Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 0538729a3a688723908cf39b381a67003cf00ccc / . / docs / LangRef.html
blob: 21144125b043478dfa5cf72ba5cdd134d67c49a2 [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">
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]8 <meta name="description"
Reid Spencer3921c742004-08-26 20:44:00 +0000[diff] [blame]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>
Bill Wendling3d10a5a2009-07-20 01:03:30 +0000[diff] [blame]23 <li><a href="#linkage">Linkage Types</a>
24 <ol>
Bill Wendling987e7eb2009-07-20 02:41:50 +0000[diff] [blame]25 <li><a href="#linkage_private">'<tt>private</tt>' Linkage</a></li>
26 <li><a href="#linkage_linker_private">'<tt>linker_private</tt>' Linkage</a></li>
Bill Wendling5e721d72010-07-01 21:55:59 +0000[diff] [blame]27 <li><a href="#linkage_linker_private_weak">'<tt>linker_private_weak</tt>' Linkage</a></li>
Bill Wendling55ae5152010-08-20 22:05:50 +0000[diff] [blame]28 <li><a href="#linkage_linker_private_weak_def_auto">'<tt>linker_private_weak_def_auto</tt>' Linkage</a></li>
Bill Wendling987e7eb2009-07-20 02:41:50 +0000[diff] [blame]29 <li><a href="#linkage_internal">'<tt>internal</tt>' Linkage</a></li>
30 <li><a href="#linkage_available_externally">'<tt>available_externally</tt>' Linkage</a></li>
31 <li><a href="#linkage_linkonce">'<tt>linkonce</tt>' Linkage</a></li>
32 <li><a href="#linkage_common">'<tt>common</tt>' Linkage</a></li>
33 <li><a href="#linkage_weak">'<tt>weak</tt>' Linkage</a></li>
34 <li><a href="#linkage_appending">'<tt>appending</tt>' Linkage</a></li>
35 <li><a href="#linkage_externweak">'<tt>extern_weak</tt>' Linkage</a></li>
Chris Lattner5a2d8752009-10-10 18:26:06 +0000[diff] [blame]36 <li><a href="#linkage_linkonce_odr">'<tt>linkonce_odr</tt>' Linkage</a></li>
Bill Wendling987e7eb2009-07-20 02:41:50 +0000[diff] [blame]37 <li><a href="#linkage_weak">'<tt>weak_odr</tt>' Linkage</a></li>
38 <li><a href="#linkage_external">'<tt>externally visible</tt>' Linkage</a></li>
39 <li><a href="#linkage_dllimport">'<tt>dllimport</tt>' Linkage</a></li>
40 <li><a href="#linkage_dllexport">'<tt>dllexport</tt>' Linkage</a></li>
Bill Wendling3d10a5a2009-07-20 01:03:30 +0000[diff] [blame]41 </ol>
42 </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]43 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]44 <li><a href="#namedtypes">Named Types</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]45 <li><a href="#globalvars">Global Variables</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]46 <li><a href="#functionstructure">Functions</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]47 <li><a href="#aliasstructure">Aliases</a></li>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]48 <li><a href="#namedmetadatastructure">Named Metadata</a></li>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]49 <li><a href="#paramattrs">Parameter Attributes</a></li>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]50 <li><a href="#fnattrs">Function Attributes</a></li>
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]51 <li><a href="#gc">Garbage Collector Names</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]52 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]53 <li><a href="#datalayout">Data Layout</a></li>
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]54 <li><a href="#pointeraliasing">Pointer Aliasing Rules</a></li>
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]55 <li><a href="#volatile">Volatile Memory Accesses</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]56 </ol>
57 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]58 <li><a href="#typesystem">Type System</a>
59 <ol>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]60 <li><a href="#t_classifications">Type Classifications</a></li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]61 <li><a href="#t_primitive">Primitive Types</a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]62 <ol>
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]63 <li><a href="#t_integer">Integer Type</a></li>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]64 <li><a href="#t_floating">Floating Point Types</a></li>
Dale Johannesen21fe99b2010-10-01 00:48:59 +0000[diff] [blame]65 <li><a href="#t_x86mmx">X86mmx Type</a></li>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]66 <li><a href="#t_void">Void Type</a></li>
67 <li><a href="#t_label">Label Type</a></li>
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]68 <li><a href="#t_metadata">Metadata Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]69 </ol>
70 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]71 <li><a href="#t_derived">Derived Types</a>
72 <ol>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]73 <li><a href="#t_aggregate">Aggregate Types</a>
74 <ol>
75 <li><a href="#t_array">Array Type</a></li>
76 <li><a href="#t_struct">Structure Type</a></li>
77 <li><a href="#t_pstruct">Packed Structure Type</a></li>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]78 <li><a href="#t_vector">Vector Type</a></li>
79 </ol>
80 </li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]81 <li><a href="#t_function">Function Type</a></li>
82 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]83 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]84 </ol>
85 </li>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]86 <li><a href="#t_uprefs">Type Up-references</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]87 </ol>
88 </li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]89 <li><a href="#constants">Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]90 <ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]91 <li><a href="#simpleconstants">Simple Constants</a></li>
Chris Lattner70882792009-02-28 18:32:25 +0000[diff] [blame]92 <li><a href="#complexconstants">Complex Constants</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]93 <li><a href="#globalconstants">Global Variable and Function Addresses</a></li>
94 <li><a href="#undefvalues">Undefined Values</a></li>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]95 <li><a href="#trapvalues">Trap Values</a></li>
Chris Lattnerf9d078e2009-10-27 21:19:13 +0000[diff] [blame]96 <li><a href="#blockaddress">Addresses of Basic Blocks</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]97 <li><a href="#constantexprs">Constant Expressions</a></li>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]98 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]99 </li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]100 <li><a href="#othervalues">Other Values</a>
101 <ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]102 <li><a href="#inlineasm">Inline Assembler Expressions</a></li>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]103 <li><a href="#metadata">Metadata Nodes and Metadata Strings</a></li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]104 </ol>
105 </li>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]106 <li><a href="#intrinsic_globals">Intrinsic Global Variables</a>
107 <ol>
108 <li><a href="#intg_used">The '<tt>llvm.used</tt>' Global Variable</a></li>
Chris Lattner401e10c2009-07-20 06:14:25 +0000[diff] [blame]109 <li><a href="#intg_compiler_used">The '<tt>llvm.compiler.used</tt>'
110 Global Variable</a></li>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]111 <li><a href="#intg_global_ctors">The '<tt>llvm.global_ctors</tt>'
112 Global Variable</a></li>
113 <li><a href="#intg_global_dtors">The '<tt>llvm.global_dtors</tt>'
114 Global Variable</a></li>
115 </ol>
116 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]117 <li><a href="#instref">Instruction Reference</a>
118 <ol>
119 <li><a href="#terminators">Terminator Instructions</a>
120 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]121 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
122 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]123 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]124 <li><a href="#i_indirectbr">'<tt>indirectbr</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]125 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]126 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]127 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]128 </ol>
129 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]130 <li><a href="#binaryops">Binary Operations</a>
131 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]132 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]133 <li><a href="#i_fadd">'<tt>fadd</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]134 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]135 <li><a href="#i_fsub">'<tt>fsub</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]136 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]137 <li><a href="#i_fmul">'<tt>fmul</tt>' Instruction</a></li>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]138 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
139 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
140 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]141 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
142 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
143 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]144 </ol>
145 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]146 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
147 <ol>
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]148 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
149 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
150 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]151 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]152 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]153 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]154 </ol>
155 </li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]156 <li><a href="#vectorops">Vector Operations</a>
157 <ol>
158 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
159 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
