Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / e26dec68a1e03c80868a38344b06c990a5131e82 / . / docs / LangRef.html
blob: af88237adc817cec11db48d6d8192c96addb09eb [file] [log] [blame]
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3<html>
4<head>
5 <title>LLVM Assembly Language Reference Manual</title>
Reid Spencer3921c742004-08-26 20:44:00 +0000[diff] [blame]6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
7 <meta name="author" content="Chris Lattner">
8 <meta name="description"
9 content="LLVM Assembly Language Reference Manual.">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]10 <link rel="stylesheet" href="llvm.css" type="text/css">
11</head>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]12
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]13<body>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]14
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]15<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]16<ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]17 <li><a href="#abstract">Abstract</a></li>
18 <li><a href="#introduction">Introduction</a></li>
19 <li><a href="#identifiers">Identifiers</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]20 <li><a href="#highlevel">High Level Structure</a>
21 <ol>
22 <li><a href="#modulestructure">Module Structure</a></li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]23 <li><a href="#linkage">Linkage Types</a></li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]24 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]25 <li><a href="#globalvars">Global Variables</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]26 <li><a href="#functionstructure">Functions</a></li>
Anton Korobeynikovc6c98af2007-04-29 18:02:48 +0000[diff] [blame]27 <li><a href="#aliasstructure">Aliases</a>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]28 <li><a href="#paramattrs">Parameter Attributes</a></li>
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]29 <li><a href="#gc">Garbage Collector Names</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]30 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]31 <li><a href="#datalayout">Data Layout</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]32 </ol>
33 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]34 <li><a href="#typesystem">Type System</a>
35 <ol>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]36 <li><a href="#t_classifications">Type Classifications</a></li>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]37 <li><a href="#t_primitive">Primitive Types</a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]38 <ol>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]39 <li><a href="#t_floating">Floating Point Types</a></li>
40 <li><a href="#t_void">Void Type</a></li>
41 <li><a href="#t_label">Label Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]42 </ol>
43 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]44 <li><a href="#t_derived">Derived Types</a>
45 <ol>
Chris Lattnerb9488a62007-12-18 06:18:21 +0000[diff] [blame]46 <li><a href="#t_integer">Integer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]47 <li><a href="#t_array">Array Type</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]48 <li><a href="#t_function">Function Type</a></li>
49 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]50 <li><a href="#t_struct">Structure Type</a></li>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]51 <li><a href="#t_pstruct">Packed Structure Type</a></li>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]52 <li><a href="#t_vector">Vector Type</a></li>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]53 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]54 </ol>
55 </li>
56 </ol>
57 </li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]58 <li><a href="#constants">Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]59 <ol>
60 <li><a href="#simpleconstants">Simple Constants</a>
61 <li><a href="#aggregateconstants">Aggregate Constants</a>
62 <li><a href="#globalconstants">Global Variable and Function Addresses</a>
63 <li><a href="#undefvalues">Undefined Values</a>
64 <li><a href="#constantexprs">Constant Expressions</a>
65 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]66 </li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]67 <li><a href="#othervalues">Other Values</a>
68 <ol>
69 <li><a href="#inlineasm">Inline Assembler Expressions</a>
70 </ol>
71 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]72 <li><a href="#instref">Instruction Reference</a>
73 <ol>
74 <li><a href="#terminators">Terminator Instructions</a>
75 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]76 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
77 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]78 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
79 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]80 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]81 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]82 </ol>
83 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]84 <li><a href="#binaryops">Binary Operations</a>
85 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]86 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
87 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
88 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]89 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
90 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
91 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]92 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
93 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
94 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]95 </ol>
96 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]97 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
98 <ol>
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]99 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
100 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
101 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]102 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]103 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]104 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]105 </ol>
106 </li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]107 <li><a href="#vectorops">Vector Operations</a>
108 <ol>
109 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
110 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
111 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]112 </ol>
113 </li>
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]114 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]115 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]116 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
117 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
118 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]119 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
120 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
121 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]122 </ol>
123 </li>
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]124 <li><a href="#convertops">Conversion Operations</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]125 <ol>
126 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
127 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
128 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
129 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
130 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]131 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
132 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
133 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
134 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]135 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
136 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]137 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]138 </ol>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]139 <li><a href="#otherops">Other Operations</a>
140 <ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]141 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
142 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]143 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]144 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]145 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]146 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]147 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]148 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]149 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]150 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]151 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]152 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]153 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
154 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]155 <li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
156 <li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
157 <li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]158 </ol>
159 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]160 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
161 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]162 <li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
163 <li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
164 <li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]165 </ol>
166 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]167 <li><a href="#int_codegen">Code Generator Intrinsics</a>
168 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]169 <li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
170 <li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
171 <li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
172 <li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
173 <li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
174 <li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
175 <li><a href="#int_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]176 </ol>
177 </li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]178 <li><a href="#int_libc">Standard C Library Intrinsics</a>
179 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]180 <li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
181 <li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
182 <li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
183 <li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
184 <li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]185 <li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
186 <li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
187 <li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]188 </ol>
189 </li>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]190 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]191 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]192 <li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]193 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
194 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
195 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]196 <li><a href="#int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic </a></li>
197 <li><a href="#int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]198 </ol>
199 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]200 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]201 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]202 <li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]203 <ol>
204 <li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]205 </ol>
206 </li>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]207 <li><a href="#int_atomics">Atomic intrinsics</a>
208 <ol>
209 <li><a href="#int_memory_barrier"><tt>llvm.memory_barrier</tt></li>
210 </ol>
211 </li>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]212 <li><a href="#int_general">General intrinsics</a>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]213 <ol>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]214 <li><a href="#int_var_annotation">
Tanya Lattner91d0b882007-09-22 00:01:26 +0000[diff] [blame]215 <tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]216 <li><a href="#int_annotation">
Tanya Lattner91d0b882007-09-22 00:01:26 +0000[diff] [blame]217 <tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]218 <li><a href="#int_trap">
219 <tt>llvm.trap</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]220 </ol>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]221 </li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]222 </ol>
223 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]224</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]225
226<div class="doc_author">
227 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
228 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]229</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]230
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]231<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]232<div class="doc_section"> <a name="abstract">Abstract </a></div>
233<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]234
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]235<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]236<p>This document is a reference manual for the LLVM assembly language.
237LLVM is an SSA based representation that provides type safety,
238low-level operations, flexibility, and the capability of representing
239'all' high-level languages cleanly. It is the common code
240representation used throughout all phases of the LLVM compilation
241strategy.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]242</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]243
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]244<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]245<div class="doc_section"> <a name="introduction">Introduction</a> </div>
246<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]247
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]248<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]249
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]250<p>The LLVM code representation is designed to be used in three
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]251different forms: as an in-memory compiler IR, as an on-disk bitcode
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]252representation (suitable for fast loading by a Just-In-Time compiler),
253and as a human readable assembly language representation. This allows
254LLVM to provide a powerful intermediate representation for efficient
255compiler transformations and analysis, while providing a natural means
256to debug and visualize the transformations. The three different forms
257of LLVM are all equivalent. This document describes the human readable
258representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]259
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]260<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]261while being expressive, typed, and extensible at the same time. It
262aims to be a "universal IR" of sorts, by being at a low enough level
263that high-level ideas may be cleanly mapped to it (similar to how
264microprocessors are "universal IR's", allowing many source languages to
265be mapped to them). By providing type information, LLVM can be used as
266the target of optimizations: for example, through pointer analysis, it
267can be proven that a C automatic variable is never accessed outside of
268the current function... allowing it to be promoted to a simple SSA
269value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]270
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]271</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]272
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]273<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]274<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]275
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]276<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]277
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]278<p>It is important to note that this document describes 'well formed'
279LLVM assembly language. There is a difference between what the parser
280accepts and what is considered 'well formed'. For example, the
281following instruction is syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]282
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]283<div class="doc_code">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]284<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]285%x = <a href="#i_add">add</a> i32 1, %x
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]286</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]287</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]288
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]289<p>...because the definition of <tt>%x</tt> does not dominate all of
290its uses. The LLVM infrastructure provides a verification pass that may
291be used to verify that an LLVM module is well formed. This pass is
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]292automatically run by the parser after parsing input assembly and by
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]293the optimizer before it outputs bitcode. The violations pointed out
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]294by the verifier pass indicate bugs in transformation passes or input to
295the parser.</p>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]296</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]297
Chris Lattnercc689392007-10-03 17:34:29 +0000[diff] [blame]298<!-- Describe the typesetting conventions here. -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]299
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]300<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]301<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]302<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]303
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]304<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]305
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]306 <p>LLVM identifiers come in two basic types: global and local. Global
307 identifiers (functions, global variables) begin with the @ character. Local
308 identifiers (register names, types) begin with the % character. Additionally,
309 there are three different formats for identifiers, for different purposes:
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]310
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]311<ol>
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]312 <li>Named values are represented as a string of characters with their prefix.
313 For example, %foo, @DivisionByZero, %a.really.long.identifier. The actual
314 regular expression used is '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]315 Identifiers which require other characters in their names can be surrounded
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]316 with quotes. In this way, anything except a <tt>&quot;</tt> character can
317 be used in a named value.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]318
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]319 <li>Unnamed values are represented as an unsigned numeric value with their
320 prefix. For example, %12, @2, %44.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]321
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]322 <li>Constants, which are described in a <a href="#constants">section about
323 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]324</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]325
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]326<p>LLVM requires that values start with a prefix for two reasons: Compilers
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]327don't need to worry about name clashes with reserved words, and the set of
328reserved words may be expanded in the future without penalty. Additionally,
329unnamed identifiers allow a compiler to quickly come up with a temporary
330variable without having to avoid symbol table conflicts.</p>
331
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]332<p>Reserved words in LLVM are very similar to reserved words in other
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]333languages. There are keywords for different opcodes
334('<tt><a href="#i_add">add</a></tt>',
335 '<tt><a href="#i_bitcast">bitcast</a></tt>',
336 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]337href="#t_void">void</a></tt>', '<tt><a href="#t_primitive">i32</a></tt>', etc...),
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]338and others. These reserved words cannot conflict with variable names, because
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]339none of them start with a prefix character ('%' or '@').</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]340
341<p>Here is an example of LLVM code to multiply the integer variable
342'<tt>%X</tt>' by 8:</p>
343
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]344<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]345
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]346<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]347<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]348%result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]349</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]350</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]351
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]352<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]353
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]354<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]355<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]356%result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]357</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]358</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]359
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]360<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]361
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]362<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]363<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]364<a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
365<a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
366%result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]367</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]368</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]369
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]370<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
371important lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]372
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]373<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]374
375 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
376 line.</li>
377
378 <li>Unnamed temporaries are created when the result of a computation is not
379 assigned to a named value.</li>
380
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]381 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]382
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]383</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]384
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]385<p>...and it also shows a convention that we follow in this document. When
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]386demonstrating instructions, we will follow an instruction with a comment that
387defines the type and name of value produced. Comments are shown in italic
388text.</p>
389
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]390</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]391
392<!-- *********************************************************************** -->
393<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
394<!-- *********************************************************************** -->
395
396<!-- ======================================================================= -->
397<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
398</div>
399
400<div class="doc_text">
401
402<p>LLVM programs are composed of "Module"s, each of which is a
403translation unit of the input programs. Each module consists of
404functions, global variables, and symbol table entries. Modules may be
405combined together with the LLVM linker, which merges function (and
406global variable) definitions, resolves forward declarations, and merges
407symbol table entries. Here is an example of the "hello world" module:</p>
408
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]409<div class="doc_code">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]410<pre><i>; Declare the string constant as a global constant...</i>
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]411<a href="#identifiers">@.LC0</a> = <a href="#linkage_internal">internal</a> <a
412 href="#globalvars">constant</a> <a href="#t_array">[13 x i8]</a> c"hello world\0A\00" <i>; [13 x i8]*</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]413
414<i>; External declaration of the puts function</i>
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]415<a href="#functionstructure">declare</a> i32 @puts(i8 *) <i>; i32(i8 *)* </i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]416
417<i>; Definition of main function</i>
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]418define i32 @main() { <i>; i32()* </i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]419 <i>; Convert [13x i8 ]* to i8 *...</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]420 %cast210 = <a
Chris Lattner6c0955b2007-06-12 17:01:15 +0000[diff] [blame]421 href="#i_getelementptr">getelementptr</a> [13 x i8 ]* @.LC0, i64 0, i64 0 <i>; i8 *</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]422
423 <i>; Call puts function to write out the string to stdout...</i>
424 <a
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]425 href="#i_call">call</a> i32 @puts(i8 * %cast210) <i>; i32</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]426 <a
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]427 href="#i_ret">ret</a> i32 0<br>}<br>
428</pre>
429</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]430
431<p>This example is made up of a <a href="#globalvars">global variable</a>
432named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
433function, and a <a href="#functionstructure">function definition</a>
434for "<tt>main</tt>".</p>
435
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]436<p>In general, a module is made up of a list of global values,
437where both functions and global variables are global values. Global values are
438represented by a pointer to a memory location (in this case, a pointer to an
439array of char, and a pointer to a function), and have one of the following <a
440href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]441
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]442</div>
443
444<!-- ======================================================================= -->
445<div class="doc_subsection">
446 <a name="linkage">Linkage Types</a>
447</div>
448
449<div class="doc_text">
450
451<p>
452All Global Variables and Functions have one of the following types of linkage:
453</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]454
455<dl>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]456
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]457 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]458
459 <dd>Global values with internal linkage are only directly accessible by
460 objects in the current module. In particular, linking code into a module with
461 an internal global value may cause the internal to be renamed as necessary to
462 avoid collisions. Because the symbol is internal to the module, all
463 references can be updated. This corresponds to the notion of the
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]464 '<tt>static</tt>' keyword in C.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]465 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]466
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]467 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]468
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]469 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
470 the same name when linkage occurs. This is typically used to implement
471 inline functions, templates, or other code which must be generated in each
472 translation unit that uses it. Unreferenced <tt>linkonce</tt> globals are
473 allowed to be discarded.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]474 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]475
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]476 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]477
478 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
479 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]480 used for globals that may be emitted in multiple translation units, but that
481 are not guaranteed to be emitted into every translation unit that uses them.
482 One example of this are common globals in C, such as "<tt>int X;</tt>" at
483 global scope.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]484 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]485
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]486 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]487
488 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
489 pointer to array type. When two global variables with appending linkage are
490 linked together, the two global arrays are appended together. This is the
491 LLVM, typesafe, equivalent of having the system linker append together
492 "sections" with identical names when .o files are linked.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]493 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]494
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]495 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt>: </dt>
496 <dd>The semantics of this linkage follow the ELF model: the symbol is weak
497 until linked, if not linked, the symbol becomes null instead of being an
498 undefined reference.
499 </dd>
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]500
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]501 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]502
503 <dd>If none of the above identifiers are used, the global is externally
504 visible, meaning that it participates in linkage and can be used to resolve
505 external symbol references.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]506 </dd>
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]507</dl>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]508
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]509 <p>
510 The next two types of linkage are targeted for Microsoft Windows platform
511 only. They are designed to support importing (exporting) symbols from (to)
512 DLLs.
513 </p>
514
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]515 <dl>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]516 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt>: </dt>
517
518 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
519 or variable via a global pointer to a pointer that is set up by the DLL
520 exporting the symbol. On Microsoft Windows targets, the pointer name is
521 formed by combining <code>_imp__</code> and the function or variable name.
522 </dd>
523
524 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt>: </dt>
525
526 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
527 pointer to a pointer in a DLL, so that it can be referenced with the
528 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
529 name is formed by combining <code>_imp__</code> and the function or variable
530 name.
531 </dd>
532
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]533</dl>
534
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]535<p><a name="linkage_external"></a>For example, since the "<tt>.LC0</tt>"
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]536variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
537variable and was linked with this one, one of the two would be renamed,
538preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
539external (i.e., lacking any linkage declarations), they are accessible
Reid Spencerac8d2762007-01-05 00:59:10 +0000[diff] [blame]540outside of the current module.</p>
541<p>It is illegal for a function <i>declaration</i>
542to have any linkage type other than "externally visible", <tt>dllimport</tt>,
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]543or <tt>extern_weak</tt>.</p>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]544<p>Aliases can have only <tt>external</tt>, <tt>internal</tt> and <tt>weak</tt>
545linkages.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]546</div>
547
548<!-- ======================================================================= -->
549<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]550 <a name="callingconv">Calling Conventions</a>
551</div>
552
553<div class="doc_text">
554
555<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
556and <a href="#i_invoke">invokes</a> can all have an optional calling convention
557specified for the call. The calling convention of any pair of dynamic
558caller/callee must match, or the behavior of the program is undefined. The
559following calling conventions are supported by LLVM, and more may be added in
560the future:</p>
561
562<dl>
563 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
564
565 <dd>This calling convention (the default if no other calling convention is
566 specified) matches the target C calling conventions. This calling convention
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]567 supports varargs function calls and tolerates some mismatch in the declared
Reid Spencerc28d2bc2006-12-31 21:30:18 +0000[diff] [blame]568 prototype and implemented declaration of the function (as does normal C).
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]569 </dd>
570
571 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
572
573 <dd>This calling convention attempts to make calls as fast as possible
574 (e.g. by passing things in registers). This calling convention allows the
575 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattner8cdc5bc2005-05-06 23:08:23 +0000[diff] [blame]576 without having to conform to an externally specified ABI. Implementations of
577 this convention should allow arbitrary tail call optimization to be supported.
578 This calling convention does not support varargs and requires the prototype of
579 all callees to exactly match the prototype of the function definition.
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]580 </dd>
581
582 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
583
584 <dd>This calling convention attempts to make code in the caller as efficient
585 as possible under the assumption that the call is not commonly executed. As
586 such, these calls often preserve all registers so that the call does not break
587 any live ranges in the caller side. This calling convention does not support
588 varargs and requires the prototype of all callees to exactly match the
589 prototype of the function definition.
590 </dd>
591
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]592 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]593
594 <dd>Any calling convention may be specified by number, allowing
595 target-specific calling conventions to be used. Target specific calling
596 conventions start at 64.
597 </dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]598</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]599
600<p>More calling conventions can be added/defined on an as-needed basis, to
601support pascal conventions or any other well-known target-independent
602convention.</p>
603
604</div>
605
606<!-- ======================================================================= -->
607<div class="doc_subsection">
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]608 <a name="visibility">Visibility Styles</a>
609</div>
610
611<div class="doc_text">
612
613<p>
614All Global Variables and Functions have one of the following visibility styles:
615</p>
616
617<dl>
618 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
619
620 <dd>On ELF, default visibility means that the declaration is visible to other
621 modules and, in shared libraries, means that the declared entity may be
622 overridden. On Darwin, default visibility means that the declaration is
623 visible to other modules. Default visibility corresponds to "external
624 linkage" in the language.
625 </dd>
626
627 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
628
629 <dd>Two declarations of an object with hidden visibility refer to the same
630 object if they are in the same shared object. Usually, hidden visibility
631 indicates that the symbol will not be placed into the dynamic symbol table,
632 so no other module (executable or shared library) can reference it
633 directly.
634 </dd>
635
Anton Korobeynikov6f9896f2007-04-29 18:35:00 +0000[diff] [blame]636 <dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
637
638 <dd>On ELF, protected visibility indicates that the symbol will be placed in
639 the dynamic symbol table, but that references within the defining module will
640 bind to the local symbol. That is, the symbol cannot be overridden by another
641 module.
642 </dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]643</dl>
644
645</div>
646
647<!-- ======================================================================= -->
648<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]649 <a name="globalvars">Global Variables</a>
650</div>
651
652<div class="doc_text">
653
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]654<p>Global variables define regions of memory allocated at compilation time
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]655instead of run-time. Global variables may optionally be initialized, may have
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]656an explicit section to be placed in, and may have an optional explicit alignment
657specified. A variable may be defined as "thread_local", which means that it
658will not be shared by threads (each thread will have a separated copy of the
659variable). A variable may be defined as a global "constant," which indicates
660that the contents of the variable will <b>never</b> be modified (enabling better
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]661optimization, allowing the global data to be placed in the read-only section of
662an executable, etc). Note that variables that need runtime initialization
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]663cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]664
665<p>
666LLVM explicitly allows <em>declarations</em> of global variables to be marked
667constant, even if the final definition of the global is not. This capability
668can be used to enable slightly better optimization of the program, but requires
669the language definition to guarantee that optimizations based on the
670'constantness' are valid for the translation units that do not include the
671definition.
