvm.bit.set.*</tt>' Intrinsic</a>
4787</div>
4788
4789<div class="doc_text">
4790<h5>Syntax:</h5>
4791<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.set</tt> on any
4792integer bit width.
4793<pre>
4794 declare i17 @llvm.bit.set.i17.i17 (i17 %val, i32 %bit)
4795 declare i52 @llvm.bit.set.i52.i52 (i52 %val, i32 %bit)
4796</pre>
4797
4798<h5>Overview:</h5>
4799<p>
4800The '<tt>llvm.bit.set</tt>' family of intrinsic functions sets a specific bit in
4801a <tt>%val</tt> and returns the result.</p>
4802
4803<h5>Arguments:</h5>
4804<p>The result and the first argument, <tt>%val</tt>, may be an integer of any
4805bit width, but they must be the same bit width. The second argument must be an
4806<tt>i32</tt>.</p>
4807
4808<h5>Semantics:</h5>
4809<p>The '<tt>llvm.bit.set</tt>' intrinsic is the equivalent of creating a bit
4810mask for the <tt>%bit</tt> requested in the width of <tt>%val</tt>, ORing that
4811mask with <tt>%val</tt> and returning the result.</p>
4812</div>
4813
4814<!-- _______________________________________________________________________ -->
4815<div class="doc_subsubsection">
4816 <a name="int_bit_clear">'<tt>llvm.bit.clear.*</tt>' Intrinsic</a>
4817</div>
4818
4819<div class="doc_text">
4820
4821<h5>Syntax:</h5>
4822<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.clear</tt> on any
4823integer bit width.
4824<pre>
4825 declare i17 @llvm.bit.clear.i17.i17 (i17 %val, i32 %bit)
4826 declare i29 @llvm.bit.clear.i29.i29 (i29 %val, i32 %bit)
4827</pre>
4828
4829<h5>Overview:</h5>
4830<p>The '<tt>llvm.bit.clear</tt>' family of intrinsic functions clears a specific
4831bit in a value and returns the result.</p>
4832
4833<h5>Arguments:</h5>
4834<p>The result and the first argument, <tt>%val</tt>, may be an integer of any
4835bit width, but they must be the same bit width. The second argument must be an
4836<tt>i32</tt>.</p>
4837
4838<h5>Semantics:</h5>
4839<p>The '<tt>llvm.bit.clear</tt>' intrinsic is the equivalent of making a bit
4840mask in the width of <tt>%val</tt> but with the bit at index <tt>%bit</tt> set
4841to zero, ANDing that mask with <tt>%val</tt> and returning the result.</p>
4842</div>
4843
4844<!-- _______________________________________________________________________ -->
4845<div class="doc_subsubsection">
4846 <a name="int_bit_concat">'<tt>llvm.bit.concat.*</tt>' Intrinsic</a>
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4847</div>
4848
4849<div class="doc_text">
4850
4851<h5>Syntax:</h5>
4852<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.concat</tt> on any
4853integer bit width.
4854<pre>
4855 declare i32 @llvm.bit.concat.i32.i17.i15 (i17 %hi, i15 %lo)
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame^]4856 declare i29 @llvm.bit.concat.i29.i12.i15 (i12 %hi, i15 %lo)
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4857</pre>
4858
4859<h5>Overview:</h5>
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame^]4860<p>The '<tt>llvm.bit.concat</tt>' family of intrinsic functions concatenates two
4861integer values to produce a longer one.</p>
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4862
4863<h5>Arguments:</h5>
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame^]4864<p>The two arguments may be any bit width. The result must be an integer type
4865whose bit width is the sum of the arguments' bit widths. The first argument,
4866<tt>%hi</tt>, represents the bits that will occupy the high order bit locations
4867in the concatenated result. The second argument, <tt>%lo</tt>, will occupy the
4868lower order bit locations in the result.</p>
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4869
4870<h5>Semantics:</h5>
4871
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame^]4872<p>The '<tt>llvm.bit.concat</tt>' intrinsic is the equivalent of two
4873<tt>zext</tt> instructions, a <tt>shl</tt> and an <tt>or</tt>. The operation
4874proceeds as follows:</p>
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4875<ol>
4876 <li>Each of the arguments is <tt>zext</tt>'d to the result bit width.</li>
4877 <li>The <tt>%hi</tt> argument is shift left by the width of the <tt>%lo</tt>
4878 argument (shifted into to high order bits).</li>
4879 <li>The shifted <tt>%hi</tt> value and <tt>%lo</tt> are <tt>or</tt>'d together
4880 to form the result.</li>
4881</ol>
4882</div>
4883
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame^]4884<!-- _______________________________________________________________________ -->
4885<div class="doc_subsubsection">
4886 <a name="int_bit_part_select">'<tt>llvm.bit.part.select.*</tt>' Intrinsic</a>
4887</div>
4888
4889<div class="doc_text">
4890
4891<h5>Syntax:</h5>
4892<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.part.select</tt>
4893on any integer bit width.