160 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]161 </ol>
162 </li>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]163 <li><a href="#aggregateops">Aggregate Operations</a>
164 <ol>
165 <li><a href="#i_extractvalue">'<tt>extractvalue</tt>' Instruction</a></li>
166 <li><a href="#i_insertvalue">'<tt>insertvalue</tt>' Instruction</a></li>
167 </ol>
168 </li>
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]169 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]170 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]171 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]172 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
173 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
174 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]175 </ol>
176 </li>
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]177 <li><a href="#convertops">Conversion Operations</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]178 <ol>
179 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
180 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
181 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
182 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
183 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]184 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
185 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
186 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
187 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]188 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
189 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]190 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]191 </ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]192 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]193 <li><a href="#otherops">Other Operations</a>
194 <ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]195 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
196 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]197 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]198 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]199 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]200 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]201 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]202 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]203 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]204 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]205 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]206 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]207 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
208 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]209 <li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
210 <li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
211 <li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]212 </ol>
213 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]214 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
215 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]216 <li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
217 <li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
218 <li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]219 </ol>
220 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]221 <li><a href="#int_codegen">Code Generator Intrinsics</a>
222 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]223 <li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
224 <li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
225 <li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
226 <li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
227 <li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
228 <li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
Dan Gohman31f1af12010-05-26 21:56:15 +0000[diff] [blame]229 <li><a href="#int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]230 </ol>
231 </li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]232 <li><a href="#int_libc">Standard C Library Intrinsics</a>
233 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]234 <li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
235 <li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
236 <li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
237 <li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
238 <li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]239 <li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
240 <li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
241 <li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]242 </ol>
243 </li>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]244 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]245 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]246 <li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]247 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
248 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
249 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]250 </ol>
251 </li>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]252 <li><a href="#int_overflow">Arithmetic with Overflow Intrinsics</a>
253 <ol>
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]254 <li><a href="#int_sadd_overflow">'<tt>llvm.sadd.with.overflow.*</tt> Intrinsics</a></li>
255 <li><a href="#int_uadd_overflow">'<tt>llvm.uadd.with.overflow.*</tt> Intrinsics</a></li>
256 <li><a href="#int_ssub_overflow">'<tt>llvm.ssub.with.overflow.*</tt> Intrinsics</a></li>
257 <li><a href="#int_usub_overflow">'<tt>llvm.usub.with.overflow.*</tt> Intrinsics</a></li>
258 <li><a href="#int_smul_overflow">'<tt>llvm.smul.with.overflow.*</tt> Intrinsics</a></li>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]259 <li><a href="#int_umul_overflow">'<tt>llvm.umul.with.overflow.*</tt> Intrinsics</a></li>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]260 </ol>
261 </li>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]262 <li><a href="#int_fp16">Half Precision Floating Point Intrinsics</a>
263 <ol>
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]264 <li><a href="#int_convert_to_fp16">'<tt>llvm.convert.to.fp16</tt>' Intrinsic</a></li>
265 <li><a href="#int_convert_from_fp16">'<tt>llvm.convert.from.fp16</tt>' Intrinsic</a></li>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]266 </ol>
267 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]268 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]269 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]270 <li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]271 <ol>
272 <li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]273 </ol>
274 </li>
Bill Wendling3c44f5b2008-11-18 22:10:53 +0000[diff] [blame]275 <li><a href="#int_atomics">Atomic intrinsics</a>
276 <ol>
277 <li><a href="#int_memory_barrier"><tt>llvm.memory_barrier</tt></a></li>
278 <li><a href="#int_atomic_cmp_swap"><tt>llvm.atomic.cmp.swap</tt></a></li>
279 <li><a href="#int_atomic_swap"><tt>llvm.atomic.swap</tt></a></li>
280 <li><a href="#int_atomic_load_add"><tt>llvm.atomic.load.add</tt></a></li>
281 <li><a href="#int_atomic_load_sub"><tt>llvm.atomic.load.sub</tt></a></li>
282 <li><a href="#int_atomic_load_and"><tt>llvm.atomic.load.and</tt></a></li>
283 <li><a href="#int_atomic_load_nand"><tt>llvm.atomic.load.nand</tt></a></li>
284 <li><a href="#int_atomic_load_or"><tt>llvm.atomic.load.or</tt></a></li>
285 <li><a href="#int_atomic_load_xor"><tt>llvm.atomic.load.xor</tt></a></li>
286 <li><a href="#int_atomic_load_max"><tt>llvm.atomic.load.max</tt></a></li>
287 <li><a href="#int_atomic_load_min"><tt>llvm.atomic.load.min</tt></a></li>
288 <li><a href="#int_atomic_load_umax"><tt>llvm.atomic.load.umax</tt></a></li>
289 <li><a href="#int_atomic_load_umin"><tt>llvm.atomic.load.umin</tt></a></li>
290 </ol>
291 </li>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]292 <li><a href="#int_memorymarkers">Memory Use Markers</a>
293 <ol>
294 <li><a href="#int_lifetime_start"><tt>llvm.lifetime.start</tt></a></li>
295 <li><a href="#int_lifetime_end"><tt>llvm.lifetime.end</tt></a></li>
296 <li><a href="#int_invariant_start"><tt>llvm.invariant.start</tt></a></li>
297 <li><a href="#int_invariant_end"><tt>llvm.invariant.end</tt></a></li>
298 </ol>
299 </li>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]300 <li><a href="#int_general">General intrinsics</a>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]301 <ol>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]302 <li><a href="#int_var_annotation">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]303 '<tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]304 <li><a href="#int_annotation">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]305 '<tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]306 <li><a href="#int_trap">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]307 '<tt>llvm.trap</tt>' Intrinsic</a></li>
308 <li><a href="#int_stackprotector">
309 '<tt>llvm.stackprotector</tt>' Intrinsic</a></li>
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]310 <li><a href="#int_objectsize">
311 '<tt>llvm.objectsize</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]312 </ol>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]313 </li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]314 </ol>
315 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]316</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]317
318<div class="doc_author">
319 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
320 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]321</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]322
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]323<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]324<div class="doc_section"> <a name="abstract">Abstract </a></div>
325<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]326
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]327<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]328