672</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]673
674<p>As SSA values, global variables define pointer values that are in
675scope (i.e. they dominate) all basic blocks in the program. Global
676variables always define a pointer to their "content" type because they
677describe a region of memory, and all memory objects in LLVM are
678accessed through pointers.</p>
679
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]680<p>A global variable may be declared to reside in a target-specifc numbered
681address space. For targets that support them, address spaces may affect how
682optimizations are performed and/or what target instructions are used to access
Christopher Lambd49e18d2007-12-12 08:44:39 +0000[diff] [blame]683the variable. The default address space is zero. The address space qualifier
684must precede any other attributes.</p>
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]685
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]686<p>LLVM allows an explicit section to be specified for globals. If the target
687supports it, it will emit globals to the section specified.</p>
688
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]689<p>An explicit alignment may be specified for a global. If not present, or if
690the alignment is set to zero, the alignment of the global is set by the target
691to whatever it feels convenient. If an explicit alignment is specified, the
692global is forced to have at least that much alignment. All alignments must be
693a power of 2.</p>
694
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]695<p>For example, the following defines a global in a numbered address space with
696an initializer, section, and alignment:</p>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]697
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]698<div class="doc_code">
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]699<pre>
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]700@G = constant float 1.0 addrspace(5), section "foo", align 4
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]701</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]702</div>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]703
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]704</div>
705
706
707<!-- ======================================================================= -->
708<div class="doc_subsection">
709 <a name="functionstructure">Functions</a>
710</div>
711
712<div class="doc_text">
713
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]714<p>LLVM function definitions consist of the "<tt>define</tt>" keyord,
715an optional <a href="#linkage">linkage type</a>, an optional
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]716<a href="#visibility">visibility style</a>, an optional
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]717<a href="#callingconv">calling convention</a>, a return type, an optional
718<a href="#paramattrs">parameter attribute</a> for the return type, a function
719name, a (possibly empty) argument list (each with optional
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]720<a href="#paramattrs">parameter attributes</a>), an optional section, an
Gordon Henriksene754abe2007-12-10 03:30:21 +0000[diff] [blame]721optional alignment, an optional <a href="#gc">garbage collector name</a>, an
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]722opening curly brace, a list of basic blocks, and a closing curly brace.
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]723
724LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
725optional <a href="#linkage">linkage type</a>, an optional
726<a href="#visibility">visibility style</a>, an optional
727<a href="#callingconv">calling convention</a>, a return type, an optional
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]728<a href="#paramattrs">parameter attribute</a> for the return type, a function
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]729name, a possibly empty list of arguments, an optional alignment, and an optional
Gordon Henriksene754abe2007-12-10 03:30:21 +0000[diff] [blame]730<a href="#gc">garbage collector name</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]731
732<p>A function definition contains a list of basic blocks, forming the CFG for
733the function. Each basic block may optionally start with a label (giving the
734basic block a symbol table entry), contains a list of instructions, and ends
735with a <a href="#terminators">terminator</a> instruction (such as a branch or
736function return).</p>
737
Chris Lattner4a3c9012007-06-08 16:52:14 +0000[diff] [blame]738<p>The first basic block in a function is special in two ways: it is immediately
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]739executed on entrance to the function, and it is not allowed to have predecessor
740basic blocks (i.e. there can not be any branches to the entry block of a
741function). Because the block can have no predecessors, it also cannot have any
742<a href="#i_phi">PHI nodes</a>.</p>
743
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]744<p>LLVM allows an explicit section to be specified for functions. If the target
745supports it, it will emit functions to the section specified.</p>
746
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]747<p>An explicit alignment may be specified for a function. If not present, or if
748the alignment is set to zero, the alignment of the function is set by the target
749to whatever it feels convenient. If an explicit alignment is specified, the
750function is forced to have at least that much alignment. All alignments must be
751a power of 2.</p>
752
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]753</div>
754
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]755
756<!-- ======================================================================= -->
757<div class="doc_subsection">
758 <a name="aliasstructure">Aliases</a>
759</div>
760<div class="doc_text">
761 <p>Aliases act as "second name" for the aliasee value (which can be either
Anton Korobeynikova80e1182007-04-28 13:45:00 +0000[diff] [blame]762 function or global variable or bitcast of global value). Aliases may have an
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]763 optional <a href="#linkage">linkage type</a>, and an
764 optional <a href="#visibility">visibility style</a>.</p>
765
766 <h5>Syntax:</h5>
767
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]768<div class="doc_code">
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]769<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]770@&lt;Name&gt; = [Linkage] [Visibility] alias &lt;AliaseeTy&gt; @&lt;Aliasee&gt;
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]771</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]772</div>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]773
774</div>
775
776
777
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]778<!-- ======================================================================= -->
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]779<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
780<div class="doc_text">
781 <p>The return type and each parameter of a function type may have a set of
782 <i>parameter attributes</i> associated with them. Parameter attributes are
783 used to communicate additional information about the result or parameters of
Duncan Sandsdc024672007-11-27 13:23:08 +0000[diff] [blame]784 a function. Parameter attributes are considered to be part of the function,
785 not of the function type, so functions with different parameter attributes
786 can have the same function type.</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]787
Reid Spencer950e9f82007-01-15 18:27:39 +0000[diff] [blame]788 <p>Parameter attributes are simple keywords that follow the type specified. If
789 multiple parameter attributes are needed, they are space separated. For
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]790 example:</p>
791
792<div class="doc_code">
793<pre>
Duncan Sandsdc024672007-11-27 13:23:08 +0000[diff] [blame]794declare i32 @printf(i8* noalias , ...) nounwind
795declare i32 @atoi(i8*) nounwind readonly
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]796</pre>
797</div>
798
Duncan Sandsdc024672007-11-27 13:23:08 +0000[diff] [blame]799 <p>Note that any attributes for the function result (<tt>nounwind</tt>,
800 <tt>readonly</tt>) come immediately after the argument list.</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]801
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]802 <p>Currently, only the following parameter attributes are defined:</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]803 <dl>
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]804 <dt><tt>zeroext</tt></dt>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]805 <dd>This indicates that the parameter should be zero extended just before
806 a call to this function.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]807
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]808 <dt><tt>signext</tt></dt>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]809 <dd>This indicates that the parameter should be sign extended just before
810 a call to this function.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]811
Anton Korobeynikov34d6dea2007-01-28 14:30:45 +0000[diff] [blame]812 <dt><tt>inreg</tt></dt>
813 <dd>This indicates that the parameter should be placed in register (if
Anton Korobeynikov66a8c8c2007-01-28 15:27:21 +0000[diff] [blame]814 possible) during assembling function call. Support for this attribute is
815 target-specific</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]816
817 <dt><tt>byval</tt></dt>
Chris Lattner0747baa2008-01-15 04:34:22 +0000[diff] [blame]818 <dd>This indicates that the pointer parameter should really be passed by
819 value to the function. The attribute implies that a hidden copy of the
820 pointee is made between the caller and the callee, so the callee is unable
821 to modify the value in the callee. This attribute is only valid on llvm
822 pointer arguments. It is generally used to pass structs and arrays by
823 value, but is also valid on scalars (even though this is silly).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]824
Anton Korobeynikov34d6dea2007-01-28 14:30:45 +0000[diff] [blame]825 <dt><tt>sret</tt></dt>
Duncan Sandse26dec62008-02-18 04:19:38 +0000[diff] [blame^]826 <dd>This indicates that the pointer parameter specifies the address of a
827 structure that is the return value of the function in the source program.
828 May only be applied to the first parameter.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]829
Zhou Shengfebca342007-06-05 05:28:26 +0000[diff] [blame]830 <dt><tt>noalias</tt></dt>
Owen Anderson117bbd32008-02-18 04:09:01 +0000[diff] [blame]831 <dd>This indicates that the parameter does not alias any global or any other
832 parameter. The caller is responsible for ensuring that this is the case,
833 usually by placing the value in a stack allocation.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]834
Reid Spencer2dc52012007-03-22 02:18:56 +0000[diff] [blame]835 <dt><tt>noreturn</tt></dt>
836 <dd>This function attribute indicates that the function never returns. This
837 indicates to LLVM that every call to this function should be treated as if
838 an <tt>unreachable</tt> instruction immediately followed the call.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]839
Reid Spencer67606122007-03-22 02:02:11 +0000[diff] [blame]840 <dt><tt>nounwind</tt></dt>
841 <dd>This function attribute indicates that the function type does not use
842 the unwind instruction and does not allow stack unwinding to propagate
843 through it.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]844
Duncan Sands50f19f52007-07-27 19:57:41 +0000[diff] [blame]845 <dt><tt>nest</tt></dt>
846 <dd>This indicates that the parameter can be excised using the
847 <a href="#int_trampoline">trampoline intrinsics</a>.</dd>
Duncan Sandsed4a2f12007-11-22 20:23:04 +0000[diff] [blame]848 <dt><tt>readonly</tt></dt>
Duncan Sandsf04d5842007-11-14 21:14:02 +0000[diff] [blame]849 <dd>This function attribute indicates that the function has no side-effects
Duncan Sandsed4a2f12007-11-22 20:23:04 +0000[diff] [blame]850 except for producing a return value or throwing an exception. The value
851 returned must only depend on the function arguments and/or global variables.
852 It may use values obtained by dereferencing pointers.</dd>
853 <dt><tt>readnone</tt></dt>
854 <dd>A <tt>readnone</tt> function has the same restrictions as a <tt>readonly</tt>
Duncan Sandsf04d5842007-11-14 21:14:02 +0000[diff] [blame]855 function, but in addition it is not allowed to dereference any pointer arguments
856 or global variables.
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]857 </dl>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]858
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]859</div>
860
861<!-- ======================================================================= -->
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]862<div class="doc_subsection">
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]863 <a name="gc">Garbage Collector Names</a>
864</div>
865
866<div class="doc_text">
867<p>Each function may specify a garbage collector name, which is simply a
868string.</p>
869
870<div class="doc_code"><pre
871>define void @f() gc "name" { ...</pre></div>
872
873<p>The compiler declares the supported values of <i>name</i>. Specifying a
874collector which will cause the compiler to alter its output in order to support
875the named garbage collection algorithm.</p>
876</div>
877
878<!-- ======================================================================= -->
879<div class="doc_subsection">
Chris Lattner1eeeb0c2006-04-08 04:40:53 +0000[diff] [blame]880 <a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]881</div>
882
883<div class="doc_text">
884<p>
885Modules may contain "module-level inline asm" blocks, which corresponds to the
886GCC "file scope inline asm" blocks. These blocks are internally concatenated by
887LLVM and treated as a single unit, but may be separated in the .ll file if
888desired. The syntax is very simple:
889</p>
890
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]891<div class="doc_code">
892<pre>
893module asm "inline asm code goes here"
894module asm "more can go here"
895</pre>
896</div>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]897
898<p>The strings can contain any character by escaping non-printable characters.
899 The escape sequence used is simply "\xx" where "xx" is the two digit hex code
900 for the number.
901</p>
902
903<p>
904 The inline asm code is simply printed to the machine code .s file when
905 assembly code is generated.
906</p>
907</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]908
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]909<!-- ======================================================================= -->
910<div class="doc_subsection">
911 <a name="datalayout">Data Layout</a>
912</div>
913
914<div class="doc_text">
915<p>A module may specify a target specific data layout string that specifies how
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]916data is to be laid out in memory. The syntax for the data layout is simply:</p>
917<pre> target datalayout = "<i>layout specification</i>"</pre>
918<p>The <i>layout specification</i> consists of a list of specifications
919separated by the minus sign character ('-'). Each specification starts with a
920letter and may include other information after the letter to define some
921aspect of the data layout. The specifications accepted are as follows: </p>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]922<dl>
923 <dt><tt>E</tt></dt>
924 <dd>Specifies that the target lays out data in big-endian form. That is, the
925 bits with the most significance have the lowest address location.</dd>
926 <dt><tt>e</tt></dt>
927 <dd>Specifies that hte target lays out data in little-endian form. That is,
928 the bits with the least significance have the lowest address location.</dd>
929 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
930 <dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
931 <i>preferred</i> alignments. All sizes are in bits. Specifying the <i>pref</i>
932 alignment is optional. If omitted, the preceding <tt>:</tt> should be omitted
933 too.</dd>
934 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
935 <dd>This specifies the alignment for an integer type of a given bit
936 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
937 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
938 <dd>This specifies the alignment for a vector type of a given bit
939 <i>size</i>.</dd>
940 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
941 <dd>This specifies the alignment for a floating point type of a given bit
942 <i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
943 (double).</dd>
944 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
945 <dd>This specifies the alignment for an aggregate type of a given bit
946 <i>size</i>.</dd>
947</dl>
948<p>When constructing the data layout for a given target, LLVM starts with a
949default set of specifications which are then (possibly) overriden by the
950specifications in the <tt>datalayout</tt> keyword. The default specifications
951are given in this list:</p>
952<ul>
953 <li><tt>E</tt> - big endian</li>
954 <li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
955 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
956 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
957 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
958 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
959 <li><tt>i64:32:64</tt> - i64 has abi alignment of 32-bits but preferred
960 alignment of 64-bits</li>
961 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
962 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
963 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
964 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
965 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
966</ul>
967<p>When llvm is determining the alignment for a given type, it uses the
968following rules:
969<ol>
970 <li>If the type sought is an exact match for one of the specifications, that
971 specification is used.</li>
972 <li>If no match is found, and the type sought is an integer type, then the
973 smallest integer type that is larger than the bitwidth of the sought type is
974 used. If none of the specifications are larger than the bitwidth then the the
975 largest integer type is used. For example, given the default specifications
976 above, the i7 type will use the alignment of i8 (next largest) while both
977 i65 and i256 will use the alignment of i64 (largest specified).</li>
978 <li>If no match is found, and the type sought is a vector type, then the
979 largest vector type that is smaller than the sought vector type will be used
980 as a fall back. This happens because <128 x double> can be implemented in
981 terms of 64 <2 x double>, for example.</li>
982</ol>
983</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]984
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]985<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]986<div class="doc_section"> <a name="typesystem">Type System</a> </div>
987<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]988
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]989<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]990
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]991<p>The LLVM type system is one of the most important features of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]992intermediate representation. Being typed enables a number of
993optimizations to be performed on the IR directly, without having to do
994extra analyses on the side before the transformation. A strong type
995system makes it easier to read the generated code and enables novel
996analyses and transformations that are not feasible to perform on normal
997three address code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]998
999</div>
1000
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1001<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1002<div class="doc_subsection"> <a name="t_classifications">Type
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1003Classifications</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1004<div class="doc_text">
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1005<p>The types fall into a few useful
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1006classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1007
1008<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1009 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1010 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1011 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1012 <td><a href="#t_integer">integer</a></td>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1013 <td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1014 </tr>
1015 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1016 <td><a href="#t_floating">floating point</a></td>
1017 <td><tt>float, double, x86_fp80, fp128, ppc_fp128</tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1018 </tr>
1019 <tr>
1020 <td><a name="t_firstclass">first class</a></td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1021 <td><a href="#t_integer">integer</a>,
1022 <a href="#t_floating">floating point</a>,
1023 <a href="#t_pointer">pointer</a>,
1024 <a href="#t_vector">vector</a>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1025 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1026 </tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1027 <tr>
1028 <td><a href="#t_primitive">primitive</a></td>
1029 <td><a href="#t_label">label</a>,
1030 <a href="#t_void">void</a>,
1031 <a href="#t_integer">integer</a>,
1032 <a href="#t_floating">floating point</a>.</td>
1033 </tr>
1034 <tr>
1035 <td><a href="#t_derived">derived</a></td>
1036 <td><a href="#t_integer">integer</a>,
1037 <a href="#t_array">array</a>,
1038 <a href="#t_function">function</a>,
1039 <a href="#t_pointer">pointer</a>,
1040 <a href="#t_struct">structure</a>,
1041 <a href="#t_pstruct">packed structure</a>,
1042 <a href="#t_vector">vector</a>,
1043 <a href="#t_opaque">opaque</a>.
1044 </tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1045 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1046</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1047
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1048<p>The <a href="#t_firstclass">first class</a> types are perhaps the
1049most important. Values of these types are the only ones which can be
1050produced by instructions, passed as arguments, or used as operands to
1051instructions. This means that all structures and arrays must be
1052manipulated either by pointer or by component.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1053</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1054
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1055<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1056<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1057
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1058<div class="doc_text">
1059<p>The primitive types are the fundamental building blocks of the LLVM
1060system.</p>
1061
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1062</div>
1063
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1064<!-- _______________________________________________________________________ -->
1065<div class="doc_subsubsection"> <a name="t_floating">Floating Point Types</a> </div>
1066
1067<div class="doc_text">
1068 <table>
1069 <tbody>
1070 <tr><th>Type</th><th>Description</th></tr>
1071 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
1072 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
1073 <tr><td><tt>fp128</tt></td><td>128-bit floating point value (112-bit mantissa)</td></tr>
1074 <tr><td><tt>x86_fp80</tt></td><td>80-bit floating point value (X87)</td></tr>
1075 <tr><td><tt>ppc_fp128</tt></td><td>128-bit floating point value (two 64-bits)</td></tr>
1076 </tbody>
1077 </table>
1078</div>
1079
1080<!-- _______________________________________________________________________ -->
1081<div class="doc_subsubsection"> <a name="t_void">Void Type</a> </div>
1082
1083<div class="doc_text">
1084<h5>Overview:</h5>
1085<p>The void type does not represent any value and has no size.</p>
1086
1087<h5>Syntax:</h5>
1088
1089<pre>
1090 void
1091</pre>
1092</div>
1093
1094<!-- _______________________________________________________________________ -->
1095<div class="doc_subsubsection"> <a name="t_label">Label Type</a> </div>
1096
1097<div class="doc_text">
1098<h5>Overview:</h5>
1099<p>The label type represents code labels.</p>
1100
1101<h5>Syntax:</h5>
1102
1103<pre>
1104 label
1105</pre>
1106</div>
1107
1108
1109<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1110<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1111
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1112<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1113
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1114<p>The real power in LLVM comes from the derived types in the system.
1115This is what allows a programmer to represent arrays, functions,
1116pointers, and other useful types. Note that these derived types may be
1117recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1118
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1119</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1120
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1121<!-- _______________________________________________________________________ -->
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1122<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
1123
1124<div class="doc_text">
1125
1126<h5>Overview:</h5>
1127<p>The integer type is a very simple derived type that simply specifies an
1128arbitrary bit width for the integer type desired. Any bit width from 1 bit to
11292^23-1 (about 8 million) can be specified.</p>
1130
1131<h5>Syntax:</h5>
1132
1133<pre>
1134 iN
1135</pre>
1136
1137<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
1138value.</p>
1139
1140<h5>Examples:</h5>
1141<table class="layout">
Chris Lattnerb9488a62007-12-18 06:18:21 +0000[diff] [blame]1142 <tbody>
1143 <tr>
1144 <td><tt>i1</tt></td>
1145 <td>a single-bit integer.</td>
1146 </tr><tr>
1147 <td><tt>i32</tt></td>
1148 <td>a 32-bit integer.</td>
1149 </tr><tr>
1150 <td><tt>i1942652</tt></td>
1151 <td>a really big integer of over 1 million bits.</td>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1152 </tr>
Chris Lattnerb9488a62007-12-18 06:18:21 +0000[diff] [blame]1153 </tbody>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1154</table>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1155</div>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1156
1157<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1158<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1159
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1160<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1161
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1162<h5>Overview:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1163
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1164<p>The array type is a very simple derived type that arranges elements
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1165sequentially in memory. The array type requires a size (number of
1166elements) and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1167
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1168<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1169
1170<pre>
1171 [&lt;# elements&gt; x &lt;elementtype&gt;]
1172</pre>
1173
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1174<p>The number of elements is a constant integer value; elementtype may
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1175be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1176
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1177<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1178<table class="layout">
1179 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1180 <td class="left"><tt>[40 x i32]</tt></td>
1181 <td class="left">Array of 40 32-bit integer values.</td>
1182 </tr>
1183 <tr class="layout">
1184 <td class="left"><tt>[41 x i32]</tt></td>
1185 <td class="left">Array of 41 32-bit integer values.</td>
1186 </tr>
1187 <tr class="layout">
1188 <td class="left"><tt>[4 x i8]</tt></td>
1189 <td class="left">Array of 4 8-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1190 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1191</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1192<p>Here are some examples of multidimensional arrays:</p>
1193<table class="layout">
1194 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1195 <td class="left"><tt>[3 x [4 x i32]]</tt></td>
1196 <td class="left">3x4 array of 32-bit integer values.</td>
1197 </tr>
1198 <tr class="layout">
1199 <td class="left"><tt>[12 x [10 x float]]</tt></td>
1200 <td class="left">12x10 array of single precision floating point values.</td>
1201 </tr>
1202 <tr class="layout">
1203 <td class="left"><tt>[2 x [3 x [4 x i16]]]</tt></td>
1204 <td class="left">2x3x4 array of 16-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1205 </tr>
1206</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1207
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1208<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
1209length array. Normally, accesses past the end of an array are undefined in
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1210LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
1211As a special case, however, zero length arrays are recognized to be variable
1212length. This allows implementation of 'pascal style arrays' with the LLVM
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1213type "{ i32, [0 x float]}", for example.</p>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1214
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1215</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1216
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1217<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1218<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1219<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1220<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1221<p>The function type can be thought of as a function signature. It
1222consists of a return type and a list of formal parameter types.
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]1223Function types are usually used to build virtual function tables
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1224(which are structures of pointers to functions), for indirect function
1225calls, and when defining a function.</p>
John Criswell009900b2003-11-25 21:45:46 +0000[diff] [blame]1226<p>
1227The return type of a function type cannot be an aggregate type.
1228</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1229<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1230<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1231<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]1232specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1233which indicates that the function takes a variable number of arguments.
1234Variable argument functions can access their arguments with the <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1235 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1236<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1237<table class="layout">
1238 <tr class="layout">
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1239 <td class="left"><tt>i32 (i32)</tt></td>
1240 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1241 </td>
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1242 </tr><tr class="layout">
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]1243 <td class="left"><tt>float&nbsp;(i16&nbsp;signext,&nbsp;i32&nbsp;*)&nbsp;*
Reid Spencerf17a0b72006-12-31 07:20:23 +0000[diff] [blame]1244 </tt></td>
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1245 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
1246 an <tt>i16</tt> that should be sign extended and a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1247 <a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1248 <tt>float</tt>.