4894<pre>
4895 declare i17 @llvm.bit.part.select.i17.i17 (i17 %val, i32 %loBit, i32 %hiBit)
4896 declare i29 @llvm.bit.part.select.i29.i29 (i29 %val, i32 %loBit, i32 %hiBit)
4897</pre>
4898
4899<h5>Overview:</h5>
4900<p>The '<tt>llvm.bit.part.select</tt>' family of intrinsic functions selects a
4901range of bits from an integer value and returns them in the same bit width as
4902the original value.</p>
4903
4904<h5>Arguments:</h5>
4905<p>The first argument, <tt>%val</tt> and the result may be integer types of
4906any bit width but they must have the same bit width. The second and third
4907arguments must be <tt>i32</tt> type.</p>
4908
4909<h5>Semantics:</h5>
4910<p>The '<tt>llvm.bit.part.select</tt>' intrinsic is the equivalent of shifting
4911<tt>%val</tt> right by <tt>%loBit</tt> bits and then ANDing it with a mask with
4912only the <tt>%hiBit - %loBit</tt> bits set, as follows:</p>
4913<ol>
4914 <li>The <tt>%val</tt> is shifted right (LSHR) by the number of bits specified
4915 by <tt>%loBits</tt>. This normalizes the value to the low order bits.</li>
4916 <li>The <tt>%loBits</tt> value is subtracted from the <tt>%hiBits</tt> value
4917 to determine the number of bits to retain.</li>
4918 <li>A mask of the retained bits is created by shifting a -1 value.</li>
4919 <li>The mask is ANDed with <tt>%val</tt> to produce the result.
4920</ol>
4921</div>
4922
4923<!-- _______________________________________________________________________ -->
4924<div class="doc_subsubsection">
4925 <a name="int_bit_part_set">'<tt>llvm.bit.part.set.*</tt>' Intrinsic</a>
4926</div>
4927
4928<div class="doc_text">
4929
4930<h5>Syntax:</h5>
4931<p>This is an overloaded intrinsic. You can use <tt>llvm.bit.part.set</tt> on
4932any integer bit width.
4933<pre>
4934 declare i17 @llvm.bit.part.set.i17.i17.i9 (i17 %val, i32 %bit, i9 %newbits)
4935 declare i29 @llvm.bit.part.set.i29.i29.i13(i29 %val, i32 %bit, i13 %newbits)
4936</pre>
4937
4938<h5>Overview:</h5>
4939<p>The '<tt>llvm.bit.part.set</tt>' family of intrinsic functions sets a range
4940of bits in a given value to a new value and returns the result.</p>
4941
4942<h5>Arguments:</h5>
4943<p>The first argument and the result may be an integer type of any bit width but
4944they must have the same bit width. The second argument must be an <tt>i32</tt>.
4945The third argument may be any any bit width less than or equal to the bit width
4946of the first argument.</p>
4947
4948<h5>Semantics:</h5>
4949<p>The '<tt>llvm.bit.part.set</tt>' intrinsic sets the value given by
4950<tt>%newbits</tt> into <tt>%val</tt> at the bit index given by <tt>%bit</tt>.
4951This is equivalent to the following sequence:</p>
4952<ol>
4953 <li>The bits in <tt>%val</tt> starting at <tt>%bit</tt> and up to the width
4954 of <tt>%newbits</tt> are cleared by ANDing them with a zero mask.</li>
4955 <li>The bits in <tt>%newbits</tt> are shifted left by <tt>%bit</tt> bits.
4956 <li>The shifted <tt>%newbits</tt> value is OR'd into <tt>%val</tt> to produce
4957 the result.</li>
4958</ol>
4959</div>
4960
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]4961<!-- ======================================================================= -->
4962<div class="doc_subsection">
4963 <a name="int_debugger">Debugger Intrinsics</a>
4964</div>
4965
4966<div class="doc_text">
4967<p>
4968The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
4969are described in the <a
4970href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
4971Debugging</a> document.
4972</p>
4973</div>
4974
4975
Jim Laskey2211f492007-03-14 19:31:19 +0000[diff] [blame]4976<!-- ======================================================================= -->
4977<div class="doc_subsection">
4978 <a name="int_eh">Exception Handling Intrinsics</a>
4979</div>
4980
4981<div class="doc_text">
4982<p> The LLVM exception handling intrinsics (which all start with
4983<tt>llvm.eh.</tt> prefix), are described in the <a
4984href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
4985Handling</a> document. </p>
4986</div>
4987
4988
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]4989<!-- *********************************************************************** -->
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]4990<hr>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]4991<address>
4992 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
4993 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
4994 <a href="http://validator.w3.org/check/referer"><img
4995 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
4996
4997 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencerca058542006-03-14 05:39:39 +0000[diff] [blame]4998 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]4999 Last modified: $Date$
5000</address>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]5001</body>
5002</html>