329<p>This document is a reference manual for the LLVM assembly language. LLVM is
330 a Static Single Assignment (SSA) based representation that provides type
331 safety, low-level operations, flexibility, and the capability of representing
332 'all' high-level languages cleanly. It is the common code representation
333 used throughout all phases of the LLVM compilation strategy.</p>
334
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]335</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]336
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]337<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]338<div class="doc_section"> <a name="introduction">Introduction</a> </div>
339<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]340
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]341<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]342
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]343<p>The LLVM code representation is designed to be used in three different forms:
344 as an in-memory compiler IR, as an on-disk bitcode representation (suitable
345 for fast loading by a Just-In-Time compiler), and as a human readable
346 assembly language representation. This allows LLVM to provide a powerful
347 intermediate representation for efficient compiler transformations and
348 analysis, while providing a natural means to debug and visualize the
349 transformations. The three different forms of LLVM are all equivalent. This
350 document describes the human readable representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]351
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]352<p>The LLVM representation aims to be light-weight and low-level while being
353 expressive, typed, and extensible at the same time. It aims to be a
354 "universal IR" of sorts, by being at a low enough level that high-level ideas
355 may be cleanly mapped to it (similar to how microprocessors are "universal
356 IR's", allowing many source languages to be mapped to them). By providing
357 type information, LLVM can be used as the target of optimizations: for
358 example, through pointer analysis, it can be proven that a C automatic
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]359 variable is never accessed outside of the current function, allowing it to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]360 be promoted to a simple SSA value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]361
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]362</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]363
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]364<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]365<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]366
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]367<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]368
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]369<p>It is important to note that this document describes 'well formed' LLVM
370 assembly language. There is a difference between what the parser accepts and
371 what is considered 'well formed'. For example, the following instruction is
372 syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]373
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]374<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]375%x = <a href="#i_add">add</a> i32 1, %x
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]376</pre>
377
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]378<p>because the definition of <tt>%x</tt> does not dominate all of its uses. The
379 LLVM infrastructure provides a verification pass that may be used to verify
380 that an LLVM module is well formed. This pass is automatically run by the
381 parser after parsing input assembly and by the optimizer before it outputs
382 bitcode. The violations pointed out by the verifier pass indicate bugs in
383 transformation passes or input to the parser.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]384
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]385</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]386
Chris Lattnercc689392007-10-03 17:34:29 +0000[diff] [blame]387<!-- Describe the typesetting conventions here. -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]388
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]389<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]390<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]391<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]392
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]393<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]394
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]395<p>LLVM identifiers come in two basic types: global and local. Global
396 identifiers (functions, global variables) begin with the <tt>'@'</tt>
397 character. Local identifiers (register names, types) begin with
398 the <tt>'%'</tt> character. Additionally, there are three different formats
399 for identifiers, for different purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]400
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]401<ol>
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]402 <li>Named values are represented as a string of characters with their prefix.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]403 For example, <tt>%foo</tt>, <tt>@DivisionByZero</tt>,
404 <tt>%a.really.long.identifier</tt>. The actual regular expression used is
405 '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'. Identifiers which require
406 other characters in their names can be surrounded with quotes. Special
407 characters may be escaped using <tt>"\xx"</tt> where <tt>xx</tt> is the
408 ASCII code for the character in hexadecimal. In this way, any character
409 can be used in a name value, even quotes themselves.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]410
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]411 <li>Unnamed values are represented as an unsigned numeric value with their
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]412 prefix. For example, <tt>%12</tt>, <tt>@2</tt>, <tt>%44</tt>.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]413
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]414 <li>Constants, which are described in a <a href="#constants">section about
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]415 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]416</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]417
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]418<p>LLVM requires that values start with a prefix for two reasons: Compilers
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]419 don't need to worry about name clashes with reserved words, and the set of
420 reserved words may be expanded in the future without penalty. Additionally,
421 unnamed identifiers allow a compiler to quickly come up with a temporary
422 variable without having to avoid symbol table conflicts.</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]423
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]424<p>Reserved words in LLVM are very similar to reserved words in other
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]425 languages. There are keywords for different opcodes
426 ('<tt><a href="#i_add">add</a></tt>',
427 '<tt><a href="#i_bitcast">bitcast</a></tt>',
428 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names
429 ('<tt><a href="#t_void">void</a></tt>',
430 '<tt><a href="#t_primitive">i32</a></tt>', etc...), and others. These
431 reserved words cannot conflict with variable names, because none of them
432 start with a prefix character (<tt>'%'</tt> or <tt>'@'</tt>).</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]433
434<p>Here is an example of LLVM code to multiply the integer variable
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]435 '<tt>%X</tt>' by 8:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]436
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]437<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]438
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]439<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]440%result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]441</pre>
442
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]443<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]444
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]445<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]446%result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]447</pre>
448
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]449<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]450
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]451<pre class="doc_code">
Gabor Greifec58f752009-10-28 13:05:07 +0000[diff] [blame]452%0 = <a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
453%1 = <a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]454%result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]455</pre>
456
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]457<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several important
458 lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]459
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]460<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]461 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]462 line.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]463
464 <li>Unnamed temporaries are created when the result of a computation is not
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]465 assigned to a named value.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]466
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]467 <li>Unnamed temporaries are numbered sequentially</li>
468</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]469
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]470<p>It also shows a convention that we follow in this document. When