1249 </td>
1250 </tr><tr class="layout">
1251 <td class="left"><tt>i32 (i8*, ...)</tt></td>
1252 <td class="left">A vararg function that takes at least one
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]1253 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1254 which returns an integer. This is the signature for <tt>printf</tt> in
1255 LLVM.
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1256 </td>
1257 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1258</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1259
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1260</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1261<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1262<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1263<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1264<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1265<p>The structure type is used to represent a collection of data members
1266together in memory. The packing of the field types is defined to match
1267the ABI of the underlying processor. The elements of a structure may
1268be any type that has a size.</p>
1269<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1270and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1271field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1272instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1273<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1274<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1275<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1276<table class="layout">
1277 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1278 <td class="left"><tt>{ i32, i32, i32 }</tt></td>
1279 <td class="left">A triple of three <tt>i32</tt> values</td>
1280 </tr><tr class="layout">
1281 <td class="left"><tt>{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}</tt></td>
1282 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1283 second element is a <a href="#t_pointer">pointer</a> to a
1284 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1285 an <tt>i32</tt>.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1286 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1287</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1288</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1289
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1290<!-- _______________________________________________________________________ -->
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1291<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1292</div>
1293<div class="doc_text">
1294<h5>Overview:</h5>
1295<p>The packed structure type is used to represent a collection of data members
1296together in memory. There is no padding between fields. Further, the alignment
1297of a packed structure is 1 byte. The elements of a packed structure may
1298be any type that has a size.</p>
1299<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1300and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1301field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1302instruction.</p>
1303<h5>Syntax:</h5>
1304<pre> &lt; { &lt;type list&gt; } &gt; <br></pre>
1305<h5>Examples:</h5>
1306<table class="layout">
1307 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1308 <td class="left"><tt>&lt; { i32, i32, i32 } &gt;</tt></td>
1309 <td class="left">A triple of three <tt>i32</tt> values</td>
1310 </tr><tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1311 <td class="left"><tt>&lt; { float, i32 (i32)* } &gt;</tt></td>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1312 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1313 second element is a <a href="#t_pointer">pointer</a> to a
1314 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1315 an <tt>i32</tt>.</td>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1316 </tr>
1317</table>
1318</div>
1319
1320<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1321<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1322<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1323<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1324<p>As in many languages, the pointer type represents a pointer or
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]1325reference to another object, which must live in memory. Pointer types may have
1326an optional address space attribute defining the target-specific numbered
1327address space where the pointed-to object resides. The default address space is
1328zero.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1329<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1330<pre> &lt;type&gt; *<br></pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1331<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1332<table class="layout">
1333 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1334 <td class="left"><tt>[4x i32]*</tt></td>
1335 <td class="left">A <a href="#t_pointer">pointer</a> to <a
1336 href="#t_array">array</a> of four <tt>i32</tt> values.</td>
1337 </tr>
1338 <tr class="layout">
1339 <td class="left"><tt>i32 (i32 *) *</tt></td>
1340 <td class="left"> A <a href="#t_pointer">pointer</a> to a <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1341 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1342 <tt>i32</tt>.</td>
1343 </tr>
1344 <tr class="layout">
1345 <td class="left"><tt>i32 addrspace(5)*</tt></td>
1346 <td class="left">A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value
1347 that resides in address space #5.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1348 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1349</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1350</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1351
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1352<!-- _______________________________________________________________________ -->
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1353<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1354<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1355
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1356<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1357
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1358<p>A vector type is a simple derived type that represents a vector
1359of elements. Vector types are used when multiple primitive data
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1360are operated in parallel using a single instruction (SIMD).
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1361A vector type requires a size (number of
Chris Lattnerb8d172f2005-11-10 01:44:22 +0000[diff] [blame]1362elements) and an underlying primitive data type. Vectors must have a power
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1363of two length (1, 2, 4, 8, 16 ...). Vector types are
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1364considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1365
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1366<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1367
1368<pre>
1369 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1370</pre>
1371
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1372<p>The number of elements is a constant integer value; elementtype may
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1373be any integer or floating point type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1374
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1375<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1376
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1377<table class="layout">
1378 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1379 <td class="left"><tt>&lt;4 x i32&gt;</tt></td>
1380 <td class="left">Vector of 4 32-bit integer values.</td>
1381 </tr>
1382 <tr class="layout">
1383 <td class="left"><tt>&lt;8 x float&gt;</tt></td>
1384 <td class="left">Vector of 8 32-bit floating-point values.</td>
1385 </tr>
1386 <tr class="layout">
1387 <td class="left"><tt>&lt;2 x i64&gt;</tt></td>
1388 <td class="left">Vector of 2 64-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1389 </tr>
1390</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1391</div>
1392
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1393<!-- _______________________________________________________________________ -->
1394<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1395<div class="doc_text">
1396
1397<h5>Overview:</h5>
1398
1399<p>Opaque types are used to represent unknown types in the system. This
Gordon Henriksen8ac04ff2007-10-14 00:34:53 +0000[diff] [blame]1400corresponds (for example) to the C notion of a forward declared structure type.
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1401In LLVM, opaque types can eventually be resolved to any type (not just a
1402structure type).</p>
1403
1404<h5>Syntax:</h5>
1405
1406<pre>
1407 opaque
1408</pre>
1409
1410<h5>Examples:</h5>
1411
1412<table class="layout">
1413 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1414 <td class="left"><tt>opaque</tt></td>
1415 <td class="left">An opaque type.</td>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1416 </tr>
1417</table>
1418</div>
1419
1420
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1421<!-- *********************************************************************** -->
1422<div class="doc_section"> <a name="constants">Constants</a> </div>
1423<!-- *********************************************************************** -->
1424
1425<div class="doc_text">
1426
1427<p>LLVM has several different basic types of constants. This section describes
1428them all and their syntax.</p>
1429
1430</div>
1431
1432<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1433<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1434
1435<div class="doc_text">
1436
1437<dl>
1438 <dt><b>Boolean constants</b></dt>
1439
1440 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1441 constants of the <tt><a href="#t_primitive">i1</a></tt> type.
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1442 </dd>
1443
1444 <dt><b>Integer constants</b></dt>
1445
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1446 <dd>Standard integers (such as '4') are constants of the <a
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]1447 href="#t_integer">integer</a> type. Negative numbers may be used with
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1448 integer types.
1449 </dd>
1450
1451 <dt><b>Floating point constants</b></dt>
1452
1453 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
1454 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1455 notation (see below). Floating point constants must have a <a
1456 href="#t_floating">floating point</a> type. </dd>
1457
1458 <dt><b>Null pointer constants</b></dt>
1459
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1460 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1461 and must be of <a href="#t_pointer">pointer type</a>.</dd>
1462
1463</dl>
1464
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1465<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1466of floating point constants. For example, the form '<tt>double
14670x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
14684.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1469(and the only time that they are generated by the disassembler) is when a
1470floating point constant must be emitted but it cannot be represented as a
1471decimal floating point number. For example, NaN's, infinities, and other
1472special values are represented in their IEEE hexadecimal format so that
1473assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1474
1475</div>
1476
1477<!-- ======================================================================= -->
1478<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
1479</div>
1480
1481<div class="doc_text">
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]1482<p>Aggregate constants arise from aggregation of simple constants
1483and smaller aggregate constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1484
1485<dl>
1486 <dt><b>Structure constants</b></dt>
1487
1488 <dd>Structure constants are represented with notation similar to structure
1489 type definitions (a comma separated list of elements, surrounded by braces
Chris Lattner64910ee2007-12-25 20:34:52 +0000[diff] [blame]1490 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* @G }</tt>",
1491 where "<tt>@G</tt>" is declared as "<tt>@G = external global i32</tt>". Structure constants
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]1492 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1493 types of elements must match those specified by the type.
1494 </dd>
1495
1496 <dt><b>Array constants</b></dt>
1497
1498 <dd>Array constants are represented with notation similar to array type
1499 definitions (a comma separated list of elements, surrounded by square brackets
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1500 (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74 ]</tt>". Array
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1501 constants must have <a href="#t_array">array type</a>, and the number and
1502 types of elements must match those specified by the type.
1503 </dd>
1504
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1505 <dt><b>Vector constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1506
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1507 <dd>Vector constants are represented with notation similar to vector type
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1508 definitions (a comma separated list of elements, surrounded by
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1509 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32 42,
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1510 i32 11, i32 74, i32 100 &gt;</tt>". Vector constants must have <a
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1511 href="#t_vector">vector type</a>, and the number and types of elements must
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1512 match those specified by the type.
1513 </dd>
1514
1515 <dt><b>Zero initialization</b></dt>
1516
1517 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
1518 value to zero of <em>any</em> type, including scalar and aggregate types.
1519 This is often used to avoid having to print large zero initializers (e.g. for
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1520 large arrays) and is always exactly equivalent to using explicit zero
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1521 initializers.
1522 </dd>
1523</dl>
1524
1525</div>
1526
1527<!-- ======================================================================= -->
1528<div class="doc_subsection">
1529 <a name="globalconstants">Global Variable and Function Addresses</a>
1530</div>
1531
1532<div class="doc_text">
1533
1534<p>The addresses of <a href="#globalvars">global variables</a> and <a
1535href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1536constants. These constants are explicitly referenced when the <a
1537href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1538href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1539file:</p>
1540
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1541<div class="doc_code">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1542<pre>
Chris Lattnera18a4242007-06-06 18:28:13 +0000[diff] [blame]1543@X = global i32 17
1544@Y = global i32 42
1545@Z = global [2 x i32*] [ i32* @X, i32* @Y ]
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1546</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1547</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1548
1549</div>
1550
1551<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1552<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1553<div class="doc_text">
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1554 <p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1555 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1556 a constant is permitted.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1557
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1558 <p>Undefined values indicate to the compiler that the program is well defined
1559 no matter what value is used, giving the compiler more freedom to optimize.
1560 </p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1561</div>
1562
1563<!-- ======================================================================= -->
1564<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1565</div>
1566
1567<div class="doc_text">
1568
1569<p>Constant expressions are used to allow expressions involving other constants
1570to be used as constants. Constant expressions may be of any <a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1571href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1572that does not have side effects (e.g. load and call are not supported). The
1573following is the syntax for constant expressions:</p>
1574
1575<dl>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1576 <dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
1577 <dd>Truncate a constant to another type. The bit size of CST must be larger
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1578 than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1579
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1580 <dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
1581 <dd>Zero extend a constant to another type. The bit size of CST must be
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1582 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1583
1584 <dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
1585 <dd>Sign extend a constant to another type. The bit size of CST must be
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1586 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1587
1588 <dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
1589 <dd>Truncate a floating point constant to another floating point type. The
1590 size of CST must be larger than the size of TYPE. Both types must be
1591 floating point.</dd>
1592
1593 <dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
1594 <dd>Floating point extend a constant to another type. The size of CST must be
1595 smaller or equal to the size of TYPE. Both types must be floating point.</dd>
1596
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]1597 <dt><b><tt>fptoui ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1598 <dd>Convert a floating point constant to the corresponding unsigned integer
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1599 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1600 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1601 of the same number of elements. If the value won't fit in the integer type,
1602 the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1603
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]1604 <dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1605 <dd>Convert a floating point constant to the corresponding signed integer
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1606 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1607 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1608 of the same number of elements. If the value won't fit in the integer type,
1609 the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1610
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]1611 <dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1612 <dd>Convert an unsigned integer constant to the corresponding floating point
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1613 constant. TYPE must be a scalar or vector floating point type. CST must be of
1614 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1615 of the same number of elements. If the value won't fit in the floating point
1616 type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1617
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]1618 <dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1619 <dd>Convert a signed integer constant to the corresponding floating point
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1620 constant. TYPE must be a scalar or vector floating point type. CST must be of
1621 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1622 of the same number of elements. If the value won't fit in the floating point
1623 type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1624
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]1625 <dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
1626 <dd>Convert a pointer typed constant to the corresponding integer constant
1627 TYPE must be an integer type. CST must be of pointer type. The CST value is
1628 zero extended, truncated, or unchanged to make it fit in TYPE.</dd>
1629
1630 <dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
1631 <dd>Convert a integer constant to a pointer constant. TYPE must be a
1632 pointer type. CST must be of integer type. The CST value is zero extended,
1633 truncated, or unchanged to make it fit in a pointer size. This one is
1634 <i>really</i> dangerous!</dd>
1635
1636 <dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1637 <dd>Convert a constant, CST, to another TYPE. The size of CST and TYPE must be
1638 identical (same number of bits). The conversion is done as if the CST value
1639 was stored to memory and read back as TYPE. In other words, no bits change
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]1640 with this operator, just the type. This can be used for conversion of
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1641 vector types to any other type, as long as they have the same bit width. For
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]1642 pointers it is only valid to cast to another pointer type.
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1643 </dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1644
1645 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1646
1647 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1648 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1649 instruction, the index list may have zero or more indexes, which are required
1650 to make sense for the type of "CSTPTR".</dd>
1651
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1652 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
1653
1654 <dd>Perform the <a href="#i_select">select operation</a> on
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]1655 constants.</dd>
1656
1657 <dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
1658 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
1659
1660 <dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
1661 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1662
1663 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
1664
1665 <dd>Perform the <a href="#i_extractelement">extractelement
1666 operation</a> on constants.
1667
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]1668 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
1669
1670 <dd>Perform the <a href="#i_insertelement">insertelement
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]1671 operation</a> on constants.</dd>
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]1672
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]1673
1674 <dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
1675
1676 <dd>Perform the <a href="#i_shufflevector">shufflevector
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]1677 operation</a> on constants.</dd>
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]1678
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1679 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1680
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1681 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1682 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1683 binary</a> operations. The constraints on operands are the same as those for
1684 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1685 values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1686</dl>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1687</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]1688
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1689<!-- *********************************************************************** -->
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1690<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
1691<!-- *********************************************************************** -->
1692
1693<!-- ======================================================================= -->
1694<div class="doc_subsection">
1695<a name="inlineasm">Inline Assembler Expressions</a>
1696</div>
1697
1698<div class="doc_text">
1699
1700<p>
1701LLVM supports inline assembler expressions (as opposed to <a href="#moduleasm">
1702Module-Level Inline Assembly</a>) through the use of a special value. This
1703value represents the inline assembler as a string (containing the instructions
1704to emit), a list of operand constraints (stored as a string), and a flag that
1705indicates whether or not the inline asm expression has side effects. An example
1706inline assembler expression is:
1707</p>
1708
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1709<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1710<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1711i32 (i32) asm "bswap $0", "=r,r"
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1712</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1713</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1714
1715<p>
1716Inline assembler expressions may <b>only</b> be used as the callee operand of
1717a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we have:
1718</p>
1719
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1720<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1721<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1722%X = call i32 asm "<a href="#int_bswap">bswap</a> $0", "=r,r"(i32 %Y)
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1723</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1724</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1725
1726<p>
1727Inline asms with side effects not visible in the constraint list must be marked
1728as having side effects. This is done through the use of the
1729'<tt>sideeffect</tt>' keyword, like so:
1730</p>
1731
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1732<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1733<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1734call void asm sideeffect "eieio", ""()
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1735</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1736</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1737
1738<p>TODO: The format of the asm and constraints string still need to be
1739documented here. Constraints on what can be done (e.g. duplication, moving, etc
1740need to be documented).
1741</p>
1742
1743</div>
1744
1745<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1746<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1747<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1748
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1749<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1750
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1751<p>The LLVM instruction set consists of several different
1752classifications of instructions: <a href="#terminators">terminator
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1753instructions</a>, <a href="#binaryops">binary instructions</a>,
1754<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1755 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1756instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1757
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1758</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1759
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1760<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1761<div class="doc_subsection"> <a name="terminators">Terminator
1762Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1763
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1764<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1765
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1766<p>As mentioned <a href="#functionstructure">previously</a>, every
1767basic block in a program ends with a "Terminator" instruction, which
1768indicates which block should be executed after the current block is
1769finished. These terminator instructions typically yield a '<tt>void</tt>'
1770value: they produce control flow, not values (the one exception being
1771the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1772<p>There are six different terminator instructions: the '<a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1773 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1774instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1775the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1776 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1777 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1778
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1779</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1780
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1781<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1782<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1783Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1784<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1785<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1786<pre> ret &lt;type&gt; &lt;value&gt; <i>; Return a value from a non-void function</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1787 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1788</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1789<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1790<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1791value) from a function back to the caller.</p>
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]1792<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1793returns a value and then causes control flow, and one that just causes
1794control flow to occur.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1795<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1796<p>The '<tt>ret</tt>' instruction may return any '<a
1797 href="#t_firstclass">first class</a>' type. Notice that a function is
1798not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1799instruction inside of the function that returns a value that does not
1800match the return type of the function.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1801<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1802<p>When the '<tt>ret</tt>' instruction is executed, control flow
1803returns back to the calling function's context. If the caller is a "<a
John Criswellfa081872004-06-25 15:16:57 +0000[diff] [blame]1804 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1805the instruction after the call. If the caller was an "<a
1806 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1807at the beginning of the "normal" destination block. If the instruction
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1808returns a value, that value shall set the call or invoke instruction's
1809return value.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1810<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1811<pre> ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1812 ret void <i>; Return from a void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1813</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1814</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1815<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1816<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1817<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1818<h5>Syntax:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1819<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 Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1820</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1821<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1822<p>The '<tt>br</tt>' instruction is used to cause control flow to
1823transfer to a different basic block in the current function. There are
1824two forms of this instruction, corresponding to a conditional branch
1825and an unconditional branch.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1826<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1827<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1828single '<tt>i1</tt>' value and two '<tt>label</tt>' values. The
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1829unconditional form of the '<tt>br</tt>' instruction takes a single
1830'<tt>label</tt>' value as a target.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1831<h5>Semantics:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1832<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1833argument is evaluated. If the value is <tt>true</tt>, control flows
1834to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1835control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1836<h5>Example:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1837<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 Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1838 href="#i_ret">ret</a> i32 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> i32 0<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1839</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1840<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1841<div class="doc_subsubsection">
1842 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1843</div>
1844
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1845<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1846<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1847
1848<pre>
1849 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1850</pre>
1851
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1852<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1853
1854<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1855several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1856instruction, allowing a branch to occur to one of many possible
1857destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1858
1859
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1860<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1861
1862<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1863comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1864an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1865table is not allowed to contain duplicate constant entries.</p>
1866
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1867<h5>Semantics:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1868
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1869<p>The <tt>switch</tt> instruction specifies a table of values and
1870destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1871table is searched for the given value. If the value is found, control flow is
1872transfered to the corresponding destination; otherwise, control flow is
1873transfered to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1874
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1875<h5>Implementation:</h5>
1876
1877<p>Depending on properties of the target machine and the particular
1878<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]1879ways. For example, it could be generated as a series of chained conditional
1880branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1881
1882<h5>Example:</h5>
1883
1884<pre>
1885 <i>; Emulate a conditional br instruction</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1886 %Val = <a href="#i_zext">zext</a> i1 %value to i32
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1887 switch i32 %Val, label %truedest [i32 0, label %falsedest ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1888
1889 <i>; Emulate an unconditional br instruction</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1890 switch i32 0, label %dest [ ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]1891
1892 <i>; Implement a jump table:</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1893 switch i32 %val, label %otherwise [ i32 0, label %onzero
1894 i32 1, label %onone
1895 i32 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1896</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1897</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1898
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1899<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1900<div class="doc_subsubsection">
1901 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1902</div>
1903
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1904<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1905
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1906<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1907
1908<pre>
1909 &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 Lattner76b8a332006-05-14 18:23:06 +0000[diff] [blame]1910 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1911</pre>
1912
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]1913<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1914
1915<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1916function, with the possibility of control flow transfer to either the
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1917'<tt>normal</tt>' label or the
1918'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1919"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1920"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1921href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1922continued at the dynamically nearest "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1923
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1924<h5>Arguments:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1925
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1926<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1927
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1928<ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1929 <li>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]1930 The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1931 convention</a> the call should use. If none is specified, the call defaults
1932 to using C calling conventions.
1933 </li>
1934 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1935 function value being invoked. In most cases, this is a direct function
1936 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1937 an arbitrary pointer to function value.
1938 </li>
1939
1940 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1941 function to be invoked. </li>
1942
1943 <li>'<tt>function args</tt>': argument list whose types match the function
1944 signature argument types. If the function signature indicates the function
1945 accepts a variable number of arguments, the extra arguments can be
1946 specified. </li>
1947
1948 <li>'<tt>normal label</tt>': the label reached when the called function
1949 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1950
1951 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1952 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1953
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1954</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1955
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1956<h5>Semantics:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1957
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1958<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1959href="#i_call">call</a></tt>' instruction in most regards. The primary
1960difference is that it establishes an association with a label, which is used by
1961the runtime library to unwind the stack.</p>
1962
1963<p>This instruction is used in languages with destructors to ensure that proper
1964cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1965exception. Additionally, this is important for implementation of
1966'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1967
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1968<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]1969<pre>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1970 %retval = invoke i32 %Test(i32 15) to label %Continue
1971 unwind label %TestCleanup <i>; {i32}:retval set</i>
1972 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Test(i32 15) to label %Continue
1973 unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1974</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1975</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1976
1977
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1978<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1979
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1980<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1981Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1982
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1983<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1984
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1985<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1986<pre>
1987 unwind
1988</pre>
1989
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1990<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1991
1992<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1993at the first callee in the dynamic call stack which used an <a
1994href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1995primarily used to implement exception handling.</p>
1996
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1997<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]1998
1999<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
2000immediately halt. The dynamic call stack is then searched for the first <a
2001href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
2002execution continues at the "exceptional" destination block specified by the
2003<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
2004dynamic call chain, undefined behavior results.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2005</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2006
2007<!-- _______________________________________________________________________ -->
2008
2009<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
2010Instruction</a> </div>
2011
2012<div class="doc_text">
2013
2014<h5>Syntax:</h5>
2015<pre>
2016 unreachable
2017</pre>
2018
2019<h5>Overview:</h5>
2020
2021<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
2022instruction is used to inform the optimizer that a particular portion of the
2023code is not reachable. This can be used to indicate that the code after a
2024no-return function cannot be reached, and other facts.</p>
2025
2026<h5>Semantics:</h5>
2027
2028<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
2029</div>
2030
2031
2032
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2033<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2034<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2035<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2036<p>Binary operators are used to do most of the computation in a
2037program. They require two operands, execute an operation on them, and
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2038produce a single value. The operands might represent
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2039multiple data, as is the case with the <a href="#t_vector">vector</a> data type.
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2040The result value of a binary operator is not
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2041necessarily the same type as its operands.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2042<p>There are several different binary operators:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2043</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2044<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2045<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
2046Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2047<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2048<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2049<pre> &lt;result&gt; = add &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2050</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2051<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2052<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2053<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2054<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2055 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2056 This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2057Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2058<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2059<p>The value produced is the integer or floating point sum of the two
2060operands.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2061<p>If an integer sum has unsigned overflow, the result returned is the
2062mathematical result modulo 2<sup>n</sup>, where n is the bit width of
2063the result.</p>
2064<p>Because LLVM integers use a two's complement representation, this
2065instruction is appropriate for both signed and unsigned integers.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2066<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2067<pre> &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2068</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2069</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2070<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2071<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
2072Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2073<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2074<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2075<pre> &lt;result&gt; = sub &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2076</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2077<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2078<p>The '<tt>sub</tt>' instruction returns the difference of its two
2079operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2080<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
2081instruction present in most other intermediate representations.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2082<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2083<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2084 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2085values.
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2086This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2087Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2088<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2089<p>The value produced is the integer or floating point difference of
2090the two operands.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2091<p>If an integer difference has unsigned overflow, the result returned is the
2092mathematical result modulo 2<sup>n</sup>, where n is the bit width of
2093the result.</p>
2094<p>Because LLVM integers use a two's complement representation, this
2095instruction is appropriate for both signed and unsigned integers.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2096<h5>Example:</h5>
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]2097<pre>
2098 &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2099 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2100</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2101</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2102<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2103<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
2104Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2105<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2106<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2107<pre> &lt;result&gt; = mul &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2108</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2109<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2110<p>The '<tt>mul</tt>' instruction returns the product of its two
2111operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2112<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2113<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2114 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2115values.
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2116This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]2117Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2118<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2119<p>The value produced is the integer or floating point product of the
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2120two operands.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2121<p>If the result of an integer multiplication has unsigned overflow,
2122the result returned is the mathematical result modulo
21232<sup>n</sup>, where n is the bit width of the result.</p>
2124<p>Because LLVM integers use a two's complement representation, and the
2125result is the same width as the operands, this instruction returns the
2126correct result for both signed and unsigned integers. If a full product
2127(e.g. <tt>i32</tt>x<tt>i32</tt>-><tt>i64</tt>) is needed, the operands
2128should be sign-extended or zero-extended as appropriate to the
2129width of the full product.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2130<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2131<pre> &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2132</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2133</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2134<!-- _______________________________________________________________________ -->
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2135<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
2136</a></div>
2137<div class="doc_text">
2138<h5>Syntax:</h5>
2139<pre> &lt;result&gt; = udiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2140</pre>
2141<h5>Overview:</h5>
2142<p>The '<tt>udiv</tt>' instruction returns the quotient of its two
2143operands.</p>
2144<h5>Arguments:</h5>
2145<p>The two arguments to the '<tt>udiv</tt>' instruction must be
2146<a href="#t_integer">integer</a> values. Both arguments must have identical
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2147types. This instruction can also take <a href="#t_vector">vector</a> versions
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2148of the values in which case the elements must be integers.</p>
2149<h5>Semantics:</h5>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2150<p>The value produced is the unsigned integer quotient of the two operands.</p>
2151<p>Note that unsigned integer division and signed integer division are distinct
2152operations; for signed integer division, use '<tt>sdiv</tt>'.</p>
2153<p>Division by zero leads to undefined behavior.</p>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2154<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2155<pre> &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2156</pre>
2157</div>
2158<!-- _______________________________________________________________________ -->
2159<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
2160</a> </div>
2161<div class="doc_text">
2162<h5>Syntax:</h5>
2163<pre> &lt;result&gt; = sdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2164</pre>
2165<h5>Overview:</h5>
2166<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two
2167operands.</p>
2168<h5>Arguments:</h5>
2169<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
2170<a href="#t_integer">integer</a> values. Both arguments must have identical
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2171types. This instruction can also take <a href="#t_vector">vector</a> versions
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2172of the values in which case the elements must be integers.</p>
2173<h5>Semantics:</h5>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2174<p>The value produced is the signed integer quotient of the two operands.</p>
2175<p>Note that signed integer division and unsigned integer division are distinct
2176operations; for unsigned integer division, use '<tt>udiv</tt>'.</p>
2177<p>Division by zero leads to undefined behavior. Overflow also leads to
2178undefined behavior; this is a rare case, but can occur, for example,
2179by doing a 32-bit division of -2147483648 by -1.</p>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2180<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2181<pre> &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2182</pre>
2183</div>
2184<!-- _______________________________________________________________________ -->
2185<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2186Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2187<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2188<h5>Syntax:</h5>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2189<pre> &lt;result&gt; = fdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2190</pre>
2191<h5>Overview:</h5>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2192<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2193operands.</p>
2194<h5>Arguments:</h5>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2195<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2196<a href="#t_floating">floating point</a> values. Both arguments must have
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2197identical types. This instruction can also take <a href="#t_vector">vector</a>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2198versions of floating point values.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2199<h5>Semantics:</h5>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2200<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2201<h5>Example:</h5>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2202<pre> &lt;result&gt; = fdiv float 4.0, %var <i>; yields {float}:result = 4.0 / %var</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2203</pre>
2204</div>
2205<!-- _______________________________________________________________________ -->
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2206<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
2207</div>
2208<div class="doc_text">
2209<h5>Syntax:</h5>
2210<pre> &lt;result&gt; = urem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2211</pre>
2212<h5>Overview:</h5>
2213<p>The '<tt>urem</tt>' instruction returns the remainder from the
2214unsigned division of its two arguments.</p>
2215<h5>Arguments:</h5>
2216<p>The two arguments to the '<tt>urem</tt>' instruction must be
2217<a href="#t_integer">integer</a> values. Both arguments must have identical
Dan Gohman80176312007-11-05 23:35:22 +0000[diff] [blame]2218types. This instruction can also take <a href="#t_vector">vector</a> versions
2219of the values in which case the elements must be integers.</p>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2220<h5>Semantics:</h5>
2221<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
2222This instruction always performs an unsigned division to get the remainder,
2223regardless of whether the arguments are unsigned or not.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2224<p>Note that unsigned integer remainder and signed integer remainder are
2225distinct operations; for signed integer remainder, use '<tt>srem</tt>'.</p>
2226<p>Taking the remainder of a division by zero leads to undefined behavior.</p>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2227<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2228<pre> &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2229</pre>
2230
2231</div>
2232<!-- _______________________________________________________________________ -->
2233<div class="doc_subsubsection"> <a name="i_srem">'<tt>srem</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2234Instruction</a> </div>
2235<div class="doc_text">
2236<h5>Syntax:</h5>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2237<pre> &lt;result&gt; = srem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2238</pre>
2239<h5>Overview:</h5>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2240<p>The '<tt>srem</tt>' instruction returns the remainder from the
Dan Gohman80176312007-11-05 23:35:22 +0000[diff] [blame]2241signed division of its two operands. This instruction can also take
2242<a href="#t_vector">vector</a> versions of the values in which case
2243the elements must be integers.</p>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]2244
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2245<h5>Arguments:</h5>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2246<p>The two arguments to the '<tt>srem</tt>' instruction must be
2247<a href="#t_integer">integer</a> values. Both arguments must have identical
2248types.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2249<h5>Semantics:</h5>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2250<p>This instruction returns the <i>remainder</i> of a division (where the result
Reid Spencerc9fdfc82007-03-24 22:23:39 +0000[diff] [blame]2251has the same sign as the dividend, <tt>var1</tt>), not the <i>modulo</i>
2252operator (where the result has the same sign as the divisor, <tt>var2</tt>) of
2253a value. For more information about the difference, see <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2254 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
Reid Spencerc9fdfc82007-03-24 22:23:39 +0000[diff] [blame]2255Math Forum</a>. For a table of how this is implemented in various languages,
Reid Spencer64f5c6c2007-03-24 22:40:44 +0000[diff] [blame]2256please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
Reid Spencerc9fdfc82007-03-24 22:23:39 +0000[diff] [blame]2257Wikipedia: modulo operation</a>.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2258<p>Note that signed integer remainder and unsigned integer remainder are
2259distinct operations; for unsigned integer remainder, use '<tt>urem</tt>'.</p>
2260<p>Taking the remainder of a division by zero leads to undefined behavior.
2261Overflow also leads to undefined behavior; this is a rare case, but can occur,
2262for example, by taking the remainder of a 32-bit division of -2147483648 by -1.
2263(The remainder doesn't actually overflow, but this rule lets srem be
2264implemented using instructions that return both the result of the division
2265and the remainder.)</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2266<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2267<pre> &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2268</pre>
2269
2270</div>
2271<!-- _______________________________________________________________________ -->
2272<div class="doc_subsubsection"> <a name="i_frem">'<tt>frem</tt>'
2273Instruction</a> </div>
2274<div class="doc_text">
2275<h5>Syntax:</h5>
2276<pre> &lt;result&gt; = frem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2277</pre>
2278<h5>Overview:</h5>
2279<p>The '<tt>frem</tt>' instruction returns the remainder from the
2280division of its two operands.</p>
2281<h5>Arguments:</h5>
2282<p>The two arguments to the '<tt>frem</tt>' instruction must be
2283<a href="#t_floating">floating point</a> values. Both arguments must have
Dan Gohman80176312007-11-05 23:35:22 +0000[diff] [blame]2284identical types. This instruction can also take <a href="#t_vector">vector</a>
2285versions of floating point values.</p>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2286<h5>Semantics:</h5>
2287<p>This instruction returns the <i>remainder</i> of a division.</p>
2288<h5>Example:</h5>
2289<pre> &lt;result&gt; = frem float 4.0, %var <i>; yields {float}:result = 4.0 % %var</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2290</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2291</div>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]2292
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]2293<!-- ======================================================================= -->
2294<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
2295Operations</a> </div>
2296<div class="doc_text">
2297<p>Bitwise binary operators are used to do various forms of
2298bit-twiddling in a program. They are generally very efficient
2299instructions and can commonly be strength reduced from other
2300instructions. They require two operands, execute an operation on them,
2301and produce a single value. The resulting value of the bitwise binary
2302operators is always the same type as its first operand.</p>
2303</div>
2304
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2305<!-- _______________________________________________________________________ -->
2306<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
2307Instruction</a> </div>
2308<div class="doc_text">
2309<h5>Syntax:</h5>
2310<pre> &lt;result&gt; = shl &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2311</pre>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2312
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2313<h5>Overview:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2314
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2315<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
2316the left a specified number of bits.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2317
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2318<h5>Arguments:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2319
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2320<p>Both arguments to the '<tt>shl</tt>' instruction must be the same <a
2321 href="#t_integer">integer</a> type.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2322
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2323<h5>Semantics:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2324
2325<p>The value produced is <tt>var1</tt> * 2<sup><tt>var2</tt></sup>. If
2326<tt>var2</tt> is (statically or dynamically) equal to or larger than the number
2327of bits in <tt>var1</tt>, the result is undefined.</p>
2328
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2329<h5>Example:</h5><pre>
2330 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
2331 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
2332 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2333 &lt;result&gt; = shl i32 1, 32 <i>; undefined</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2334</pre>
2335</div>
2336<!-- _______________________________________________________________________ -->
2337<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
2338Instruction</a> </div>
2339<div class="doc_text">
2340<h5>Syntax:</h5>
2341<pre> &lt;result&gt; = lshr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2342</pre>
2343
2344<h5>Overview:</h5>
2345<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2346operand shifted to the right a specified number of bits with zero fill.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2347
2348<h5>Arguments:</h5>
2349<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
2350<a href="#t_integer">integer</a> type.</p>
2351
2352<h5>Semantics:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2353
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2354<p>This instruction always performs a logical shift right operation. The most
2355significant bits of the result will be filled with zero bits after the
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2356shift. If <tt>var2</tt> is (statically or dynamically) equal to or larger than
2357the number of bits in <tt>var1</tt>, the result is undefined.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2358
2359<h5>Example:</h5>
2360<pre>
2361 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
2362 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
2363 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
2364 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2365 &lt;result&gt; = lshr i32 1, 32 <i>; undefined</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2366</pre>
2367</div>
2368
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]2369<!-- _______________________________________________________________________ -->
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2370<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
2371Instruction</a> </div>
2372<div class="doc_text">
2373
2374<h5>Syntax:</h5>
2375<pre> &lt;result&gt; = ashr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2376</pre>
2377
2378<h5>Overview:</h5>
2379<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2380operand shifted to the right a specified number of bits with sign extension.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2381
2382<h5>Arguments:</h5>
2383<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
2384<a href="#t_integer">integer</a> type.</p>
2385
2386<h5>Semantics:</h5>
2387<p>This instruction always performs an arithmetic shift right operation,
2388The most significant bits of the result will be filled with the sign bit
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2389of <tt>var1</tt>. If <tt>var2</tt> is (statically or dynamically) equal to or
2390larger than the number of bits in <tt>var1</tt>, the result is undefined.
2391</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2392
2393<h5>Example:</h5>
2394<pre>
2395 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
2396 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
2397 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
2398 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2399 &lt;result&gt; = ashr i32 1, 32 <i>; undefined</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2400</pre>
2401</div>
2402
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2403<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2404<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
2405Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2406<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2407<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2408<pre> &lt;result&gt; = and &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2409</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2410<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2411<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
2412its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2413<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2414<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]2415 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2416identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2417<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2418<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2419<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2420<div style="align: center">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2421<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2422 <tbody>
2423 <tr>
2424 <td>In0</td>
2425 <td>In1</td>
2426 <td>Out</td>
2427 </tr>
2428 <tr>
2429 <td>0</td>
2430 <td>0</td>
2431 <td>0</td>
2432 </tr>
2433 <tr>
2434 <td>0</td>
2435 <td>1</td>
2436 <td>0</td>
2437 </tr>
2438 <tr>
2439 <td>1</td>
2440 <td>0</td>
2441 <td>0</td>
2442 </tr>
2443 <tr>
2444 <td>1</td>
2445 <td>1</td>
2446 <td>1</td>
2447 </tr>
2448 </tbody>
2449</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2450</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2451<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2452<pre> &lt;result&gt; = and i32 4, %var <i>; yields {i32}:result = 4 &amp; %var</i>
2453 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
2454 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2455</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2456</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2457<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2458<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2459<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2460<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2461<pre> &lt;result&gt; = or &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2462</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2463<h5>Overview:</h5>
2464<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
2465or of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2466<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2467<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]2468 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2469identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2470<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2471<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2472<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2473<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2474<table border="1" cellspacing="0" cellpadding="4">
2475 <tbody>
2476 <tr>
2477 <td>In0</td>
2478 <td>In1</td>
2479 <td>Out</td>
2480 </tr>
2481 <tr>
2482 <td>0</td>
2483 <td>0</td>
2484 <td>0</td>
2485 </tr>
2486 <tr>
2487 <td>0</td>
2488 <td>1</td>
2489 <td>1</td>
2490 </tr>
2491 <tr>
2492 <td>1</td>
2493 <td>0</td>
2494 <td>1</td>
2495 </tr>
2496 <tr>
2497 <td>1</td>
2498 <td>1</td>
2499 <td>1</td>
2500 </tr>
2501 </tbody>
2502</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2503</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2504<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2505<pre> &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
2506 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
2507 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2508</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2509</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2510<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2511<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
2512Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2513<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2514<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2515<pre> &lt;result&gt; = xor &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2516</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2517<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2518<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
2519or of its two operands. The <tt>xor</tt> is used to implement the
2520"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2521<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2522<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]2523 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2524identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2525<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2526<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2527<p> </p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2528<div style="align: center">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2529<table border="1" cellspacing="0" cellpadding="4">
2530 <tbody>
2531 <tr>
2532 <td>In0</td>
2533 <td>In1</td>
2534 <td>Out</td>
2535 </tr>
2536 <tr>
2537 <td>0</td>
2538 <td>0</td>
2539 <td>0</td>
2540 </tr>
2541 <tr>
2542 <td>0</td>
2543 <td>1</td>
2544 <td>1</td>
2545 </tr>
2546 <tr>
2547 <td>1</td>
2548 <td>0</td>
2549 <td>1</td>
2550 </tr>
2551 <tr>
2552 <td>1</td>
2553 <td>1</td>
2554 <td>0</td>
2555 </tr>
2556 </tbody>
2557</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2558</div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2559<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2560<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2561<pre> &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
2562 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
2563 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
2564 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2565</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2566</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2567
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2568<!-- ======================================================================= -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2569<div class="doc_subsection">
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2570 <a name="vectorops">Vector Operations</a>
2571</div>
2572
2573<div class="doc_text">
2574
2575<p>LLVM supports several instructions to represent vector operations in a
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2576target-independent manner. These instructions cover the element-access and
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2577vector-specific operations needed to process vectors effectively. While LLVM
2578does directly support these vector operations, many sophisticated algorithms
2579will want to use target-specific intrinsics to take full advantage of a specific
2580target.</p>
2581
2582</div>
2583
2584<!-- _______________________________________________________________________ -->
2585<div class="doc_subsubsection">
2586 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
2587</div>
2588
2589<div class="doc_text">
2590
2591<h5>Syntax:</h5>
2592
2593<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2594 &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 Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2595</pre>
2596
2597<h5>Overview:</h5>
2598
2599<p>
2600The '<tt>extractelement</tt>' instruction extracts a single scalar
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2601element from a vector at a specified index.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2602</p>
2603
2604
2605<h5>Arguments:</h5>
2606
2607<p>
2608The first operand of an '<tt>extractelement</tt>' instruction is a
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2609value of <a href="#t_vector">vector</a> type. The second operand is
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2610an index indicating the position from which to extract the element.
2611The index may be a variable.</p>
2612
2613<h5>Semantics:</h5>
2614
2615<p>
2616The result is a scalar of the same type as the element type of
2617<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
2618<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
2619results are undefined.
2620</p>
2621
2622<h5>Example:</h5>
2623
2624<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2625 %result = extractelement &lt;4 x i32&gt; %vec, i32 0 <i>; yields i32</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2626</pre>
2627</div>
2628
2629
2630<!-- _______________________________________________________________________ -->
2631<div class="doc_subsubsection">
2632 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
2633</div>
2634
2635<div class="doc_text">
2636
2637<h5>Syntax:</h5>
2638
2639<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2640 &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 Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2641</pre>
2642
2643<h5>Overview:</h5>
2644
2645<p>
2646The '<tt>insertelement</tt>' instruction inserts a scalar
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2647element into a vector at a specified index.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2648</p>
2649
2650
2651<h5>Arguments:</h5>
2652
2653<p>
2654The first operand of an '<tt>insertelement</tt>' instruction is a
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2655value of <a href="#t_vector">vector</a> type. The second operand is a
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2656scalar value whose type must equal the element type of the first
2657operand. The third operand is an index indicating the position at
2658which to insert the value. The index may be a variable.</p>
2659
2660<h5>Semantics:</h5>
2661
2662<p>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2663The result is a vector of the same type as <tt>val</tt>. Its
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2664element values are those of <tt>val</tt> except at position
2665<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
2666exceeds the length of <tt>val</tt>, the results are undefined.
2667</p>
2668
2669<h5>Example:</h5>
2670
2671<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2672 %result = insertelement &lt;4 x i32&gt; %vec, i32 1, i32 0 <i>; yields &lt;4 x i32&gt;</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2673</pre>
2674</div>
2675
2676<!-- _______________________________________________________________________ -->
2677<div class="doc_subsubsection">
2678 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
2679</div>
2680
2681<div class="doc_text">
2682
2683<h5>Syntax:</h5>
2684
2685<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2686 &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 Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2687</pre>
2688
2689<h5>Overview:</h5>
2690
2691<p>
2692The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
2693from two input vectors, returning a vector of the same type.
2694</p>
2695
2696<h5>Arguments:</h5>
2697
2698<p>
2699The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
2700with types that match each other and types that match the result of the
2701instruction. The third argument is a shuffle mask, which has the same number
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2702of elements as the other vector type, but whose element type is always 'i32'.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2703</p>
2704
2705<p>
2706The shuffle mask operand is required to be a constant vector with either
2707constant integer or undef values.
2708</p>
2709
2710<h5>Semantics:</h5>
2711
2712<p>
2713The elements of the two input vectors are numbered from left to right across
2714both of the vectors. The shuffle mask operand specifies, for each element of
2715the result vector, which element of the two input registers the result element
2716gets. The element selector may be undef (meaning "don't care") and the second
2717operand may be undef if performing a shuffle from only one vector.