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]471 demonstrating instructions, we will follow an instruction with a comment that
472 defines the type and name of value produced. Comments are shown in italic
473 text.</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]474
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]475</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]476
477<!-- *********************************************************************** -->
478<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
479<!-- *********************************************************************** -->
480
481<!-- ======================================================================= -->
482<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
483</div>
484
485<div class="doc_text">
486
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]487<p>LLVM programs are composed of "Module"s, each of which is a translation unit
488 of the input programs. Each module consists of functions, global variables,
489 and symbol table entries. Modules may be combined together with the LLVM
490 linker, which merges function (and global variable) definitions, resolves
491 forward declarations, and merges symbol table entries. Here is an example of
492 the "hello world" module:</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]493
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]494<pre class="doc_code">
Chris Lattner63e4ccb2010-08-17 17:13:42 +0000[diff] [blame]495<i>; Declare the string constant as a global constant.</i>&nbsp;
496<a href="#identifiers">@.LC0</a> = <a href="#linkage_internal">internal</a> <a href="#globalvars">constant</a> <a href="#t_array">[13 x i8]</a> c"hello world\0A\00" <i>; [13 x i8]*</i>&nbsp;
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]497
Chris Lattner63e4ccb2010-08-17 17:13:42 +0000[diff] [blame]498<i>; External declaration of the puts function</i>&nbsp;
499<a href="#functionstructure">declare</a> i32 @puts(i8*) <i>; i32 (i8*)* </i>&nbsp;
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]500
501<i>; Definition of main function</i>
Chris Lattner63e4ccb2010-08-17 17:13:42 +0000[diff] [blame]502define i32 @main() { <i>; i32()* </i>&nbsp;
503 <i>; Convert [13 x i8]* to i8 *...</i>&nbsp;
504 %cast210 = <a href="#i_getelementptr">getelementptr</a> [13 x i8]* @.LC0, i64 0, i64 0 <i>; i8*</i>&nbsp;
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]505
Chris Lattner63e4ccb2010-08-17 17:13:42 +0000[diff] [blame]506 <i>; Call puts function to write out the string to stdout.</i>&nbsp;
507 <a href="#i_call">call</a> i32 @puts(i8* %cast210) <i>; i32</i>&nbsp;
508 <a href="#i_ret">ret</a> i32 0&nbsp;
509}
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]510
511<i>; Named metadata</i>
512!1 = metadata !{i32 41}
513!foo = !{!1, null}
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]514</pre>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]515
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]516<p>This example is made up of a <a href="#globalvars">global variable</a> named
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]517 "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>" function,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]518 a <a href="#functionstructure">function definition</a> for
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]519 "<tt>main</tt>" and <a href="#namedmetadatastructure">named metadata</a>
520 "<tt>foo"</tt>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]521
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]522<p>In general, a module is made up of a list of global values, where both
523 functions and global variables are global values. Global values are
524 represented by a pointer to a memory location (in this case, a pointer to an
525 array of char, and a pointer to a function), and have one of the
526 following <a href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]527
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]528</div>
529
530<!-- ======================================================================= -->
531<div class="doc_subsection">
532 <a name="linkage">Linkage Types</a>
533</div>
534
535<div class="doc_text">
536
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]537<p>All Global Variables and Functions have one of the following types of
538 linkage:</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]539
540<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]541 <dt><tt><b><a name="linkage_private">private</a></b></tt></dt>
Bill Wendling5e721d72010-07-01 21:55:59 +0000[diff] [blame]542 <dd>Global values with "<tt>private</tt>" linkage are only directly accessible
543 by objects in the current module. In particular, linking code into a
544 module with an private global value may cause the private to be renamed as
545 necessary to avoid collisions. Because the symbol is private to the
546 module, all references can be updated. This doesn't show up in any symbol
547 table in the object file.</dd>
Rafael Espindolabb46f522009-01-15 20:18:42 +0000[diff] [blame]548
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]549 <dt><tt><b><a name="linkage_linker_private">linker_private</a></b></tt></dt>
Bill Wendling5e721d72010-07-01 21:55:59 +0000[diff] [blame]550 <dd>Similar to <tt>private</tt>, but the symbol is passed through the
551 assembler and evaluated by the linker. Unlike normal strong symbols, they
552 are removed by the linker from the final linked image (executable or
553 dynamic library).</dd>
554
555 <dt><tt><b><a name="linkage_linker_private_weak">linker_private_weak</a></b></tt></dt>
556 <dd>Similar to "<tt>linker_private</tt>", but the symbol is weak. Note that
557 <tt>linker_private_weak</tt> symbols are subject to coalescing by the
558 linker. The symbols are removed by the linker from the final linked image
559 (executable or dynamic library).</dd>
Bill Wendling3d10a5a2009-07-20 01:03:30 +0000[diff] [blame]560
Bill Wendling55ae5152010-08-20 22:05:50 +0000[diff] [blame]561 <dt><tt><b><a name="linkage_linker_private_weak_def_auto">linker_private_weak_def_auto</a></b></tt></dt>
562 <dd>Similar to "<tt>linker_private_weak</tt>", but it's known that the address
563 of the object is not taken. For instance, functions that had an inline
564 definition, but the compiler decided not to inline it. Note,
565 unlike <tt>linker_private</tt> and <tt>linker_private_weak</tt>,
566 <tt>linker_private_weak_def_auto</tt> may have only <tt>default</tt>
567 visibility. The symbols are removed by the linker from the final linked
568 image (executable or dynamic library).</dd>
569
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]570 <dt><tt><b><a name="linkage_internal">internal</a></b></tt></dt>
Bill Wendling07d31772010-06-29 22:34:52 +0000[diff] [blame]571 <dd>Similar to private, but the value shows as a local symbol
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]572 (<tt>STB_LOCAL</tt> in the case of ELF) in the object file. This
573 corresponds to the notion of the '<tt>static</tt>' keyword in C.</dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]574
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]575 <dt><tt><b><a name="linkage_available_externally">available_externally</a></b></tt></dt>
Chris Lattner266c7bb2009-04-13 05:44:34 +0000[diff] [blame]576 <dd>Globals with "<tt>available_externally</tt>" linkage are never emitted
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]577 into the object file corresponding to the LLVM module. They exist to
578 allow inlining and other optimizations to take place given knowledge of
579 the definition of the global, which is known to be somewhere outside the
580 module. Globals with <tt>available_externally</tt> linkage are allowed to
581 be discarded at will, and are otherwise the same as <tt>linkonce_odr</tt>.
582 This linkage type is only allowed on definitions, not declarations.</dd>
Chris Lattner266c7bb2009-04-13 05:44:34 +0000[diff] [blame]583
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]584 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt></dt>
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]585 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
Chris Lattner873187c2010-01-09 19:15:14 +0000[diff] [blame]586 the same name when linkage occurs. This can be used to implement
587 some forms of inline functions, templates, or other code which must be
588 generated in each translation unit that uses it, but where the body may
589 be overridden with a more definitive definition later. Unreferenced
590 <tt>linkonce</tt> globals are allowed to be discarded. Note that
591 <tt>linkonce</tt> linkage does not actually allow the optimizer to
592 inline the body of this function into callers because it doesn't know if
593 this definition of the function is the definitive definition within the
594 program or whether it will be overridden by a stronger definition.
595 To enable inlining and other optimizations, use "<tt>linkonce_odr</tt>"
596 linkage.</dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]597
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]598 <dt><tt><b><a name="linkage_weak">weak</a></b></tt></dt>
Chris Lattner26d054d2009-08-05 05:21:07 +0000[diff] [blame]599 <dd>"<tt>weak</tt>" linkage has the same merging semantics as
600 <tt>linkonce</tt> linkage, except that unreferenced globals with
601 <tt>weak</tt> linkage may not be discarded. This is used for globals that
602 are declared "weak" in C source code.</dd>
603
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]604 <dt><tt><b><a name="linkage_common">common</a></b></tt></dt>
Chris Lattner26d054d2009-08-05 05:21:07 +0000[diff] [blame]605 <dd>"<tt>common</tt>" linkage is most similar to "<tt>weak</tt>" linkage, but
606 they are used for tentative definitions in C, such as "<tt>int X;</tt>" at
607 global scope.
608 Symbols with "<tt>common</tt>" linkage are merged in the same way as
609 <tt>weak symbols</tt>, and they may not be deleted if unreferenced.