2718</p>
2719
2720<h5>Example:</h5>
2721
2722<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2723 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2724 &lt;4 x i32&gt; &lt;i32 0, i32 4, i32 1, i32 5&gt; <i>; yields &lt;4 x i32&gt;</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2725 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; undef,
2726 &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 Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2727</pre>
2728</div>
2729
Tanya Lattner09474292006-04-14 19:24:33 +0000[diff] [blame]2730
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2731<!-- ======================================================================= -->
2732<div class="doc_subsection">
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]2733 <a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2734</div>
2735
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2736<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2737
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2738<p>A key design point of an SSA-based representation is how it
2739represents memory. In LLVM, no memory locations are in SSA form, which
2740makes things very simple. This section describes how to read, write,
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2741allocate, and free memory in LLVM.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2742
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2743</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2744
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2745<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2746<div class="doc_subsubsection">
2747 <a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
2748</div>
2749
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2750<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2751
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2752<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2753
2754<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2755 &lt;result&gt; = malloc &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2756</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2757
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2758<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2759
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2760<p>The '<tt>malloc</tt>' instruction allocates memory from the system
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]2761heap and returns a pointer to it. The object is always allocated in the generic
2762address space (address space zero).</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2763
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2764<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2765
2766<p>The '<tt>malloc</tt>' instruction allocates
2767<tt>sizeof(&lt;type&gt;)*NumElements</tt>
John Criswell6e4ca612004-02-24 16:13:56 +0000[diff] [blame]2768bytes of memory from the operating system and returns a pointer of the
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2769appropriate type to the program. If "NumElements" is specified, it is the
Gabor Greif1acd2ee2008-02-09 22:24:34 +0000[diff] [blame]2770number of elements allocated, otherwise "NumElements" is defaulted to be one.
2771If an alignment is specified, the value result of the allocation is guaranteed to
2772be aligned to at least that boundary. If not specified, or if zero, the target can
2773choose to align the allocation on any convenient boundary.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2774
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2775<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2776
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2777<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2778
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2779<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
2780a pointer is returned.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2781
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2782<h5>Example:</h5>
2783
2784<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2785 %array = malloc [4 x i8 ] <i>; yields {[%4 x i8]*}:array</i>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2786
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]2787 %size = <a href="#i_add">add</a> i32 2, 2 <i>; yields {i32}:size = i32 4</i>
2788 %array1 = malloc i8, i32 4 <i>; yields {i8*}:array1</i>
2789 %array2 = malloc [12 x i8], i32 %size <i>; yields {[12 x i8]*}:array2</i>
2790 %array3 = malloc i32, i32 4, align 1024 <i>; yields {i32*}:array3</i>
2791 %array4 = malloc i32, align 1024 <i>; yields {i32*}:array4</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2792</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2793</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2794
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2795<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2796<div class="doc_subsubsection">
2797 <a name="i_free">'<tt>free</tt>' Instruction</a>
2798</div>
2799
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2800<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2801
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2802<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2803
2804<pre>
2805 free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2806</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2807
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2808<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2809
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2810<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]2811memory heap to be reallocated in the future.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2812
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2813<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2814
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2815<p>'<tt>value</tt>' shall be a pointer value that points to a value
2816that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
2817instruction.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2818
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2819<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2820
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2821<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2822after this instruction executes.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2823
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2824<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2825
2826<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2827 %array = <a href="#i_malloc">malloc</a> [4 x i8] <i>; yields {[4 x i8]*}:array</i>
2828 free [4 x i8]* %array
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2829</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2830</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2831
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2832<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2833<div class="doc_subsubsection">
2834 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
2835</div>
2836
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2837<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2838
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2839<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2840
2841<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2842 &lt;result&gt; = alloca &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2843</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2844
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2845<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2846
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2847<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
2848currently executing function, to be automatically released when this function
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]2849returns to its caller. The object is always allocated in the generic address
2850space (address space zero).</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2851
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2852<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2853
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2854<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2855bytes of memory on the runtime stack, returning a pointer of the
Gabor Greif1acd2ee2008-02-09 22:24:34 +0000[diff] [blame]2856appropriate type to the program. If "NumElements" is specified, it is the
2857number of elements allocated, otherwise "NumElements" is defaulted to be one.
2858If an alignment is specified, the value result of the allocation is guaranteed
2859to be aligned to at least that boundary. If not specified, or if zero, the target
2860can choose to align the allocation on any convenient boundary.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2861
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2862<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2863
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2864<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2865
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]2866<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2867memory is automatically released when the function returns. The '<tt>alloca</tt>'
2868instruction is commonly used to represent automatic variables that must
2869have an address available. When the function returns (either with the <tt><a
John Criswelldae2e932005-05-12 16:55:34 +0000[diff] [blame]2870 href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2871instructions), the memory is reclaimed.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2872
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2873<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2874
2875<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2876 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2877 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
2878 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2879 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2880</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2881</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2882
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2883<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2884<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
2885Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2886<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2887<h5>Syntax:</h5>
Christopher Lamb2330e4d2007-04-21 08:16:25 +0000[diff] [blame]2888<pre> &lt;result&gt; = load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;]<br> &lt;result&gt; = volatile load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;]<br></pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2889<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2890<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2891<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2892<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]2893address from which to load. The pointer must point to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]2894 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]2895marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2896the number or order of execution of this <tt>load</tt> with other
2897volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
2898instructions. </p>
Chris Lattnera31d1d72008-01-06 21:04:43 +0000[diff] [blame]2899<p>
2900The optional "align" argument specifies the alignment of the operation
2901(that is, the alignment of the memory address). A value of 0 or an
2902omitted "align" argument means that the operation has the preferential
2903alignment for the target. It is the responsibility of the code emitter
2904to ensure that the alignment information is correct. Overestimating
2905the alignment results in an undefined behavior. Underestimating the
2906alignment may produce less efficient code. An alignment of 1 is always
2907safe.
2908</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2909<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2910<p>The location of memory pointed to is loaded.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2911<h5>Examples:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2912<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2913 <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2914 href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
2915 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2916</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2917</div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2918<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2919<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
2920Instruction</a> </div>
Reid Spencer035ab572006-11-09 21:18:01 +0000[diff] [blame]2921<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2922<h5>Syntax:</h5>
Christopher Lamb2330e4d2007-04-21 08:16:25 +0000[diff] [blame]2923<pre> store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
2924 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2925</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2926<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2927<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2928<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2929<p>There are two arguments to the '<tt>store</tt>' instruction: a value
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2930to store and an address at which to store it. The type of the '<tt>&lt;pointer&gt;</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2931operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]2932operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2933optimizer is not allowed to modify the number or order of execution of
2934this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
2935 href="#i_store">store</a></tt> instructions.</p>
Chris Lattnera31d1d72008-01-06 21:04:43 +0000[diff] [blame]2936<p>
2937The optional "align" argument specifies the alignment of the operation
2938(that is, the alignment of the memory address). A value of 0 or an
2939omitted "align" argument means that the operation has the preferential
2940alignment for the target. It is the responsibility of the code emitter
2941to ensure that the alignment information is correct. Overestimating
2942the alignment results in an undefined behavior. Underestimating the
2943alignment may produce less efficient code. An alignment of 1 is always
2944safe.
2945</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2946<h5>Semantics:</h5>
2947<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
2948at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2949<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2950<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling8c6c72d2007-10-22 05:10:05 +0000[diff] [blame]2951 store i32 3, i32* %ptr <i>; yields {void}</i>
2952 %val = <a href="#i_load">load</a> i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2953</pre>
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]2954</div>
2955
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]2956<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2957<div class="doc_subsubsection">
2958 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
2959</div>
2960
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2961<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2962<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2963<pre>
2964 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
2965</pre>
2966
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2967<h5>Overview:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2968
2969<p>
2970The '<tt>getelementptr</tt>' instruction is used to get the address of a
2971subelement of an aggregate data structure.</p>
2972
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2973<h5>Arguments:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2974
Reid Spencer85f5b5b2006-12-04 21:29:24 +0000[diff] [blame]2975<p>This instruction takes a list of integer operands that indicate what
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2976elements of the aggregate object to index to. The actual types of the arguments
2977provided depend on the type of the first pointer argument. The
2978'<tt>getelementptr</tt>' instruction is used to index down through the type
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]2979levels of a structure or to a specific index in an array. When indexing into a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2980structure, only <tt>i32</tt> integer constants are allowed. When indexing
Reid Spencer85f5b5b2006-12-04 21:29:24 +0000[diff] [blame]2981into an array or pointer, only integers of 32 or 64 bits are allowed, and will
2982be sign extended to 64-bit values.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2983
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2984<p>For example, let's consider a C code fragment and how it gets
2985compiled to LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2986
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2987<div class="doc_code">
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2988<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2989struct RT {
2990 char A;
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]2991 int B[10][20];
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2992 char C;
2993};
2994struct ST {
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]2995 int X;
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2996 double Y;
2997 struct RT Z;
2998};
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]2999
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]3000int *foo(struct ST *s) {
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3001 return &amp;s[1].Z.B[5][13];
3002}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3003</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3004</div>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3005
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3006<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3007
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3008<div class="doc_code">
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3009<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3010%RT = type { i8 , [10 x [20 x i32]], i8 }
3011%ST = type { i32, double, %RT }
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3012
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3013define i32* %foo(%ST* %s) {
3014entry:
3015 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
3016 ret i32* %reg
3017}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3018</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3019</div>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3020
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3021<h5>Semantics:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3022
3023<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]3024on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
Reid Spencer85f5b5b2006-12-04 21:29:24 +0000[diff] [blame]3025and <a href="#t_array">array</a> types can use a 32-bit or 64-bit
Reid Spencer42ddd842006-12-03 16:53:48 +0000[diff] [blame]3026<a href="#t_integer">integer</a> type but the value will always be sign extended
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3027to 64-bits. <a href="#t_struct">Structure</a> types require <tt>i32</tt>
Reid Spencer42ddd842006-12-03 16:53:48 +0000[diff] [blame]3028<b>constants</b>.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3029
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3030<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3031type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3032}</tt>' type, a structure. The second index indexes into the third element of
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3033the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
3034i8 }</tt>' type, another structure. The third index indexes into the second
3035element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3036array. The two dimensions of the array are subscripted into, yielding an
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3037'<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
3038to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3039
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3040<p>Note that it is perfectly legal to index partially through a
3041structure, returning a pointer to an inner element. Because of this,
3042the LLVM code for the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3043
3044<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3045 define i32* %foo(%ST* %s) {
3046 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3047 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
3048 %t3 = getelementptr %RT* %t2, i32 0, i32 1 <i>; yields [10 x [20 x i32]]*:%t3</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3049 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
3050 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
3051 ret i32* %t5
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3052 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3053</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]3054
3055<p>Note that it is undefined to access an array out of bounds: array and
3056pointer indexes must always be within the defined bounds of the array type.
3057The one exception for this rules is zero length arrays. These arrays are
3058defined to be accessible as variable length arrays, which requires access
3059beyond the zero'th element.</p>
3060
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]3061<p>The getelementptr instruction is often confusing. For some more insight
3062into how it works, see <a href="GetElementPtr.html">the getelementptr
3063FAQ</a>.</p>
3064
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3065<h5>Example:</h5>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]3066
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3067<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3068 <i>; yields [12 x i8]*:aptr</i>
3069 %aptr = getelementptr {i32, [12 x i8]}* %sptr, i64 0, i32 1
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3070</pre>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3071</div>
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]3072
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3073<!-- ======================================================================= -->
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3074<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3075</div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3076<div class="doc_text">
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3077<p>The instructions in this category are the conversion instructions (casting)
3078which all take a single operand and a type. They perform various bit conversions
3079on the operand.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3080</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3081
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3082<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3083<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3084 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
3085</div>
3086<div class="doc_text">
3087
3088<h5>Syntax:</h5>
3089<pre>
3090 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3091</pre>
3092
3093<h5>Overview:</h5>
3094<p>
3095The '<tt>trunc</tt>' instruction truncates its operand to the type <tt>ty2</tt>.
3096</p>
3097
3098<h5>Arguments:</h5>
3099<p>
3100The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
3101be an <a href="#t_integer">integer</a> type, and a type that specifies the size
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3102and type of the result, which must be an <a href="#t_integer">integer</a>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3103type. The bit size of <tt>value</tt> must be larger than the bit size of
3104<tt>ty2</tt>. Equal sized types are not allowed.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3105
3106<h5>Semantics:</h5>
3107<p>
3108The '<tt>trunc</tt>' instruction truncates the high order bits in <tt>value</tt>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3109and converts the remaining bits to <tt>ty2</tt>. Since the source size must be
3110larger than the destination size, <tt>trunc</tt> cannot be a <i>no-op cast</i>.
3111It will always truncate bits.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3112
3113<h5>Example:</h5>
3114<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3115 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3116 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
3117 %Y = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3118</pre>
3119</div>
3120
3121<!-- _______________________________________________________________________ -->
3122<div class="doc_subsubsection">
3123 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
3124</div>
3125<div class="doc_text">
3126
3127<h5>Syntax:</h5>
3128<pre>
3129 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3130</pre>
3131
3132<h5>Overview:</h5>
3133<p>The '<tt>zext</tt>' instruction zero extends its operand to type
3134<tt>ty2</tt>.</p>
3135
3136
3137<h5>Arguments:</h5>
3138<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3139<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3140also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3141<tt>value</tt> must be smaller than the bit size of the destination type,
3142<tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3143
3144<h5>Semantics:</h5>
3145<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
Chris Lattnerd1d25172007-05-24 19:13:27 +0000[diff] [blame]3146bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3147
Reid Spencerb5929522007-01-12 15:46:11 +0000[diff] [blame]3148<p>When zero extending from i1, the result will always be either 0 or 1.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3149
3150<h5>Example:</h5>
3151<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3152 %X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3153 %Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3154</pre>
3155</div>
3156
3157<!-- _______________________________________________________________________ -->
3158<div class="doc_subsubsection">
3159 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
3160</div>
3161<div class="doc_text">
3162
3163<h5>Syntax:</h5>
3164<pre>
3165 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3166</pre>
3167
3168<h5>Overview:</h5>
3169<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
3170
3171<h5>Arguments:</h5>
3172<p>
3173The '<tt>sext</tt>' instruction takes a value to cast, which must be of
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3174<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3175also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3176<tt>value</tt> must be smaller than the bit size of the destination type,
3177<tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3178
3179<h5>Semantics:</h5>
3180<p>
3181The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
3182bit (highest order bit) of the <tt>value</tt> until it reaches the bit size of
Chris Lattnerd1d25172007-05-24 19:13:27 +0000[diff] [blame]3183the type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3184
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3185<p>When sign extending from i1, the extension always results in -1 or 0.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3186
3187<h5>Example:</h5>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3188<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3189 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3190 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3191</pre>
3192</div>
3193
3194<!-- _______________________________________________________________________ -->
3195<div class="doc_subsubsection">
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]3196 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
3197</div>
3198
3199<div class="doc_text">
3200
3201<h5>Syntax:</h5>
3202
3203<pre>
3204 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3205</pre>
3206
3207<h5>Overview:</h5>
3208<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
3209<tt>ty2</tt>.</p>
3210
3211
3212<h5>Arguments:</h5>
3213<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
3214 point</a> value to cast and a <a href="#t_floating">floating point</a> type to
3215cast it to. The size of <tt>value</tt> must be larger than the size of
3216<tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
3217<i>no-op cast</i>.</p>
3218
3219<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3220<p> The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
3221<a href="#t_floating">floating point</a> type to a smaller
3222<a href="#t_floating">floating point</a> type. If the value cannot fit within
3223the destination type, <tt>ty2</tt>, then the results are undefined.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]3224
3225<h5>Example:</h5>
3226<pre>
3227 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
3228 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
3229</pre>
3230</div>
3231
3232<!-- _______________________________________________________________________ -->
3233<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3234 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
3235</div>
3236<div class="doc_text">
3237
3238<h5>Syntax:</h5>
3239<pre>
3240 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3241</pre>
3242
3243<h5>Overview:</h5>
3244<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
3245floating point value.</p>
3246
3247<h5>Arguments:</h5>
3248<p>The '<tt>fpext</tt>' instruction takes a
3249<a href="#t_floating">floating point</a> <tt>value</tt> to cast,
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3250and a <a href="#t_floating">floating point</a> type to cast it to. The source
3251type must be smaller than the destination type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3252
3253<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3254<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]3255<a href="#t_floating">floating point</a> type to a larger
3256<a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3257used to make a <i>no-op cast</i> because it always changes bits. Use
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3258<tt>bitcast</tt> to make a <i>no-op cast</i> for a floating point cast.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3259
3260<h5>Example:</h5>
3261<pre>
3262 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
3263 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
3264</pre>
3265</div>
3266
3267<!-- _______________________________________________________________________ -->
3268<div class="doc_subsubsection">
Reid Spencer24d6da52007-01-21 00:29:26 +0000[diff] [blame]3269 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3270</div>
3271<div class="doc_text">
3272
3273<h5>Syntax:</h5>
3274<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3275 &lt;result&gt; = fptoui &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3276</pre>
3277
3278<h5>Overview:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3279<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3280unsigned integer equivalent of type <tt>ty2</tt>.
3281</p>
3282
3283<h5>Arguments:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3284<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3285scalar or vector <a href="#t_floating">floating point</a> value, and a type
3286to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3287type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3288vector integer type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3289
3290<h5>Semantics:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3291<p> The '<tt>fptoui</tt>' instruction converts its
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3292<a href="#t_floating">floating point</a> operand into the nearest (rounding
3293towards zero) unsigned integer value. If the value cannot fit in <tt>ty2</tt>,
3294the results are undefined.</p>
3295
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3296<h5>Example:</h5>
3297<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3298 %X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]3299 %Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3300 %X = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3301</pre>
3302</div>
3303
3304<!-- _______________________________________________________________________ -->
3305<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3306 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3307</div>
3308<div class="doc_text">
3309
3310<h5>Syntax:</h5>
3311<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3312 &lt;result&gt; = fptosi &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3313</pre>
3314
3315<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3316<p>The '<tt>fptosi</tt>' instruction converts
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3317<a href="#t_floating">floating point</a> <tt>value</tt> to type <tt>ty2</tt>.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3318</p>
3319
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3320<h5>Arguments:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3321<p> The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3322scalar or vector <a href="#t_floating">floating point</a> value, and a type
3323to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3324type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3325vector integer type with the same number of elements as <tt>ty</tt></p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3326
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3327<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3328<p>The '<tt>fptosi</tt>' instruction converts its
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3329<a href="#t_floating">floating point</a> operand into the nearest (rounding
3330towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
3331the results are undefined.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3332
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]3333<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3334<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3335 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]3336 %Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3337 %X = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3338</pre>
3339</div>
3340
3341<!-- _______________________________________________________________________ -->
3342<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3343 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3344</div>
3345<div class="doc_text">
3346
3347<h5>Syntax:</h5>
3348<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3349 &lt;result&gt; = uitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3350</pre>
3351
3352<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3353<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3354integer and converts that value to the <tt>ty2</tt> type.</p>
3355
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3356<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3357<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be a
3358scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3359to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3360type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3361floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3362
3363<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3364<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3365integer quantity and converts it to the corresponding floating point value. If
Jeff Cohencb757312007-04-22 14:56:37 +0000[diff] [blame]3366the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3367
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3368<h5>Example:</h5>
3369<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3370 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3371 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3372</pre>
3373</div>
3374
3375<!-- _______________________________________________________________________ -->
3376<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3377 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3378</div>
3379<div class="doc_text">
3380
3381<h5>Syntax:</h5>
3382<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3383 &lt;result&gt; = sitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3384</pre>
3385
3386<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3387<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3388integer and converts that value to the <tt>ty2</tt> type.</p>
3389
3390<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3391<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be a
3392scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3393to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3394type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3395floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3396
3397<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3398<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3399integer quantity and converts it to the corresponding floating point value. If
Jeff Cohencb757312007-04-22 14:56:37 +0000[diff] [blame]3400the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3401
3402<h5>Example:</h5>
3403<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3404 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3405 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3406</pre>
3407</div>
3408
3409<!-- _______________________________________________________________________ -->
3410<div class="doc_subsubsection">
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3411 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
3412</div>
3413<div class="doc_text">
3414
3415<h5>Syntax:</h5>
3416<pre>
3417 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3418</pre>
3419
3420<h5>Overview:</h5>
3421<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
3422the integer type <tt>ty2</tt>.</p>
3423
3424<h5>Arguments:</h5>
3425<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]3426must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3427<tt>ty2</tt>, which must be an <a href="#t_integer">integer</a> type.
3428
3429<h5>Semantics:</h5>
3430<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
3431<tt>ty2</tt> by interpreting the pointer value as an integer and either
3432truncating or zero extending that value to the size of the integer type. If
3433<tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
3434<tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3435are the same size, then nothing is done (<i>no-op cast</i>) other than a type
3436change.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3437
3438<h5>Example:</h5>
3439<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3440 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
3441 %Y = ptrtoint i32* %x to i64 <i>; yields zero extension on 32-bit architecture</i>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3442</pre>
3443</div>
3444
3445<!-- _______________________________________________________________________ -->
3446<div class="doc_subsubsection">
3447 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
3448</div>
3449<div class="doc_text">
3450
3451<h5>Syntax:</h5>
3452<pre>
3453 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3454</pre>
3455
3456<h5>Overview:</h5>
3457<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to
3458a pointer type, <tt>ty2</tt>.</p>
3459
3460<h5>Arguments:</h5>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]3461<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3462value to cast, and a type to cast it to, which must be a
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]3463<a href="#t_pointer">pointer</a> type.