Chris Lattnercd81f5d2009-08-05 05:41:44 +0000[diff] [blame]610 <tt>common</tt> symbols may not have an explicit section,
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]611 must have a zero initializer, and may not be marked '<a
Chris Lattnercd81f5d2009-08-05 05:41:44 +0000[diff] [blame]612 href="#globalvars"><tt>constant</tt></a>'. Functions and aliases may not
613 have common linkage.</dd>
Chris Lattner26d054d2009-08-05 05:21:07 +0000[diff] [blame]614
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]615
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]616 <dt><tt><b><a name="linkage_appending">appending</a></b></tt></dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]617 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]618 pointer to array type. When two global variables with appending linkage
619 are linked together, the two global arrays are appended together. This is
620 the LLVM, typesafe, equivalent of having the system linker append together
621 "sections" with identical names when .o files are linked.</dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]622
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]623 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]624 <dd>The semantics of this linkage follow the ELF object file model: the symbol
625 is weak until linked, if not linked, the symbol becomes null instead of
626 being an undefined reference.</dd>
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]627
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]628 <dt><tt><b><a name="linkage_linkonce_odr">linkonce_odr</a></b></tt></dt>
629 <dt><tt><b><a name="linkage_weak_odr">weak_odr</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]630 <dd>Some languages allow differing globals to be merged, such as two functions
631 with different semantics. Other languages, such as <tt>C++</tt>, ensure
Bill Wendling5e721d72010-07-01 21:55:59 +0000[diff] [blame]632 that only equivalent globals are ever merged (the "one definition rule"
633 &mdash; "ODR"). Such languages can use the <tt>linkonce_odr</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]634 and <tt>weak_odr</tt> linkage types to indicate that the global will only
635 be merged with equivalent globals. These linkage types are otherwise the
636 same as their non-<tt>odr</tt> versions.</dd>
Duncan Sands667d4b82009-03-07 15:45:40 +0000[diff] [blame]637
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]638 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]639 <dd>If none of the above identifiers are used, the global is externally
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]640 visible, meaning that it participates in linkage and can be used to
641 resolve external symbol references.</dd>
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]642</dl>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]643
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]644<p>The next two types of linkage are targeted for Microsoft Windows platform
645 only. They are designed to support importing (exporting) symbols from (to)
646 DLLs (Dynamic Link Libraries).</p>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]647
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]648<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]649 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt></dt>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]650 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]651 or variable via a global pointer to a pointer that is set up by the DLL
652 exporting the symbol. On Microsoft Windows targets, the pointer name is
653 formed by combining <code>__imp_</code> and the function or variable
654 name.</dd>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]655
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]656 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt></dt>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]657 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]658 pointer to a pointer in a DLL, so that it can be referenced with the
659 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
660 name is formed by combining <code>__imp_</code> and the function or
661 variable name.</dd>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]662</dl>
663
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]664<p>For example, since the "<tt>.LC0</tt>" variable is defined to be internal, if
665 another module defined a "<tt>.LC0</tt>" variable and was linked with this
666 one, one of the two would be renamed, preventing a collision. Since
667 "<tt>main</tt>" and "<tt>puts</tt>" are external (i.e., lacking any linkage
668 declarations), they are accessible outside of the current module.</p>
669
670<p>It is illegal for a function <i>declaration</i> to have any linkage type
671 other than "externally visible", <tt>dllimport</tt>
672 or <tt>extern_weak</tt>.</p>
673
Duncan Sands667d4b82009-03-07 15:45:40 +0000[diff] [blame]674<p>Aliases can have only <tt>external</tt>, <tt>internal</tt>, <tt>weak</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]675 or <tt>weak_odr</tt> linkages.</p>
676
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]677</div>
678
679<!-- ======================================================================= -->
680<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]681 <a name="callingconv">Calling Conventions</a>
682</div>
683
684<div class="doc_text">
685
686<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]687 and <a href="#i_invoke">invokes</a> can all have an optional calling
688 convention specified for the call. The calling convention of any pair of
689 dynamic caller/callee must match, or the behavior of the program is
690 undefined. The following calling conventions are supported by LLVM, and more
691 may be added in the future:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]692
693<dl>
694 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]695 <dd>This calling convention (the default if no other calling convention is
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]696 specified) matches the target C calling conventions. This calling
697 convention supports varargs function calls and tolerates some mismatch in
698 the declared prototype and implemented declaration of the function (as
699 does normal C).</dd>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]700
701 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]702 <dd>This calling convention attempts to make calls as fast as possible
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]703 (e.g. by passing things in registers). This calling convention allows the
704 target to use whatever tricks it wants to produce fast code for the
705 target, without having to conform to an externally specified ABI
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]706 (Application Binary Interface).
707 <a href="CodeGenerator.html#tailcallopt">Tail calls can only be optimized
Chris Lattner29689432010-03-11 00:22:57 +0000[diff] [blame]708 when this or the GHC convention is used.</a> This calling convention
709 does not support varargs and requires the prototype of all callees to
710 exactly match the prototype of the function definition.</dd>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]711
712 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]713 <dd>This calling convention attempts to make code in the caller as efficient
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]714 as possible under the assumption that the call is not commonly executed.
715 As such, these calls often preserve all registers so that the call does
716 not break any live ranges in the caller side. This calling convention
717 does not support varargs and requires the prototype of all callees to
718 exactly match the prototype of the function definition.</dd>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]719
Chris Lattner29689432010-03-11 00:22:57 +0000[diff] [blame]720 <dt><b>"<tt>cc <em>10</em></tt>" - GHC convention</b>:</dt>
721 <dd>This calling convention has been implemented specifically for use by the
722 <a href="http://www.haskell.org/ghc">Glasgow Haskell Compiler (GHC)</a>.
723 It passes everything in registers, going to extremes to achieve this by
724 disabling callee save registers. This calling convention should not be
725 used lightly but only for specific situations such as an alternative to
726 the <em>register pinning</em> performance technique often used when
727 implementing functional programming languages.At the moment only X86
728 supports this convention and it has the following limitations:
729 <ul>
730 <li>On <em>X86-32</em> only supports up to 4 bit type parameters. No
731 floating point types are supported.</li>
732 <li>On <em>X86-64</em> only supports up to 10 bit type parameters and
733 6 floating point parameters.</li>
734 </ul>
735 This calling convention supports
736 <a href="CodeGenerator.html#tailcallopt">tail call optimization</a> but
737 requires both the caller and callee are using it.
738 </dd>
739
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]740 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]741 <dd>Any calling convention may be specified by number, allowing
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]742 target-specific calling conventions to be used. Target specific calling
743 conventions start at 64.</dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]744</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]745
746<p>More calling conventions can be added/defined on an as-needed basis, to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]747 support Pascal conventions or any other well-known target-independent
748 convention.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]749
750</div>
751
752<!-- ======================================================================= -->
753<div class="doc_subsection">
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]754 <a name="visibility">Visibility Styles</a>
755</div>
756
757<div class="doc_text">
758
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]759<p>All Global Variables and Functions have one of the following visibility
760 styles:</p>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]761
762<dl>
763 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]764 <dd>On targets that use the ELF object file format, default visibility means
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]765 that the declaration is visible to other modules and, in shared libraries,
766 means that the declared entity may be overridden. On Darwin, default
767 visibility means that the declaration is visible to other modules. Default
768 visibility corresponds to "external linkage" in the language.</dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]769
770 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]771 <dd>Two declarations of an object with hidden visibility refer to the same
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]772 object if they are in the same shared object. Usually, hidden visibility
773 indicates that the symbol will not be placed into the dynamic symbol
774 table, so no other module (executable or shared library) can reference it
775 directly.</dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]776
Anton Korobeynikov6f9896f2007-04-29 18:35:00 +0000[diff] [blame]777 <dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
Anton Korobeynikov6f9896f2007-04-29 18:35:00 +0000[diff] [blame]778 <dd>On ELF, protected visibility indicates that the symbol will be placed in
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]779 the dynamic symbol table, but that references within the defining module
780 will bind to the local symbol. That is, the symbol cannot be overridden by
781 another module.</dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]782</dl>
783
784</div>
785
786<!-- ======================================================================= -->
787<div class="doc_subsection">
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]788 <a name="namedtypes">Named Types</a>
789</div>
790
791<div class="doc_text">
792
793<p>LLVM IR allows you to specify name aliases for certain types. This can make
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]794 it easier to read the IR and make the IR more condensed (particularly when
795 recursive types are involved). An example of a name specification is:</p>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]796
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]797<pre class="doc_code">
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]798%mytype = type { %mytype*, i32 }
799</pre>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]800
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]801<p>You may give a name to any <a href="#typesystem">type</a> except
Chris Lattnerdc65f222010-08-17 23:26:04 +0000[diff] [blame]802 "<a href="#t_void">void</a>". Type name aliases may be used anywhere a type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]803 is expected with the syntax "%mytype".</p>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]804
805<p>Note that type names are aliases for the structural type that they indicate,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]806 and that you can therefore specify multiple names for the same type. This
807 often leads to confusing behavior when dumping out a .ll file. Since LLVM IR
808 uses structural typing, the name is not part of the type. When printing out
809 LLVM IR, the printer will pick <em>one name</em> to render all types of a
810 particular shape. This means that if you have code where two different
811 source types end up having the same LLVM type, that the dumper will sometimes
812 print the "wrong" or unexpected type. This is an important design point and
813 isn't going to change.</p>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]814
815</div>
816
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]817<!-- ======================================================================= -->
818<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]819 <a name="globalvars">Global Variables</a>
820</div>
821
822<div class="doc_text">
823
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]824<p>Global variables define regions of memory allocated at compilation time
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]825 instead of run-time. Global variables may optionally be initialized, may
826 have an explicit section to be placed in, and may have an optional explicit
827 alignment specified. A variable may be defined as "thread_local", which
828 means that it will not be shared by threads (each thread will have a
829 separated copy of the variable). A variable may be defined as a global
830 "constant," which indicates that the contents of the variable
831 will <b>never</b> be modified (enabling better optimization, allowing the
832 global data to be placed in the read-only section of an executable, etc).
833 Note that variables that need runtime initialization cannot be marked
834 "constant" as there is a store to the variable.</p>
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]835
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]836<p>LLVM explicitly allows <em>declarations</em> of global variables to be marked
837 constant, even if the final definition of the global is not. This capability
838 can be used to enable slightly better optimization of the program, but
839 requires the language definition to guarantee that optimizations based on the
840 'constantness' are valid for the translation units that do not include the
841 definition.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]842
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]843<p>As SSA values, global variables define pointer values that are in scope
844 (i.e. they dominate) all basic blocks in the program. Global variables
845 always define a pointer to their "content" type because they describe a
846 region of memory, and all memory objects in LLVM are accessed through
847 pointers.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]848
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]849<p>A global variable may be declared to reside in a target-specific numbered
850 address space. For targets that support them, address spaces may affect how
851 optimizations are performed and/or what target instructions are used to
852 access the variable. The default address space is zero. The address space
853 qualifier must precede any other attributes.</p>
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]854
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]855<p>LLVM allows an explicit section to be specified for globals. If the target
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]856 supports it, it will emit globals to the section specified.</p>
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]857
Chris Lattnerce99fa92010-04-28 00:13:42 +0000[diff] [blame]858<p>An explicit alignment may be specified for a global, which must be a power
859 of 2. If not present, or if the alignment is set to zero, the alignment of
860 the global is set by the target to whatever it feels convenient. If an
861 explicit alignment is specified, the global is forced to have exactly that
Chris Lattner2d4b8ee2010-04-28 00:31:12 +0000[diff] [blame]862 alignment. Targets and optimizers are not allowed to over-align the global
863 if the global has an assigned section. In this case, the extra alignment
864 could be observable: for example, code could assume that the globals are
865 densely packed in their section and try to iterate over them as an array,
866 alignment padding would break this iteration.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]867
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]868<p>For example, the following defines a global in a numbered address space with
869 an initializer, section, and alignment:</p>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]870
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]871<pre class="doc_code">
Dan Gohman398873c2009-01-11 00:40:00 +0000[diff] [blame]872@G = addrspace(5) constant float 1.0, section "foo", align 4
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]873</pre>
874
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]875</div>
876
877
878<!-- ======================================================================= -->
879<div class="doc_subsection">
880 <a name="functionstructure">Functions</a>
881</div>
882
883<div class="doc_text">
884
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]885<p>LLVM function definitions consist of the "<tt>define</tt>" keyword, an
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]886 optional <a href="#linkage">linkage type</a>, an optional
887 <a href="#visibility">visibility style</a>, an optional
888 <a href="#callingconv">calling convention</a>, a return type, an optional
889 <a href="#paramattrs">parameter attribute</a> for the return type, a function
890 name, a (possibly empty) argument list (each with optional
891 <a href="#paramattrs">parameter attributes</a>), optional
892 <a href="#fnattrs">function attributes</a>, an optional section, an optional
893 alignment, an optional <a href="#gc">garbage collector name</a>, an opening
894 curly brace, a list of basic blocks, and a closing curly brace.</p>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]895
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]896<p>LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
897 optional <a href="#linkage">linkage type</a>, an optional
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]898 <a href="#visibility">visibility style</a>, an optional
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]899 <a href="#callingconv">calling convention</a>, a return type, an optional
900 <a href="#paramattrs">parameter attribute</a> for the return type, a function
901 name, a possibly empty list of arguments, an optional alignment, and an
902 optional <a href="#gc">garbage collector name</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]903
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]904<p>A function definition contains a list of basic blocks, forming the CFG
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]905 (Control Flow Graph) for the function. Each basic block may optionally start
906 with a label (giving the basic block a symbol table entry), contains a list
907 of instructions, and ends with a <a href="#terminators">terminator</a>
908 instruction (such as a branch or function return).</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]909
Chris Lattner4a3c9012007-06-08 16:52:14 +0000[diff] [blame]910<p>The first basic block in a function is special in two ways: it is immediately
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]911 executed on entrance to the function, and it is not allowed to have
912 predecessor basic blocks (i.e. there can not be any branches to the entry
913 block of a function). Because the block can have no predecessors, it also
914 cannot have any <a href="#i_phi">PHI nodes</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]915
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]916<p>LLVM allows an explicit section to be specified for functions. If the target
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]917 supports it, it will emit functions to the section specified.</p>
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]918
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]919<p>An explicit alignment may be specified for a function. If not present, or if
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]920 the alignment is set to zero, the alignment of the function is set by the
921 target to whatever it feels convenient. If an explicit alignment is
922 specified, the function is forced to have at least that much alignment. All
923 alignments must be a power of 2.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]924
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]925<h5>Syntax:</h5>
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]926<pre class="doc_code">
Chris Lattner50ad45c2008-10-13 16:55:18 +0000[diff] [blame]927define [<a href="#linkage">linkage</a>] [<a href="#visibility">visibility</a>]
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]928 [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>]
929 &lt;ResultType&gt; @&lt;FunctionName&gt; ([argument list])
930 [<a href="#fnattrs">fn Attrs</a>] [section "name"] [align N]
931 [<a href="#gc">gc</a>] { ... }
932</pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]933
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]934</div>
935
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]936<!-- ======================================================================= -->
937<div class="doc_subsection">
938 <a name="aliasstructure">Aliases</a>
939</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]940
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]941<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]942
943<p>Aliases act as "second name" for the aliasee value (which can be either
944 function, global variable, another alias or bitcast of global value). Aliases
945 may have an optional <a href="#linkage">linkage type</a>, and an
946 optional <a href="#visibility">visibility style</a>.</p>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]947
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]948<h5>Syntax:</h5>
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]949<pre class="doc_code">
Duncan Sands0b23ac12008-09-12 20:48:21 +0000[diff] [blame]950@&lt;Name&gt; = alias [Linkage] [Visibility] &lt;AliaseeTy&gt; @&lt;Aliasee&gt;
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]951</pre>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]952
953</div>
954
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]955<!-- ======================================================================= -->
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]956<div class="doc_subsection">
957 <a name="namedmetadatastructure">Named Metadata</a>
958</div>
959
960<div class="doc_text">
961
Chris Lattnere6a5ddd2010-01-15 21:50:19 +0000[diff] [blame]962<p>Named metadata is a collection of metadata. <a href="#metadata">Metadata
Dan Gohman872814a2010-07-21 18:54:18 +0000[diff] [blame]963 nodes</a> (but not metadata strings) are the only valid operands for
Chris Lattnere6a5ddd2010-01-15 21:50:19 +0000[diff] [blame]964 a named metadata.</p>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]965
966<h5>Syntax:</h5>
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]967<pre class="doc_code">
Dan Gohman872814a2010-07-21 18:54:18 +0000[diff] [blame]968; Some unnamed metadata nodes, which are referenced by the named metadata.