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3464
3465<h5>Semantics:</h5>
3466<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
3467<tt>ty2</tt> by applying either a zero extension or a truncation depending on
3468the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
3469size of a pointer then a truncation is done. If <tt>value</tt> is smaller than
3470the size of a pointer then a zero extension is done. If they are the same size,
3471nothing is done (<i>no-op cast</i>).</p>
3472
3473<h5>Example:</h5>
3474<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3475 %X = inttoptr i32 255 to i32* <i>; yields zero extension on 64-bit architecture</i>
3476 %X = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
3477 %Y = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3478</pre>
3479</div>
3480
3481<!-- _______________________________________________________________________ -->
3482<div class="doc_subsubsection">
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3483 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3484</div>
3485<div class="doc_text">
3486
3487<h5>Syntax:</h5>
3488<pre>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3489 &lt;result&gt; = bitcast &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3490</pre>
3491
3492<h5>Overview:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3493<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3494<tt>ty2</tt> without changing any bits.</p>
3495
3496<h5>Arguments:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3497<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3498a first class value, and a type to cast it to, which must also be a <a
3499 href="#t_firstclass">first class</a> type. The bit sizes of <tt>value</tt>
Reid Spencer19b569f2007-01-09 20:08:58 +0000[diff] [blame]3500and the destination type, <tt>ty2</tt>, must be identical. If the source
3501type is a pointer, the destination type must also be a pointer.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3502
3503<h5>Semantics:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3504<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3505<tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
3506this conversion. The conversion is done as if the <tt>value</tt> had been
3507stored to memory and read back as type <tt>ty2</tt>. Pointer types may only be
3508converted to other pointer types with this instruction. To convert pointers to
3509other types, use the <a href="#i_inttoptr">inttoptr</a> or
3510<a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3511
3512<h5>Example:</h5>
3513<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3514 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3515 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
3516 %Z = bitcast <2xint> %V to i64; <i>; yields i64: %V</i>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]3517</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3518</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3519
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3520<!-- ======================================================================= -->
3521<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
3522<div class="doc_text">
3523<p>The instructions in this category are the "miscellaneous"
3524instructions, which defy better classification.</p>
3525</div>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3526
3527<!-- _______________________________________________________________________ -->
3528<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
3529</div>
3530<div class="doc_text">
3531<h5>Syntax:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3532<pre> &lt;result&gt; = icmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {i1}:result</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3533</pre>
3534<h5>Overview:</h5>
3535<p>The '<tt>icmp</tt>' instruction returns a boolean value based on comparison
3536of its two integer operands.</p>
3537<h5>Arguments:</h5>
3538<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3539the condition code indicating the kind of comparison to perform. It is not
3540a value, just a keyword. The possible condition code are:
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3541<ol>
3542 <li><tt>eq</tt>: equal</li>
3543 <li><tt>ne</tt>: not equal </li>
3544 <li><tt>ugt</tt>: unsigned greater than</li>
3545 <li><tt>uge</tt>: unsigned greater or equal</li>
3546 <li><tt>ult</tt>: unsigned less than</li>
3547 <li><tt>ule</tt>: unsigned less or equal</li>
3548 <li><tt>sgt</tt>: signed greater than</li>
3549 <li><tt>sge</tt>: signed greater or equal</li>
3550 <li><tt>slt</tt>: signed less than</li>
3551 <li><tt>sle</tt>: signed less or equal</li>
3552</ol>
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3553<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Reid Spencer350f8aa2007-01-04 05:19:58 +0000[diff] [blame]3554<a href="#t_pointer">pointer</a> typed. They must also be identical types.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3555<h5>Semantics:</h5>
3556<p>The '<tt>icmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3557the condition code given as <tt>cond</tt>. The comparison performed always
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3558yields a <a href="#t_primitive">i1</a> result, as follows:
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3559<ol>
3560 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
3561 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3562 </li>
3563 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
3564 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3565 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
3566 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3567 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
3568 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3569 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
3570 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3571 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
3572 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3573 <li><tt>sgt</tt>: interprets the operands as signed values and yields
3574 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3575 <li><tt>sge</tt>: interprets the operands as signed values and yields
3576 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3577 <li><tt>slt</tt>: interprets the operands as signed values and yields
3578 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3579 <li><tt>sle</tt>: interprets the operands as signed values and yields
3580 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3581</ol>
3582<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3583values are compared as if they were integers.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3584
3585<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3586<pre> &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
3587 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
3588 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
3589 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
3590 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
3591 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3592</pre>
3593</div>
3594
3595<!-- _______________________________________________________________________ -->
3596<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
3597</div>
3598<div class="doc_text">
3599<h5>Syntax:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3600<pre> &lt;result&gt; = fcmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {i1}:result</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3601</pre>
3602<h5>Overview:</h5>
3603<p>The '<tt>fcmp</tt>' instruction returns a boolean value based on comparison
3604of its floating point operands.</p>
3605<h5>Arguments:</h5>
3606<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3607the condition code indicating the kind of comparison to perform. It is not
3608a value, just a keyword. The possible condition code are:
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3609<ol>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3610 <li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3611 <li><tt>oeq</tt>: ordered and equal</li>
3612 <li><tt>ogt</tt>: ordered and greater than </li>
3613 <li><tt>oge</tt>: ordered and greater than or equal</li>
3614 <li><tt>olt</tt>: ordered and less than </li>
3615 <li><tt>ole</tt>: ordered and less than or equal</li>
3616 <li><tt>one</tt>: ordered and not equal</li>
3617 <li><tt>ord</tt>: ordered (no nans)</li>
3618 <li><tt>ueq</tt>: unordered or equal</li>
3619 <li><tt>ugt</tt>: unordered or greater than </li>
3620 <li><tt>uge</tt>: unordered or greater than or equal</li>
3621 <li><tt>ult</tt>: unordered or less than </li>
3622 <li><tt>ule</tt>: unordered or less than or equal</li>
3623 <li><tt>une</tt>: unordered or not equal</li>
3624 <li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3625 <li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3626</ol>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3627<p><i>Ordered</i> means that neither operand is a QNAN while
Reid Spencer93a49852006-12-06 07:08:07 +0000[diff] [blame]3628<i>unordered</i> means that either operand may be a QNAN.</p>
Reid Spencer350f8aa2007-01-04 05:19:58 +0000[diff] [blame]3629<p>The <tt>val1</tt> and <tt>val2</tt> arguments must be
3630<a href="#t_floating">floating point</a> typed. They must have identical
3631types.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3632<h5>Semantics:</h5>
3633<p>The '<tt>fcmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3634the condition code given as <tt>cond</tt>. The comparison performed always
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3635yields a <a href="#t_primitive">i1</a> result, as follows:
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3636<ol>
3637 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3638 <li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3639 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3640 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3641 <tt>var1</tt> is greather than <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3642 <li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3643 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3644 <li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3645 <tt>var1</tt> is less than <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3646 <li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3647 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3648 <li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3649 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3650 <li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
3651 <li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3652 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3653 <li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3654 <tt>var1</tt> is greater than <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3655 <li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3656 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3657 <li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3658 <tt>var1</tt> is less than <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3659 <li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3660 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3661 <li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3662 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]3663 <li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3664 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
3665</ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]3666
3667<h5>Example:</h5>
3668<pre> &lt;result&gt; = fcmp oeq float 4.0, 5.0 <i>; yields: result=false</i>
3669 &lt;result&gt; = icmp one float 4.0, 5.0 <i>; yields: result=true</i>
3670 &lt;result&gt; = icmp olt float 4.0, 5.0 <i>; yields: result=true</i>
3671 &lt;result&gt; = icmp ueq double 1.0, 2.0 <i>; yields: result=false</i>
3672</pre>
3673</div>
3674
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3675<!-- _______________________________________________________________________ -->
3676<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
3677Instruction</a> </div>
3678<div class="doc_text">
3679<h5>Syntax:</h5>
3680<pre> &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...<br></pre>
3681<h5>Overview:</h5>
3682<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
3683the SSA graph representing the function.</p>
3684<h5>Arguments:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3685<p>The type of the incoming values is specified with the first type
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3686field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
3687as arguments, with one pair for each predecessor basic block of the
3688current block. Only values of <a href="#t_firstclass">first class</a>
3689type may be used as the value arguments to the PHI node. Only labels
3690may be used as the label arguments.</p>
3691<p>There must be no non-phi instructions between the start of a basic
3692block and the PHI instructions: i.e. PHI instructions must be first in
3693a basic block.</p>
3694<h5>Semantics:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3695<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
3696specified by the pair corresponding to the predecessor basic block that executed
3697just prior to the current block.</p>
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3698<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3699<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 Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3700</div>
3701
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3702<!-- _______________________________________________________________________ -->
3703<div class="doc_subsubsection">
3704 <a name="i_select">'<tt>select</tt>' Instruction</a>
3705</div>
3706
3707<div class="doc_text">
3708
3709<h5>Syntax:</h5>
3710
3711<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3712 &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 Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3713</pre>
3714
3715<h5>Overview:</h5>
3716
3717<p>
3718The '<tt>select</tt>' instruction is used to choose one value based on a
3719condition, without branching.
3720</p>
3721
3722
3723<h5>Arguments:</h5>
3724
3725<p>
3726The '<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.
3727</p>
3728
3729<h5>Semantics:</h5>
3730
3731<p>
3732If the boolean condition evaluates to true, the instruction returns the first
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]3733value argument; otherwise, it returns the second value argument.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3734</p>
3735
3736<h5>Example:</h5>
3737
3738<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3739 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3740</pre>
3741</div>
3742
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]3743
3744<!-- _______________________________________________________________________ -->
3745<div class="doc_subsubsection">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3746 <a name="i_call">'<tt>call</tt>' Instruction</a>
3747</div>
3748
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3749<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3750
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3751<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3752<pre>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]3753 &lt;result&gt; = [tail] call [<a href="#callingconv">cconv</a>] &lt;ty&gt; [&lt;fnty&gt;*] &lt;fnptrval&gt;(&lt;param list&gt;)
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3754</pre>
3755
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3756<h5>Overview:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3757
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3758<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3759
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3760<h5>Arguments:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3761
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3762<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3763
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3764<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3765 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3766 <p>The optional "tail" marker indicates whether the callee function accesses
3767 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3768 function call is eligible for tail call optimization. Note that calls may
3769 be marked "tail" even if they do not occur before a <a
3770 href="#i_ret"><tt>ret</tt></a> instruction.
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3771 </li>
3772 <li>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]3773 <p>The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3774 convention</a> the call should use. If none is specified, the call defaults
3775 to using C calling conventions.
3776 </li>
3777 <li>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]3778 <p>'<tt>ty</tt>': the type of the call instruction itself which is also
3779 the type of the return value. Functions that return no value are marked
3780 <tt><a href="#t_void">void</a></tt>.</p>
3781 </li>
3782 <li>
3783 <p>'<tt>fnty</tt>': shall be the signature of the pointer to function
3784 value being invoked. The argument types must match the types implied by
3785 this signature. This type can be omitted if the function is not varargs
3786 and if the function type does not return a pointer to a function.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3787 </li>
3788 <li>
3789 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
3790 be invoked. In most cases, this is a direct function invocation, but
3791 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]3792 to function value.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3793 </li>
3794 <li>
3795 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencera7e302a2005-05-01 22:22:57 +0000[diff] [blame]3796 function signature argument types. All arguments must be of
3797 <a href="#t_firstclass">first class</a> type. If the function signature
3798 indicates the function accepts a variable number of arguments, the extra
3799 arguments can be specified.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3800 </li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3801</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3802
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3803<h5>Semantics:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3804
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3805<p>The '<tt>call</tt>' instruction is used to cause control flow to
3806transfer to a specified function, with its incoming arguments bound to
3807the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
3808instruction in the called function, control flow continues with the
3809instruction after the function call, and the return value of the
3810function is bound to the result argument. This is a simpler case of
3811the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3812
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3813<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3814
3815<pre>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]3816 %retval = call i32 @test(i32 %argc)
3817 call i32 (i8 *, ...)* @printf(i8 * %msg, i32 12, i8 42);
3818 %X = tail call i32 @foo()
3819 %Y = tail call <a href="#callingconv">fastcc</a> i32 @foo()
3820 %Z = call void %foo(i8 97 signext)
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]3821</pre>
3822
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3823</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3824
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3825<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3826<div class="doc_subsubsection">
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]3827 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3828</div>
3829
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3830<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3831
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]3832<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3833
3834<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3835 &lt;resultval&gt; = va_arg &lt;va_list*&gt; &lt;arglist&gt;, &lt;argty&gt;
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3836</pre>
3837
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]3838<h5>Overview:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3839
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3840<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3841the "variable argument" area of a function call. It is used to implement the
3842<tt>va_arg</tt> macro in C.</p>
3843
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]3844<h5>Arguments:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3845
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3846<p>This instruction takes a <tt>va_list*</tt> value and the type of
3847the argument. It returns a value of the specified argument type and
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3848increments the <tt>va_list</tt> to point to the next argument. The
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3849actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3850
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]3851<h5>Semantics:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3852
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3853<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
3854type from the specified <tt>va_list</tt> and causes the
3855<tt>va_list</tt> to point to the next argument. For more information,
3856see the variable argument handling <a href="#int_varargs">Intrinsic
3857Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3858
3859<p>It is legal for this instruction to be called in a function which does not
3860take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3861function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3862
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3863<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]3864href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3865argument.</p>
3866
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]3867<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]3868
3869<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
3870
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3871</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3872
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3873<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3874<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
3875<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3876
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3877<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3878
3879<p>LLVM supports the notion of an "intrinsic function". These functions have
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]3880well known names and semantics and are required to follow certain restrictions.
3881Overall, these intrinsics represent an extension mechanism for the LLVM
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3882language that does not require changing all of the transformations in LLVM when
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]3883adding to the language (or the bitcode reader/writer, the parser, etc...).</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3884
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]3885<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3886prefix is reserved in LLVM for intrinsic names; thus, function names may not
3887begin with this prefix. Intrinsic functions must always be external functions:
3888you cannot define the body of intrinsic functions. Intrinsic functions may
3889only be used in call or invoke instructions: it is illegal to take the address
3890of an intrinsic function. Additionally, because intrinsic functions are part
3891of the LLVM language, it is required if any are added that they be documented
3892here.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3893
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]3894<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents
3895a family of functions that perform the same operation but on different data
3896types. Because LLVM can represent over 8 million different integer types,
3897overloading is used commonly to allow an intrinsic function to operate on any
3898integer type. One or more of the argument types or the result type can be
3899overloaded to accept any integer type. Argument types may also be defined as
3900exactly matching a previous argument's type or the result type. This allows an
3901intrinsic function which accepts multiple arguments, but needs all of them to
3902be of the same type, to only be overloaded with respect to a single argument or
3903the result.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3904
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]3905<p>Overloaded intrinsics will have the names of its overloaded argument types
3906encoded into its function name, each preceded by a period. Only those types
3907which are overloaded result in a name suffix. Arguments whose type is matched
3908against another type do not. For example, the <tt>llvm.ctpop</tt> function can
3909take an integer of any width and returns an integer of exactly the same integer
3910width. This leads to a family of functions such as
3911<tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29 %val)</tt>.
3912Only one type, the return type, is overloaded, and only one type suffix is
3913required. Because the argument's type is matched against the return type, it
3914does not require its own name suffix.</p>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]3915
3916<p>To learn how to add an intrinsic function, please see the
3917<a href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3918</p>
3919
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3920</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3921
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3922<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3923<div class="doc_subsection">
3924 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
3925</div>
3926
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3927<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]3928
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3929<p>Variable argument support is defined in LLVM with the <a
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]3930 href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3931intrinsic functions. These functions are related to the similarly
3932named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]3933
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3934<p>All of these functions operate on arguments that use a
3935target-specific value type "<tt>va_list</tt>". The LLVM assembly
3936language reference manual does not define what this type is, so all
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3937transformations should be prepared to handle these functions regardless of
3938the type used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]3939
Chris Lattner374ab302006-05-15 17:26:46 +0000[diff] [blame]3940<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3941instruction and the variable argument handling intrinsic functions are
3942used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]3943
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3944<div class="doc_code">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3945<pre>
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]3946define i32 @test(i32 %X, ...) {
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3947 ; Initialize variable argument processing
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3948 %ap = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]3949 %ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]3950 call void @llvm.va_start(i8* %ap2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3951
3952 ; Read a single integer argument
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3953 %tmp = va_arg i8** %ap, i32
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3954
3955 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3956 %aq = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]3957 %aq2 = bitcast i8** %aq to i8*
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3958 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]3959 call void @llvm.va_end(i8* %aq2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3960
3961 ; Stop processing of arguments.
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]3962 call void @llvm.va_end(i8* %ap2)
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3963 ret i32 %tmp
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3964}
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]3965
3966declare void @llvm.va_start(i8*)
3967declare void @llvm.va_copy(i8*, i8*)
3968declare void @llvm.va_end(i8*)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]3969</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3970</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3971
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3972</div>
3973
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3974<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3975<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]3976 <a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]3977</div>
3978
3979
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3980<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3981<h5>Syntax:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]3982<pre> declare void %llvm.va_start(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3983<h5>Overview:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3984<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
3985<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
3986href="#i_va_arg">va_arg</a></tt>.</p>
3987
3988<h5>Arguments:</h5>
3989
3990<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
3991
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]3992<h5>Semantics:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3993
3994<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
3995macro available in C. In a target-dependent way, it initializes the
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3996<tt>va_list</tt> element to which the argument points, so that the next call to
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]3997<tt>va_arg</tt> will produce the first variable argument passed to the function.
3998Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3999last argument of the function as the compiler can figure that out.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4000
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4001</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4002
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4003<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4004<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4005 <a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4006</div>
4007
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4008<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4009<h5>Syntax:</h5>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4010<pre> declare void @llvm.va_end(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4011<h5>Overview:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4012
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4013<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*&lt;arglist&gt;</tt>,
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4014which has been initialized previously with <tt><a href="#int_va_start">llvm.va_start</a></tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4015or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4016
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4017<h5>Arguments:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4018
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4019<p>The argument is a pointer to a <tt>va_list</tt> to destroy.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4020
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4021<h5>Semantics:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4022
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4023<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4024macro available in C. In a target-dependent way, it destroys the
4025<tt>va_list</tt> element to which the argument points. Calls to <a
4026href="#int_va_start"><tt>llvm.va_start</tt></a> and <a href="#int_va_copy">
4027<tt>llvm.va_copy</tt></a> must be matched exactly with calls to
4028<tt>llvm.va_end</tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4029
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4030</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4031
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4032<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4033<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4034 <a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4035</div>
4036
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4037<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4038
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4039<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4040
4041<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4042 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4043</pre>
4044
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4045<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4046
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4047<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
4048from the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4049
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4050<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4051
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4052<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]4053The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4054
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4055
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4056<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4057
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4058<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
4059macro available in C. In a target-dependent way, it copies the source
4060<tt>va_list</tt> element into the destination <tt>va_list</tt> element. This
4061intrinsic is necessary because the <tt><a href="#int_va_start">
4062llvm.va_start</a></tt> intrinsic may be arbitrarily complex and require, for
4063example, memory allocation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4064
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4065</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4066
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4067<!-- ======================================================================= -->
4068<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4069 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
4070</div>
4071
4072<div class="doc_text">
4073
4074<p>
4075LLVM support for <a href="GarbageCollection.html">Accurate Garbage
4076Collection</a> requires the implementation and generation of these intrinsics.
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4077These intrinsics allow identification of <a href="#int_gcroot">GC roots on the
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4078stack</a>, as well as garbage collector implementations that require <a
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4079href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a> barriers.