969!0 = metadata !{metadata !"zero"}
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]970!1 = metadata !{metadata !"one"}
Dan Gohman872814a2010-07-21 18:54:18 +0000[diff] [blame]971!2 = metadata !{metadata !"two"}
Dan Gohman1005bc52010-07-13 19:48:13 +0000[diff] [blame]972; A named metadata.
Dan Gohman872814a2010-07-21 18:54:18 +0000[diff] [blame]973!name = !{!0, !1, !2}
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]974</pre>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]975
976</div>
977
978<!-- ======================================================================= -->
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]979<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]980
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]981<div class="doc_text">
982
983<p>The return type and each parameter of a function type may have a set of
984 <i>parameter attributes</i> associated with them. Parameter attributes are
985 used to communicate additional information about the result or parameters of
986 a function. Parameter attributes are considered to be part of the function,
987 not of the function type, so functions with different parameter attributes
988 can have the same function type.</p>
989
990<p>Parameter attributes are simple keywords that follow the type specified. If
991 multiple parameter attributes are needed, they are space separated. For
992 example:</p>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]993
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]994<pre class="doc_code">
Nick Lewyckyb6a7d252009-02-15 23:06:14 +0000[diff] [blame]995declare i32 @printf(i8* noalias nocapture, ...)
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]996declare i32 @atoi(i8 zeroext)
997declare signext i8 @returns_signed_char()
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]998</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]999
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1000<p>Note that any attributes for the function result (<tt>nounwind</tt>,
1001 <tt>readonly</tt>) come immediately after the argument list.</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1002
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1003<p>Currently, only the following parameter attributes are defined:</p>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]1004
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1005<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1006 <dt><tt><b>zeroext</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1007 <dd>This indicates to the code generator that the parameter or return value
1008 should be zero-extended to a 32-bit value by the caller (for a parameter)
1009 or the callee (for a return value).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]1010
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1011 <dt><tt><b>signext</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1012 <dd>This indicates to the code generator that the parameter or return value
1013 should be sign-extended to a 32-bit value by the caller (for a parameter)
1014 or the callee (for a return value).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]1015
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1016 <dt><tt><b>inreg</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1017 <dd>This indicates that this parameter or return value should be treated in a
1018 special target-dependent fashion during while emitting code for a function
1019 call or return (usually, by putting it in a register as opposed to memory,
1020 though some targets use it to distinguish between two different kinds of
1021 registers). Use of this attribute is target-specific.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]1022
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1023 <dt><tt><b><a name="byval">byval</a></b></tt></dt>
Chris Lattnera6fd81d2010-11-20 23:49:06 +0000[diff] [blame]1024 <dd><p>This indicates that the pointer parameter should really be passed by
1025 value to the function. The attribute implies that a hidden copy of the
1026 pointee
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1027 is made between the caller and the callee, so the callee is unable to
1028 modify the value in the callee. This attribute is only valid on LLVM
1029 pointer arguments. It is generally used to pass structs and arrays by
1030 value, but is also valid on pointers to scalars. The copy is considered
1031 to belong to the caller not the callee (for example,
1032 <tt><a href="#readonly">readonly</a></tt> functions should not write to
1033 <tt>byval</tt> parameters). This is not a valid attribute for return
Chris Lattnera6fd81d2010-11-20 23:49:06 +0000[diff] [blame]1034 values.</p>
1035
1036 <p>The byval attribute also supports specifying an alignment with
1037 the align attribute. It indicates the alignment of the stack slot to
1038 form and the known alignment of the pointer specified to the call site. If
1039 the alignment is not specified, then the code generator makes a
1040 target-specific assumption.</p></dd>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1041
Dan Gohmanff235352010-07-02 23:18:08 +0000[diff] [blame]1042 <dt><tt><b><a name="sret">sret</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1043 <dd>This indicates that the pointer parameter specifies the address of a
1044 structure that is the return value of the function in the source program.
1045 This pointer must be guaranteed by the caller to be valid: loads and
1046 stores to the structure may be assumed by the callee to not to trap. This
1047 may only be applied to the first parameter. This is not a valid attribute
1048 for return values. </dd>
1049
Dan Gohmanff235352010-07-02 23:18:08 +0000[diff] [blame]1050 <dt><tt><b><a name="noalias">noalias</a></b></tt></dt>
Dan Gohman1e109622010-07-02 18:41:32 +0000[diff] [blame]1051 <dd>This indicates that pointer values
1052 <a href="#pointeraliasing"><i>based</i></a> on the argument or return
Dan Gohmanefca7f92010-07-02 23:46:54 +0000[diff] [blame]1053 value do not alias pointer values which are not <i>based</i> on it,
1054 ignoring certain "irrelevant" dependencies.
1055 For a call to the parent function, dependencies between memory
1056 references from before or after the call and from those during the call
1057 are "irrelevant" to the <tt>noalias</tt> keyword for the arguments and
1058 return value used in that call.
Dan Gohman1e109622010-07-02 18:41:32 +0000[diff] [blame]1059 The caller shares the responsibility with the callee for ensuring that
1060 these requirements are met.
1061 For further details, please see the discussion of the NoAlias response in
Dan Gohmanff70fe42010-07-06 15:26:33 +0000[diff] [blame]1062 <a href="AliasAnalysis.html#MustMayNo">alias analysis</a>.<br>
1063<br>
John McCall191d4ee2010-07-06 21:07:14 +0000[diff] [blame]1064 Note that this definition of <tt>noalias</tt> is intentionally
1065 similar to the definition of <tt>restrict</tt> in C99 for function
Chris Lattner211244a2010-07-06 20:51:35 +0000[diff] [blame]1066 arguments, though it is slightly weaker.
Dan Gohmanff70fe42010-07-06 15:26:33 +0000[diff] [blame]1067<br>
1068 For function return values, C99's <tt>restrict</tt> is not meaningful,
1069 while LLVM's <tt>noalias</tt> is.