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4080Front-ends for type-safe garbage collected languages should generate these
4081intrinsics to make use of the LLVM garbage collectors. For more details, see <a
4082href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
4083</p>
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]4084
4085<p>The garbage collection intrinsics only operate on objects in the generic
4086 address space (address space zero).</p>
4087
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4088</div>
4089
4090<!-- _______________________________________________________________________ -->
4091<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4092 <a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4093</div>
4094
4095<div class="doc_text">
4096
4097<h5>Syntax:</h5>
4098
4099<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4100 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4101</pre>
4102
4103<h5>Overview:</h5>
4104
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]4105<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4106the code generator, and allows some metadata to be associated with it.</p>
4107
4108<h5>Arguments:</h5>
4109
4110<p>The first argument specifies the address of a stack object that contains the
4111root pointer. The second pointer (which must be either a constant or a global
4112value address) contains the meta-data to be associated with the root.</p>
4113
4114<h5>Semantics:</h5>
4115
4116<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
4117location. At compile-time, the code generator generates information to allow
Gordon Henriksene1433f22007-12-25 02:31:26 +0000[diff] [blame]4118the runtime to find the pointer at GC safe points. The '<tt>llvm.gcroot</tt>'
4119intrinsic may only be used in a function which <a href="#gc">specifies a GC
4120algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4121
4122</div>
4123
4124
4125<!-- _______________________________________________________________________ -->
4126<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4127 <a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4128</div>
4129
4130<div class="doc_text">
4131
4132<h5>Syntax:</h5>
4133
4134<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4135 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4136</pre>
4137
4138<h5>Overview:</h5>
4139
4140<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
4141locations, allowing garbage collector implementations that require read
4142barriers.</p>
4143
4144<h5>Arguments:</h5>
4145
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]4146<p>The second argument is the address to read from, which should be an address
4147allocated from the garbage collector. The first object is a pointer to the
4148start of the referenced object, if needed by the language runtime (otherwise
4149null).</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4150
4151<h5>Semantics:</h5>
4152
4153<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
4154instruction, but may be replaced with substantially more complex code by the
Gordon Henriksene1433f22007-12-25 02:31:26 +0000[diff] [blame]4155garbage collector runtime, as needed. The '<tt>llvm.gcread</tt>' intrinsic
4156may only be used in a function which <a href="#gc">specifies a GC
4157algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4158
4159</div>
4160
4161
4162<!-- _______________________________________________________________________ -->
4163<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4164 <a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4165</div>
4166
4167<div class="doc_text">
4168
4169<h5>Syntax:</h5>
4170
4171<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4172 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4173</pre>
4174
4175<h5>Overview:</h5>
4176
4177<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
4178locations, allowing garbage collector implementations that require write
4179barriers (such as generational or reference counting collectors).</p>
4180
4181<h5>Arguments:</h5>
4182
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]4183<p>The first argument is the reference to store, the second is the start of the
4184object to store it to, and the third is the address of the field of Obj to
4185store to. If the runtime does not require a pointer to the object, Obj may be
4186null.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4187
4188<h5>Semantics:</h5>
4189
4190<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
4191instruction, but may be replaced with substantially more complex code by the
Gordon Henriksene1433f22007-12-25 02:31:26 +0000[diff] [blame]4192garbage collector runtime, as needed. The '<tt>llvm.gcwrite</tt>' intrinsic
4193may only be used in a function which <a href="#gc">specifies a GC
4194algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4195
4196</div>
4197
4198
4199
4200<!-- ======================================================================= -->
4201<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4202 <a name="int_codegen">Code Generator Intrinsics</a>
4203</div>
4204
4205<div class="doc_text">
4206<p>
4207These intrinsics are provided by LLVM to expose special features that may only
4208be implemented with code generator support.
4209</p>
4210
4211</div>
4212
4213<!-- _______________________________________________________________________ -->
4214<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4215 <a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4216</div>
4217
4218<div class="doc_text">
4219
4220<h5>Syntax:</h5>
4221<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4222 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4223</pre>
4224
4225<h5>Overview:</h5>
4226
4227<p>
Chris Lattner32b5d712006-10-15 20:05:59 +0000[diff] [blame]4228The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
4229target-specific value indicating the return address of the current function
4230or one of its callers.
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4231</p>
4232
4233<h5>Arguments:</h5>
4234
4235<p>
4236The argument to this intrinsic indicates which function to return the address
4237for. Zero indicates the calling function, one indicates its caller, etc. The
4238argument is <b>required</b> to be a constant integer value.
4239</p>
4240
4241<h5>Semantics:</h5>
4242
4243<p>
4244The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
4245the return address of the specified call frame, or zero if it cannot be
4246identified. The value returned by this intrinsic is likely to be incorrect or 0
4247for arguments other than zero, so it should only be used for debugging purposes.
4248</p>
4249
4250<p>
4251Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]4252aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4253source-language caller.
4254</p>
4255</div>
4256
4257
4258<!-- _______________________________________________________________________ -->
4259<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4260 <a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4261</div>
4262
4263<div class="doc_text">
4264
4265<h5>Syntax:</h5>
4266<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4267 declare i8 *@llvm.frameaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4268</pre>
4269
4270<h5>Overview:</h5>
4271
4272<p>
Chris Lattner32b5d712006-10-15 20:05:59 +0000[diff] [blame]4273The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
4274target-specific frame pointer value for the specified stack frame.
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4275</p>
4276
4277<h5>Arguments:</h5>
4278
4279<p>
4280The argument to this intrinsic indicates which function to return the frame
4281pointer for. Zero indicates the calling function, one indicates its caller,
4282etc. The argument is <b>required</b> to be a constant integer value.
4283</p>
4284
4285<h5>Semantics:</h5>
4286
4287<p>
4288The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
4289the frame address of the specified call frame, or zero if it cannot be
4290identified. The value returned by this intrinsic is likely to be incorrect or 0
4291for arguments other than zero, so it should only be used for debugging purposes.
4292</p>
4293
4294<p>
4295Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]4296aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4297source-language caller.
4298</p>
4299</div>
4300
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]4301<!-- _______________________________________________________________________ -->
4302<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4303 <a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4304</div>
4305
4306<div class="doc_text">
4307
4308<h5>Syntax:</h5>
4309<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4310 declare i8 *@llvm.stacksave()
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4311</pre>
4312
4313<h5>Overview:</h5>
4314
4315<p>
4316The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4317the function stack, for use with <a href="#int_stackrestore">
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4318<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
4319features like scoped automatic variable sized arrays in C99.
4320</p>
4321
4322<h5>Semantics:</h5>
4323
4324<p>
4325This intrinsic returns a opaque pointer value that can be passed to <a
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4326href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4327<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
4328<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
4329state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
4330practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
4331that were allocated after the <tt>llvm.stacksave</tt> was executed.
4332</p>
4333
4334</div>
4335
4336<!-- _______________________________________________________________________ -->
4337<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4338 <a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4339</div>
4340
4341<div class="doc_text">
4342
4343<h5>Syntax:</h5>
4344<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4345 declare void @llvm.stackrestore(i8 * %ptr)
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4346</pre>
4347
4348<h5>Overview:</h5>
4349
4350<p>
4351The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
4352the function stack to the state it was in when the corresponding <a
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4353href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4354useful for implementing language features like scoped automatic variable sized
4355arrays in C99.
4356</p>
4357
4358<h5>Semantics:</h5>
4359
4360<p>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4361See the description for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]4362</p>
4363
4364</div>
4365
4366
4367<!-- _______________________________________________________________________ -->
4368<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4369 <a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]4370</div>
4371
4372<div class="doc_text">
4373
4374<h5>Syntax:</h5>
4375<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4376 declare void @llvm.prefetch(i8* &lt;address&gt;, i32 &lt;rw&gt;, i32 &lt;locality&gt;)
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]4377</pre>
4378
4379<h5>Overview:</h5>
4380
4381
4382<p>
4383The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4384a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
4385no
4386effect on the behavior of the program but can change its performance
Chris Lattner2a615362005-02-28 19:47:14 +0000[diff] [blame]4387characteristics.
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]4388</p>
4389
4390<h5>Arguments:</h5>
4391
4392<p>
4393<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
4394determining if the fetch should be for a read (0) or write (1), and
4395<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattneraeffb4a2005-03-07 20:31:38 +0000[diff] [blame]4396locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]4397<tt>locality</tt> arguments must be constant integers.
4398</p>
4399
4400<h5>Semantics:</h5>
4401
4402<p>
4403This intrinsic does not modify the behavior of the program. In particular,
4404prefetches cannot trap and do not produce a value. On targets that support this
4405intrinsic, the prefetch can provide hints to the processor cache for better
4406performance.
4407</p>
4408
4409</div>
4410
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]4411<!-- _______________________________________________________________________ -->
4412<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4413 <a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]4414</div>
4415
4416<div class="doc_text">
4417
4418<h5>Syntax:</h5>
4419<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4420 declare void @llvm.pcmarker(i32 &lt;id&gt;)
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]4421</pre>
4422
4423<h5>Overview:</h5>
4424
4425
4426<p>
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4427The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
4428(PC) in a region of
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]4429code to simulators and other tools. The method is target specific, but it is
4430expected that the marker will use exported symbols to transmit the PC of the marker.
Jeff Cohen25d4f7e2005-11-11 02:15:27 +0000[diff] [blame]4431The marker makes no guarantees that it will remain with any specific instruction
Chris Lattnerd07c3f42005-11-15 06:07:55 +0000[diff] [blame]4432after optimizations. It is possible that the presence of a marker will inhibit
Chris Lattnerb3e7afd2006-03-24 07:16:10 +0000[diff] [blame]4433optimizations. The intended use is to be inserted after optimizations to allow
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4434correlations of simulation runs.
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]4435</p>
4436
4437<h5>Arguments:</h5>
4438
4439<p>
4440<tt>id</tt> is a numerical id identifying the marker.
4441</p>
4442
4443<h5>Semantics:</h5>
4444
4445<p>
4446This intrinsic does not modify the behavior of the program. Backends that do not
4447support this intrinisic may ignore it.
4448</p>
4449
4450</div>
4451
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]4452<!-- _______________________________________________________________________ -->
4453<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4454 <a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]4455</div>
4456
4457<div class="doc_text">
4458
4459<h5>Syntax:</h5>
4460<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4461 declare i64 @llvm.readcyclecounter( )
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]4462</pre>
4463
4464<h5>Overview:</h5>
4465
4466
4467<p>
4468The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
4469counter register (or similar low latency, high accuracy clocks) on those targets
4470that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
4471As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
4472should only be used for small timings.
4473</p>
4474
4475<h5>Semantics:</h5>
4476
4477<p>
4478When directly supported, reading the cycle counter should not modify any memory.
4479Implementations are allowed to either return a application specific value or a
4480system wide value. On backends without support, this is lowered to a constant 0.
4481</p>
4482
4483</div>
4484
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4485<!-- ======================================================================= -->
4486<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4487 <a name="int_libc">Standard C Library Intrinsics</a>
4488</div>
4489
4490<div class="doc_text">
4491<p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4492LLVM provides intrinsics for a few important standard C library functions.
4493These intrinsics allow source-language front-ends to pass information about the
4494alignment of the pointer arguments to the code generator, providing opportunity
4495for more efficient code generation.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4496</p>
4497
4498</div>
4499
4500<!-- _______________________________________________________________________ -->
4501<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4502 <a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4503</div>
4504
4505<div class="doc_text">
4506
4507<h5>Syntax:</h5>
4508<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4509 declare void @llvm.memcpy.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4510 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4511 declare void @llvm.memcpy.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4512 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4513</pre>
4514
4515<h5>Overview:</h5>
4516
4517<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4518The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4519location to the destination location.
4520</p>
4521
4522<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4523Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
4524intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4525</p>
4526
4527<h5>Arguments:</h5>
4528
4529<p>
4530The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4531the source. The third argument is an integer argument
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4532specifying the number of bytes to copy, and the fourth argument is the alignment
4533of the source and destination locations.
4534</p>
4535
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]4536<p>
4537If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattnerf0afc2c2006-03-04 00:02:10 +0000[diff] [blame]4538the caller guarantees that both the source and destination pointers are aligned
4539to that boundary.
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]4540</p>
4541
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4542<h5>Semantics:</h5>
4543
4544<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4545The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4546location to the destination location, which are not allowed to overlap. It
4547copies "len" bytes of memory over. If the argument is known to be aligned to
4548some boundary, this can be specified as the fourth argument, otherwise it should
4549be set to 0 or 1.
4550</p>
4551</div>
4552
4553
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4554<!-- _______________________________________________________________________ -->
4555<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4556 <a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4557</div>
4558
4559<div class="doc_text">
4560
4561<h5>Syntax:</h5>
4562<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4563 declare void @llvm.memmove.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4564 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4565 declare void @llvm.memmove.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4566 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4567</pre>
4568
4569<h5>Overview:</h5>
4570
4571<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4572The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the source
4573location to the destination location. It is similar to the
Chris Lattner4b2cbcf2008-01-06 19:51:52 +0000[diff] [blame]4574'<tt>llvm.memcpy</tt>' intrinsic but allows the two memory locations to overlap.
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4575</p>
4576
4577<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4578Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
4579intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4580</p>
4581
4582<h5>Arguments:</h5>
4583
4584<p>
4585The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4586the source. The third argument is an integer argument
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4587specifying the number of bytes to copy, and the fourth argument is the alignment
4588of the source and destination locations.
4589</p>
4590
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]4591<p>
4592If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattnerf0afc2c2006-03-04 00:02:10 +0000[diff] [blame]4593the caller guarantees that the source and destination pointers are aligned to
4594that boundary.
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]4595</p>
4596
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4597<h5>Semantics:</h5>
4598
4599<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4600The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]4601location to the destination location, which may overlap. It
4602copies "len" bytes of memory over. If the argument is known to be aligned to
4603some boundary, this can be specified as the fourth argument, otherwise it should
4604be set to 0 or 1.
4605</p>
4606</div>
4607
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4608
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4609<!-- _______________________________________________________________________ -->
4610<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4611 <a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4612</div>
4613
4614<div class="doc_text">
4615
4616<h5>Syntax:</h5>
4617<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4618 declare void @llvm.memset.i32(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4619 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4620 declare void @llvm.memset.i64(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4621 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4622</pre>
4623
4624<h5>Overview:</h5>
4625
4626<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4627The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a particular
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4628byte value.
4629</p>
4630
4631<p>
4632Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
4633does not return a value, and takes an extra alignment argument.
4634</p>
4635
4636<h5>Arguments:</h5>
4637
4638<p>
4639The first argument is a pointer to the destination to fill, the second is the
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4640byte value to fill it with, the third argument is an integer
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4641argument specifying the number of bytes to fill, and the fourth argument is the
4642known alignment of destination location.
4643</p>
4644
4645<p>
4646If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattnerf0afc2c2006-03-04 00:02:10 +0000[diff] [blame]4647the caller guarantees that the destination pointer is aligned to that boundary.
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4648</p>
4649
4650<h5>Semantics:</h5>
4651
4652<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]4653The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting at
4654the
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4655destination location. If the argument is known to be aligned to some boundary,
4656this can be specified as the fourth argument, otherwise it should be set to 0 or
46571.
4658</p>
4659</div>
4660
4661
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]4662<!-- _______________________________________________________________________ -->
4663<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4664 <a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]4665</div>
4666
4667<div class="doc_text">
4668
4669<h5>Syntax:</h5>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]4670<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4671floating point or vector of floating point type. Not all targets support all
4672types however.
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]4673<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]4674 declare float @llvm.sqrt.f32(float %Val)
4675 declare double @llvm.sqrt.f64(double %Val)
4676 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
4677 declare fp128 @llvm.sqrt.f128(fp128 %Val)
4678 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]4679</pre>
4680
4681<h5>Overview:</h5>
4682
4683<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]4684The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4685returning the same value as the libm '<tt>sqrt</tt>' functions would. Unlike
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]4686<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
Chris Lattner103e2d72008-01-29 07:00:44 +0000[diff] [blame]4687negative numbers other than -0.0 (which allows for better optimization, because
4688there is no need to worry about errno being set). <tt>llvm.sqrt(-0.0)</tt> is
4689defined to return -0.0 like IEEE sqrt.
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]4690</p>
4691
4692<h5>Arguments:</h5>
4693
4694<p>
4695The argument and return value are floating point numbers of the same type.
4696</p>
4697
4698<h5>Semantics:</h5>
4699
4700<p>
Dan Gohmand6257fe2007-07-16 14:37:41 +0000[diff] [blame]4701This function returns the sqrt of the specified operand if it is a nonnegative
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]4702floating point number.
4703</p>
4704</div>
4705
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]4706<!-- _______________________________________________________________________ -->
4707<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4708 <a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]4709</div>
4710
4711<div class="doc_text">
4712
4713<h5>Syntax:</h5>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]4714<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4715floating point or vector of floating point type. Not all targets support all
4716types however.
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]4717<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]4718 declare float @llvm.powi.f32(float %Val, i32 %power)
4719 declare double @llvm.powi.f64(double %Val, i32 %power)
4720 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
4721 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
4722 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]4723</pre>
4724
4725<h5>Overview:</h5>
4726
4727<p>
4728The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
4729specified (positive or negative) power. The order of evaluation of
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4730multiplications is not defined. When a vector of floating point type is
4731used, the second argument remains a scalar integer value.
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]4732</p>
4733
4734<h5>Arguments:</h5>
4735
4736<p>
4737The second argument is an integer power, and the first is a value to raise to
4738that power.
4739</p>
4740
4741<h5>Semantics:</h5>
4742
4743<p>
4744This function returns the first value raised to the second power with an
4745unspecified sequence of rounding operations.</p>
4746</div>
4747
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4748<!-- _______________________________________________________________________ -->
4749<div class="doc_subsubsection">
4750 <a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
4751</div>
4752
4753<div class="doc_text">
4754
4755<h5>Syntax:</h5>
4756<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
4757floating point or vector of floating point type. Not all targets support all
4758types however.
4759<pre>
4760 declare float @llvm.sin.f32(float %Val)
4761 declare double @llvm.sin.f64(double %Val)
4762 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
4763 declare fp128 @llvm.sin.f128(fp128 %Val)
4764 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
4765</pre>
4766
4767<h5>Overview:</h5>
4768
4769<p>
4770The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.
4771</p>
4772
4773<h5>Arguments:</h5>
4774
4775<p>
4776The argument and return value are floating point numbers of the same type.
4777</p>
4778
4779<h5>Semantics:</h5>
4780
4781<p>
4782This function returns the sine of the specified operand, returning the
4783same values as the libm <tt>sin</tt> functions would, and handles error
Dan Gohmanba83b7e2007-10-17 18:05:13 +0000[diff] [blame]4784conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4785</div>
4786
4787<!-- _______________________________________________________________________ -->
4788<div class="doc_subsubsection">
4789 <a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
4790</div>
4791
4792<div class="doc_text">
4793
4794<h5>Syntax:</h5>
4795<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
4796floating point or vector of floating point type. Not all targets support all
4797types however.
4798<pre>
4799 declare float @llvm.cos.f32(float %Val)
4800 declare double @llvm.cos.f64(double %Val)
4801 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
4802 declare fp128 @llvm.cos.f128(fp128 %Val)
4803 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
4804</pre>
4805
4806<h5>Overview:</h5>
4807
4808<p>
4809The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.
4810</p>
4811
4812<h5>Arguments:</h5>
4813
4814<p>
4815The argument and return value are floating point numbers of the same type.
4816</p>
4817
4818<h5>Semantics:</h5>
4819
4820<p>
4821This function returns the cosine of the specified operand, returning the
4822same values as the libm <tt>cos</tt> functions would, and handles error
Dan Gohmanba83b7e2007-10-17 18:05:13 +0000[diff] [blame]4823conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4824</div>
4825
4826<!-- _______________________________________________________________________ -->
4827<div class="doc_subsubsection">
4828 <a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
4829</div>
4830
4831<div class="doc_text">
4832
4833<h5>Syntax:</h5>
4834<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
4835floating point or vector of floating point type. Not all targets support all
4836types however.
4837<pre>
4838 declare float @llvm.pow.f32(float %Val, float %Power)
4839 declare double @llvm.pow.f64(double %Val, double %Power)
4840 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
4841 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
4842 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
4843</pre>
4844
4845<h5>Overview:</h5>
4846
4847<p>
4848The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
4849specified (positive or negative) power.
4850</p>
4851
4852<h5>Arguments:</h5>
4853
4854<p>
4855The second argument is a floating point power, and the first is a value to
4856raise to that power.
4857</p>
4858
4859<h5>Semantics:</h5>
4860
4861<p>
4862This function returns the first value raised to the second power,
4863returning the
4864same values as the libm <tt>pow</tt> functions would, and handles error
Dan Gohmanba83b7e2007-10-17 18:05:13 +0000[diff] [blame]4865conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]4866</div>
4867
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]4868
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4869<!-- ======================================================================= -->
4870<div class="doc_subsection">
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4871 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4872</div>
4873
4874<div class="doc_text">
4875<p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4876LLVM provides intrinsics for a few important bit manipulation operations.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4877These allow efficient code generation for some algorithms.
4878</p>
4879
4880</div>
4881
4882<!-- _______________________________________________________________________ -->
4883<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4884 <a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4885</div>
4886
4887<div class="doc_text">
4888
4889<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4890<p>This is an overloaded intrinsic function. You can use bswap on any integer
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4891type that is an even number of bytes (i.e. BitWidth % 16 == 0).
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4892<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4893 declare i16 @llvm.bswap.i16(i16 &lt;id&gt;)
4894 declare i32 @llvm.bswap.i32(i32 &lt;id&gt;)
4895 declare i64 @llvm.bswap.i64(i64 &lt;id&gt;)
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4896</pre>
4897
4898<h5>Overview:</h5>
4899
4900<p>
Reid Spencer338ea092007-04-02 02:25:19 +0000[diff] [blame]4901The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4902values with an even number of bytes (positive multiple of 16 bits). These are
4903useful for performing operations on data that is not in the target's native
4904byte order.
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4905</p>
4906
4907<h5>Semantics:</h5>
4908
4909<p>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4910The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4911and low byte of the input i16 swapped. Similarly, the <tt>llvm.bswap.i32</tt>
4912intrinsic returns an i32 value that has the four bytes of the input i32
4913swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the returned
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4914i32 will have its bytes in 3, 2, 1, 0 order. The <tt>llvm.bswap.i48</tt>,
4915<tt>llvm.bswap.i64</tt> and other intrinsics extend this concept to
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4916additional even-byte lengths (6 bytes, 8 bytes and more, respectively).
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]4917</p>
4918
4919</div>
4920
4921<!-- _______________________________________________________________________ -->
4922<div class="doc_subsubsection">
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]4923 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4924</div>
4925
4926<div class="doc_text">
4927
4928<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4929<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
4930width. Not all targets support all bit widths however.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4931<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4932 declare i8 @llvm.ctpop.i8 (i8 &lt;src&gt;)
4933 declare i16 @llvm.ctpop.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4934 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4935 declare i64 @llvm.ctpop.i64(i64 &lt;src&gt;)
4936 declare i256 @llvm.ctpop.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4937</pre>
4938
4939<h5>Overview:</h5>
4940
4941<p>
Chris Lattnerec6cb612006-01-16 22:38:59 +0000[diff] [blame]4942The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set in a
4943value.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4944</p>
4945
4946<h5>Arguments:</h5>
4947
4948<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]4949The only argument is the value to be counted. The argument may be of any
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]4950integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4951</p>
4952
4953<h5>Semantics:</h5>
4954
4955<p>
4956The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
4957</p>
4958</div>
4959
4960<!-- _______________________________________________________________________ -->
4961<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]4962 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4963</div>
4964
4965<div class="doc_text">
4966
4967<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4968<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
4969integer bit width. Not all targets support all bit widths however.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4970<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4971 declare i8 @llvm.ctlz.i8 (i8 &lt;src&gt;)
4972 declare i16 @llvm.ctlz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4973 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4974 declare i64 @llvm.ctlz.i64(i64 &lt;src&gt;)
4975 declare i256 @llvm.ctlz.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4976</pre>
4977
4978<h5>Overview:</h5>
4979
4980<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]4981The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
4982leading zeros in a variable.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4983</p>
4984
4985<h5>Arguments:</h5>
4986
4987<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]4988The only argument is the value to be counted. The argument may be of any
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]4989integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4990</p>
4991
4992<h5>Semantics:</h5>
4993
4994<p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]4995The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
4996in a variable. If the src == 0 then the result is the size in bits of the type
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4997of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]4998</p>
4999</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]5000
5001
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5002
5003<!-- _______________________________________________________________________ -->
5004<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]5005 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5006</div>
5007
5008<div class="doc_text">
5009
5010<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5011<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
5012integer bit width. Not all targets support all bit widths however.
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5013<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5014 declare i8 @llvm.cttz.i8 (i8 &lt;src&gt;)
5015 declare i16 @llvm.cttz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5016 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5017 declare i64 @llvm.cttz.i64(i64 &lt;src&gt;)
5018 declare i256 @llvm.cttz.i256(i256 &lt;src&gt;)
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5019</pre>
5020
5021<h5>Overview:</h5>
5022
5023<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]5024The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
5025trailing zeros.
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5026</p>
5027
5028<h5>Arguments:</h5>
5029
5030<p>
5031The only argument is the value to be counted. The argument may be of any
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]5032integer type. The return type must match the argument type.
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5033</p>
5034
5035<h5>Semantics:</h5>
5036
5037<p>
5038The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
5039in a variable. If the src == 0 then the result is the size in bits of the type
5040of src. For example, <tt>llvm.cttz(2) = 1</tt>.
5041</p>
5042</div>
5043
Reid Spencer497d93e2007-04-01 08:27:01 +0000[diff] [blame]5044<!-- _______________________________________________________________________ -->
5045<div class="doc_subsubsection">
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5046 <a name="int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic</a>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5047</div>
5048
5049<div class="doc_text">
5050
5051<h5>Syntax:</h5>
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5052<p>This is an overloaded intrinsic. You can use <tt>llvm.part.select</tt>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5053on any integer bit width.
5054<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5055 declare i17 @llvm.part.select.i17 (i17 %val, i32 %loBit, i32 %hiBit)
5056 declare i29 @llvm.part.select.i29 (i29 %val, i32 %loBit, i32 %hiBit)
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5057</pre>
5058
5059<h5>Overview:</h5>
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5060<p>The '<tt>llvm.part.select</tt>' family of intrinsic functions selects a
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5061range of bits from an integer value and returns them in the same bit width as
5062the original value.</p>
5063
5064<h5>Arguments:</h5>
5065<p>The first argument, <tt>%val</tt> and the result may be integer types of
5066any bit width but they must have the same bit width. The second and third
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5067arguments must be <tt>i32</tt> type since they specify only a bit index.</p>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5068
5069<h5>Semantics:</h5>
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5070<p>The operation of the '<tt>llvm.part.select</tt>' intrinsic has two modes
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5071of operation: forwards and reverse. If <tt>%loBit</tt> is greater than
5072<tt>%hiBits</tt> then the intrinsic operates in reverse mode. Otherwise it
5073operates in forward mode.</p>
5074<p>In forward mode, this intrinsic is the equivalent of shifting <tt>%val</tt>
5075right by <tt>%loBit</tt> bits and then ANDing it with a mask with
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5076only the <tt>%hiBit - %loBit</tt> bits set, as follows:</p>
5077<ol>
5078 <li>The <tt>%val</tt> is shifted right (LSHR) by the number of bits specified
5079 by <tt>%loBits</tt>. This normalizes the value to the low order bits.</li>
5080 <li>The <tt>%loBits</tt> value is subtracted from the <tt>%hiBits</tt> value
5081 to determine the number of bits to retain.</li>
5082 <li>A mask of the retained bits is created by shifting a -1 value.</li>
5083 <li>The mask is ANDed with <tt>%val</tt> to produce the result.
5084</ol>
Reid Spencerd6a85b52007-05-14 16:14:57 +0000[diff] [blame]5085<p>In reverse mode, a similar computation is made except that the bits are
5086returned in the reverse order. So, for example, if <tt>X</tt> has the value
5087<tt>i16 0x0ACF (101011001111)</tt> and we apply
5088<tt>part.select(i16 X, 8, 3)</tt> to it, we get back the value
5089<tt>i16 0x0026 (000000100110)</tt>.</p>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5090</div>
5091
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5092<div class="doc_subsubsection">
5093 <a name="int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic</a>
5094</div>
5095
5096<div class="doc_text">
5097
5098<h5>Syntax:</h5>
5099<p>This is an overloaded intrinsic. You can use <tt>llvm.part.set</tt>
5100on any integer bit width.
5101<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5102 declare i17 @llvm.part.set.i17.i9 (i17 %val, i9 %repl, i32 %lo, i32 %hi)
5103 declare i29 @llvm.part.set.i29.i9 (i29 %val, i9 %repl, i32 %lo, i32 %hi)
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5104</pre>
5105
5106<h5>Overview:</h5>
5107<p>The '<tt>llvm.part.set</tt>' family of intrinsic functions replaces a range
5108of bits in an integer value with another integer value. It returns the integer
5109with the replaced bits.</p>
5110
5111<h5>Arguments:</h5>
5112<p>The first argument, <tt>%val</tt> and the result may be integer types of
5113any bit width but they must have the same bit width. <tt>%val</tt> is the value
5114whose bits will be replaced. The second argument, <tt>%repl</tt> may be an
5115integer of any bit width. The third and fourth arguments must be <tt>i32</tt>
5116type since they specify only a bit index.</p>
5117
5118<h5>Semantics:</h5>
5119<p>The operation of the '<tt>llvm.part.set</tt>' intrinsic has two modes
5120of operation: forwards and reverse. If <tt>%lo</tt> is greater than
5121<tt>%hi</tt> then the intrinsic operates in reverse mode. Otherwise it
5122operates in forward mode.</p>
5123<p>For both modes, the <tt>%repl</tt> value is prepared for use by either
5124truncating it down to the size of the replacement area or zero extending it
5125up to that size.</p>
5126<p>In forward mode, the bits between <tt>%lo</tt> and <tt>%hi</tt> (inclusive)
5127are replaced with corresponding bits from <tt>%repl</tt>. That is the 0th bit
5128in <tt>%repl</tt> replaces the <tt>%lo</tt>th bit in <tt>%val</tt> and etc. up
5129to the <tt>%hi</tt>th bit.
Reid Spencerc6749c42007-05-14 16:50:20 +0000[diff] [blame]5130<p>In reverse mode, a similar computation is made except that the bits are
5131reversed. That is, the <tt>0</tt>th bit in <tt>%repl</tt> replaces the
5132<tt>%hi</tt> bit in <tt>%val</tt> and etc. down to the <tt>%lo</tt>th bit.
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5133<h5>Examples:</h5>
5134<pre>
Reid Spencerf0dbf642007-04-12 01:03:03 +0000[diff] [blame]5135 llvm.part.set(0xFFFF, 0, 4, 7) -&gt; 0xFF0F
Reid Spencerc6749c42007-05-14 16:50:20 +0000[diff] [blame]5136 llvm.part.set(0xFFFF, 0, 7, 4) -&gt; 0xFF0F
5137 llvm.part.set(0xFFFF, 1, 7, 4) -&gt; 0xFF8F
5138 llvm.part.set(0xFFFF, F, 8, 3) -&gt; 0xFFE7
Reid Spencerf0dbf642007-04-12 01:03:03 +0000[diff] [blame]5139 llvm.part.set(0xFFFF, 0, 3, 8) -&gt; 0xFE07
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]5140</pre>
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5141</div>
5142
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5143<!-- ======================================================================= -->
5144<div class="doc_subsection">
5145 <a name="int_debugger">Debugger Intrinsics</a>
5146</div>
5147
5148<div class="doc_text">
5149<p>
5150The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
5151are described in the <a
5152href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
5153Debugging</a> document.
5154</p>
5155</div>
5156
5157
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]5158<!-- ======================================================================= -->
5159<div class="doc_subsection">
5160 <a name="int_eh">Exception Handling Intrinsics</a>
5161</div>
5162
5163<div class="doc_text">
5164<p> The LLVM exception handling intrinsics (which all start with
5165<tt>llvm.eh.</tt> prefix), are described in the <a
5166href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
5167Handling</a> document. </p>
5168</div>
5169
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5170<!-- ======================================================================= -->
5171<div class="doc_subsection">
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5172 <a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5173</div>
5174
5175<div class="doc_text">
5176<p>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5177 This intrinsic makes it possible to excise one parameter, marked with
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5178 the <tt>nest</tt> attribute, from a function. The result is a callable
5179 function pointer lacking the nest parameter - the caller does not need
5180 to provide a value for it. Instead, the value to use is stored in
5181 advance in a "trampoline", a block of memory usually allocated
5182 on the stack, which also contains code to splice the nest value into the
5183 argument list. This is used to implement the GCC nested function address
5184 extension.
5185</p>
5186<p>
5187 For example, if the function is
5188 <tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
Bill Wendling03295ca2007-09-22 09:23:55 +0000[diff] [blame]5189 pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as follows:</p>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5190<pre>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5191 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
5192 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
5193 %p = call i8* @llvm.init.trampoline( i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8*), i8* %nval )
5194 %fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5195</pre>
Bill Wendling03295ca2007-09-22 09:23:55 +0000[diff] [blame]5196 <p>The call <tt>%val = call i32 %fp( i32 %x, i32 %y )</tt> is then equivalent
5197 to <tt>%val = call i32 %f( i8* %nval, i32 %x, i32 %y )</tt>.</p>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5198</div>
5199
5200<!-- _______________________________________________________________________ -->
5201<div class="doc_subsubsection">
5202 <a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
5203</div>
5204<div class="doc_text">
5205<h5>Syntax:</h5>
5206<pre>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5207declare i8* @llvm.init.trampoline(i8* &lt;tramp&gt;, i8* &lt;func&gt;, i8* &lt;nval&gt;)
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5208</pre>
5209<h5>Overview:</h5>
5210<p>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5211 This fills the memory pointed to by <tt>tramp</tt> with code
5212 and returns a function pointer suitable for executing it.
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5213</p>
5214<h5>Arguments:</h5>
5215<p>
5216 The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
5217 pointers. The <tt>tramp</tt> argument must point to a sufficiently large
5218 and sufficiently aligned block of memory; this memory is written to by the
Duncan Sandsc00c2ba2007-08-22 23:39:54 +0000[diff] [blame]5219 intrinsic. Note that the size and the alignment are target-specific - LLVM
5220 currently provides no portable way of determining them, so a front-end that
5221 generates this intrinsic needs to have some target-specific knowledge.
5222 The <tt>func</tt> argument must hold a function bitcast to an <tt>i8*</tt>.
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5223</p>
5224<h5>Semantics:</h5>
5225<p>
5226 The block of memory pointed to by <tt>tramp</tt> is filled with target
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5227 dependent code, turning it into a function. A pointer to this function is
5228 returned, but needs to be bitcast to an
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5229 <a href="#int_trampoline">appropriate function pointer type</a>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5230 before being called. The new function's signature is the same as that of
5231 <tt>func</tt> with any arguments marked with the <tt>nest</tt> attribute
5232 removed. At most one such <tt>nest</tt> argument is allowed, and it must be
5233 of pointer type. Calling the new function is equivalent to calling
5234 <tt>func</tt> with the same argument list, but with <tt>nval</tt> used for the
5235 missing <tt>nest</tt> argument. If, after calling
5236 <tt>llvm.init.trampoline</tt>, the memory pointed to by <tt>tramp</tt> is
5237 modified, then the effect of any later call to the returned function pointer is
5238 undefined.
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5239</p>
5240</div>
5241
5242<!-- ======================================================================= -->
5243<div class="doc_subsection">
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]5244 <a name="int_atomics">Atomic Operations and Synchronization Intrinsics</a>
5245</div>
5246
5247<div class="doc_text">
5248<p>
5249 These intrinsic functions expand the "universal IR" of LLVM to represent
5250 hardware constructs for atomic operations and memory synchronization. This
5251 provides an interface to the hardware, not an interface to the programmer. It
5252 is aimed at a low enough level to allow any programming models or APIs which
5253 need atomic behaviors to map cleanly onto it. It is also modeled primarily on
5254 hardware behavior. Just as hardware provides a "universal IR" for source
5255 languages, it also provides a starting point for developing a "universal"
5256 atomic operation and synchronization IR.
5257</p>
5258<p>
5259 These do <em>not</em> form an API such as high-level threading libraries,
5260 software transaction memory systems, atomic primitives, and intrinsic
5261 functions as found in BSD, GNU libc, atomic_ops, APR, and other system and
5262 application libraries. The hardware interface provided by LLVM should allow
5263 a clean implementation of all of these APIs and parallel programming models.
5264 No one model or paradigm should be selected above others unless the hardware
5265 itself ubiquitously does so.
5266
5267</p>
5268</div>
5269
5270<!-- _______________________________________________________________________ -->
5271<div class="doc_subsubsection">
5272 <a name="int_memory_barrier">'<tt>llvm.memory.barrier</tt>' Intrinsic</a>
5273</div>
5274<div class="doc_text">
5275<h5>Syntax:</h5>
5276<pre>
5277declare void @llvm.memory.barrier( i1 &lt;ll&gt;, i1 &lt;ls&gt;, i1 &lt;sl&gt;, i1 &lt;ss&gt;,
5278i1 &lt;device&gt; )
5279
5280</pre>
5281<h5>Overview:</h5>
5282<p>
5283 The <tt>llvm.memory.barrier</tt> intrinsic guarantees ordering between
5284 specific pairs of memory access types.
5285</p>
5286<h5>Arguments:</h5>
5287<p>
5288 The <tt>llvm.memory.barrier</tt> intrinsic requires five boolean arguments.
5289 The first four arguments enables a specific barrier as listed below. The fith
5290 argument specifies that the barrier applies to io or device or uncached memory.
5291
5292</p>
5293 <ul>
5294 <li><tt>ll</tt>: load-load barrier</li>
5295 <li><tt>ls</tt>: load-store barrier</li>
5296 <li><tt>sl</tt>: store-load barrier</li>
5297 <li><tt>ss</tt>: store-store barrier</li>
5298 <li><tt>device</tt>: barrier applies to device and uncached memory also.
5299 </ul>
5300<h5>Semantics:</h5>
5301<p>
5302 This intrinsic causes the system to enforce some ordering constraints upon
5303 the loads and stores of the program. This barrier does not indicate
5304 <em>when</em> any events will occur, it only enforces an <em>order</em> in
5305 which they occur. For any of the specified pairs of load and store operations
5306 (f.ex. load-load, or store-load), all of the first operations preceding the
5307 barrier will complete before any of the second operations succeeding the
5308 barrier begin. Specifically the semantics for each pairing is as follows:
5309</p>
5310 <ul>
5311 <li><tt>ll</tt>: All loads before the barrier must complete before any load
5312 after the barrier begins.</li>
5313
5314 <li><tt>ls</tt>: All loads before the barrier must complete before any
5315 store after the barrier begins.</li>
5316 <li><tt>ss</tt>: All stores before the barrier must complete before any
5317 store after the barrier begins.</li>
5318 <li><tt>sl</tt>: All stores before the barrier must complete before any
5319 load after the barrier begins.</li>
5320 </ul>
5321<p>
5322 These semantics are applied with a logical "and" behavior when more than one
5323 is enabled in a single memory barrier intrinsic.
5324</p>
5325<p>
5326 Backends may implement stronger barriers than those requested when they do not
5327 support as fine grained a barrier as requested. Some architectures do not
5328 need all types of barriers and on such architectures, these become noops.
5329</p>
5330<h5>Example:</h5>
5331<pre>
5332%ptr = malloc i32
5333 store i32 4, %ptr
5334
5335%result1 = load i32* %ptr <i>; yields {i32}:result1 = 4</i>
5336 call void @llvm.memory.barrier( i1 false, i1 true, i1 false, i1 false )
5337 <i>; guarantee the above finishes</i>
5338 store i32 8, %ptr <i>; before this begins</i>
5339</pre>
5340</div>
5341
5342
5343<!-- ======================================================================= -->
5344<div class="doc_subsection">
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5345 <a name="int_general">General Intrinsics</a>
5346</div>
5347
5348<div class="doc_text">
5349<p> This class of intrinsics is designed to be generic and has
5350no specific purpose. </p>
5351</div>
5352
5353<!-- _______________________________________________________________________ -->
5354<div class="doc_subsubsection">
5355 <a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
5356</div>
5357
5358<div class="doc_text">
5359
5360<h5>Syntax:</h5>
5361<pre>
Tanya Lattnerd2e84422007-06-18 23:42:37 +0000[diff] [blame]5362 declare void @llvm.var.annotation(i8* &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5363</pre>
5364
5365<h5>Overview:</h5>
5366
5367<p>
5368The '<tt>llvm.var.annotation</tt>' intrinsic
5369</p>
5370
5371<h5>Arguments:</h5>
5372
5373<p>
Tanya Lattnerd2e84422007-06-18 23:42:37 +0000[diff] [blame]5374The first argument is a pointer to a value, the second is a pointer to a
5375global string, the third is a pointer to a global string which is the source
5376file name, and the last argument is the line number.
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5377</p>
5378
5379<h5>Semantics:</h5>
5380
5381<p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]5382This intrinsic allows annotation of local variables with arbitrary strings.
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5383This can be useful for special purpose optimizations that want to look for these
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]5384annotations. These have no other defined use, they are ignored by code
5385generation and optimization.
5386</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5387</div>
5388
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]5389<!-- _______________________________________________________________________ -->
5390<div class="doc_subsubsection">
Tanya Lattnere1a8da02007-09-21 23:57:59 +0000[diff] [blame]5391 <a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]5392</div>
5393
5394<div class="doc_text">
5395
5396<h5>Syntax:</h5>
Tanya Lattner39cfba62007-09-21 23:56:27 +0000[diff] [blame]5397<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
5398any integer bit width.
5399</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]5400<pre>
Tanya Lattnerd3989a82007-09-22 00:03:01 +0000[diff] [blame]5401 declare i8 @llvm.annotation.i8(i8 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5402 declare i16 @llvm.annotation.i16(i16 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5403 declare i32 @llvm.annotation.i32(i32 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5404 declare i64 @llvm.annotation.i64(i64 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5405 declare i256 @llvm.annotation.i256(i256 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]5406</pre>
5407
5408<h5>Overview:</h5>
Tanya Lattner39cfba62007-09-21 23:56:27 +0000[diff] [blame]5409
5410<p>
5411The '<tt>llvm.annotation</tt>' intrinsic.
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]5412</p>
5413
5414<h5>Arguments:</h5>
5415
5416<p>
5417The first argument is an integer value (result of some expression),
5418the second is a pointer to a global string, the third is a pointer to a global
5419string which is the source file name, and the last argument is the line number.
Tanya Lattner39cfba62007-09-21 23:56:27 +0000[diff] [blame]5420It returns the value of the first argument.
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]5421</p>
5422
5423<h5>Semantics:</h5>
5424
5425<p>
5426This intrinsic allows annotations to be put on arbitrary expressions
5427with arbitrary strings. This can be useful for special purpose optimizations
5428that want to look for these annotations. These have no other defined use, they
5429are ignored by code generation and optimization.
5430</div>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]5431
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]5432<!-- _______________________________________________________________________ -->
5433<div class="doc_subsubsection">
5434 <a name="int_trap">'<tt>llvm.trap</tt>' Intrinsic</a>
5435</div>
5436
5437<div class="doc_text">
5438
5439<h5>Syntax:</h5>
5440<pre>
5441 declare void @llvm.trap()
5442</pre>
5443
5444<h5>Overview:</h5>
5445
5446<p>
5447The '<tt>llvm.trap</tt>' intrinsic
5448</p>
5449
5450<h5>Arguments:</h5>
5451
5452<p>
5453None
5454</p>
5455
5456<h5>Semantics:</h5>
5457
5458<p>
5459This intrinsics is lowered to the target dependent trap instruction. If the
5460target does not have a trap instruction, this intrinsic will be lowered to the
5461call of the abort() function.
5462</p>
5463</div>
5464
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5465<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5466<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]5467<address>
5468 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
5469 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
5470 <a href="http://validator.w3.org/check/referer"><img
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]5471 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!"></a>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]5472
5473 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencer05fe4b02006-03-14 05:39:39 +0000[diff] [blame]5474 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]5475 Last modified: $Date$
5476</address>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]5477
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5478</body>
5479</html>