1070 </dd>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1071
Dan Gohmanff235352010-07-02 23:18:08 +0000[diff] [blame]1072 <dt><tt><b><a name="nocapture">nocapture</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1073 <dd>This indicates that the callee does not make any copies of the pointer
1074 that outlive the callee itself. This is not a valid attribute for return
1075 values.</dd>
1076
Dan Gohmanff235352010-07-02 23:18:08 +0000[diff] [blame]1077 <dt><tt><b><a name="nest">nest</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1078 <dd>This indicates that the pointer parameter can be excised using the
1079 <a href="#int_trampoline">trampoline intrinsics</a>. This is not a valid
1080 attribute for return values.</dd>
1081</dl>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1082
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1083</div>
1084
1085<!-- ======================================================================= -->
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1086<div class="doc_subsection">
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1087 <a name="gc">Garbage Collector Names</a>
1088</div>
1089
1090<div class="doc_text">
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1091
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1092<p>Each function may specify a garbage collector name, which is simply a
1093 string:</p>
1094
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]1095<pre class="doc_code">
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1096define void @f() gc "name" { ... }
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1097</pre>
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1098
1099<p>The compiler declares the supported values of <i>name</i>. Specifying a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1100 collector which will cause the compiler to alter its output in order to
1101 support the named garbage collection algorithm.</p>
1102
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1103</div>
1104
1105<!-- ======================================================================= -->
1106<div class="doc_subsection">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1107 <a name="fnattrs">Function Attributes</a>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1108</div>
1109
1110<div class="doc_text">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1111
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1112<p>Function attributes are set to communicate additional information about a
1113 function. Function attributes are considered to be part of the function, not
1114 of the function type, so functions with different parameter attributes can
1115 have the same function type.</p>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1116
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1117<p>Function attributes are simple keywords that follow the type specified. If
1118 multiple attributes are needed, they are space separated. For example:</p>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1119
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]1120<pre class="doc_code">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1121define void @f() noinline { ... }
1122define void @f() alwaysinline { ... }
1123define void @f() alwaysinline optsize { ... }
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1124define void @f() optsize { ... }
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1125</pre>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1126
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1127<dl>
Charles Davis1e063d12010-02-12 00:31:15 +0000[diff] [blame]1128 <dt><tt><b>alignstack(&lt;<em>n</em>&gt;)</b></tt></dt>
1129 <dd>This attribute indicates that, when emitting the prologue and epilogue,
1130 the backend should forcibly align the stack pointer. Specify the
1131 desired alignment, which must be a power of two, in parentheses.
1132
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1133 <dt><tt><b>alwaysinline</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1134 <dd>This attribute indicates that the inliner should attempt to inline this
1135 function into callers whenever possible, ignoring any active inlining size
1136 threshold for this caller.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1137
Charles Davis970bfcc2010-10-25 15:37:09 +0000[diff] [blame]1138 <dt><tt><b>hotpatch</b></tt></dt>
Charles Davis6f12e292010-10-25 16:29:03 +0000[diff] [blame]1139 <dd>This attribute indicates that the function should be 'hotpatchable',
Charles Davis0076d202010-10-25 19:07:39 +0000[diff] [blame]1140 meaning the function can be patched and/or hooked even while it is
1141 loaded into memory. On x86, the function prologue will be preceded
1142 by six bytes of padding and will begin with a two-byte instruction.
1143 Most of the functions in the Windows system DLLs in Windows XP SP2 or
1144 higher were compiled in this fashion.</dd>
Charles Davis970bfcc2010-10-25 15:37:09 +0000[diff] [blame]1145
Jakob Stoklund Olesen570a4a52010-02-06 01:16:28 +0000[diff] [blame]1146 <dt><tt><b>inlinehint</b></tt></dt>
1147 <dd>This attribute indicates that the source code contained a hint that inlining
1148 this function is desirable (such as the "inline" keyword in C/C++). It
1149 is just a hint; it imposes no requirements on the inliner.</dd>
1150
Nick Lewycky76ec37a2010-07-06 18:24:09 +0000[diff] [blame]1151 <dt><tt><b>naked</b></tt></dt>
1152 <dd>This attribute disables prologue / epilogue emission for the function.
1153 This can have very system-specific consequences.</dd>
1154
1155 <dt><tt><b>noimplicitfloat</b></tt></dt>
1156 <dd>This attributes disables implicit floating point instructions.</dd>
1157
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1158 <dt><tt><b>noinline</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1159 <dd>This attribute indicates that the inliner should never inline this
1160 function in any situation. This attribute may not be used together with
1161 the <tt>alwaysinline</tt> attribute.</dd>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1162
Nick Lewycky76ec37a2010-07-06 18:24:09 +0000[diff] [blame]1163 <dt><tt><b>noredzone</b></tt></dt>
1164 <dd>This attribute indicates that the code generator should not use a red
1165 zone, even if the target-specific ABI normally permits it.</dd>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1166
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1167 <dt><tt><b>noreturn</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1168 <dd>This function attribute indicates that the function never returns
1169 normally. This produces undefined behavior at runtime if the function
1170 ever does dynamically return.</dd>
Bill Wendling31359ba2008-11-13 01:02:51 +0000[diff] [blame]1171
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1172 <dt><tt><b>nounwind</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1173 <dd>This function attribute indicates that the function never returns with an
1174 unwind or exceptional control flow. If the function does unwind, its
1175 runtime behavior is undefined.</dd>
Bill Wendlingfbaa7ed2008-11-26 19:07:40 +0000[diff] [blame]1176
Nick Lewycky76ec37a2010-07-06 18:24:09 +0000[diff] [blame]1177 <dt><tt><b>optsize</b></tt></dt>
1178 <dd>This attribute suggests that optimization passes and code generator passes
1179 make choices that keep the code size of this function low, and otherwise
1180 do optimizations specifically to reduce code size.</dd>
1181
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1182 <dt><tt><b>readnone</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1183 <dd>This attribute indicates that the function computes its result (or decides
1184 to unwind an exception) based strictly on its arguments, without
1185 dereferencing any pointer arguments or otherwise accessing any mutable
1186 state (e.g. memory, control registers, etc) visible to caller functions.
1187 It does not write through any pointer arguments
1188 (including <tt><a href="#byval">byval</a></tt> arguments) and never
1189 changes any state visible to callers. This means that it cannot unwind
1190 exceptions by calling the <tt>C++</tt> exception throwing methods, but
1191 could use the <tt>unwind</tt> instruction.</dd>
Devang Patel5d96fda2009-06-12 19:45:19 +0000[diff] [blame]1192
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1193 <dt><tt><b><a name="readonly">readonly</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1194 <dd>This attribute indicates that the function does not write through any
1195 pointer arguments (including <tt><a href="#byval">byval</a></tt>
1196 arguments) or otherwise modify any state (e.g. memory, control registers,
1197 etc) visible to caller functions. It may dereference pointer arguments
1198 and read state that may be set in the caller. A readonly function always
1199 returns the same value (or unwinds an exception identically) when called
1200 with the same set of arguments and global state. It cannot unwind an
1201 exception by calling the <tt>C++</tt> exception throwing methods, but may
1202 use the <tt>unwind</tt> instruction.</dd>
Anton Korobeynikovc5ec8a72009-07-17 18:07:26 +0000[diff] [blame]1203
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1204 <dt><tt><b><a name="ssp">ssp</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1205 <dd>This attribute indicates that the function should emit a stack smashing
1206 protector. It is in the form of a "canary"&mdash;a random value placed on
1207 the stack before the local variables that's checked upon return from the
1208 function to see if it has been overwritten. A heuristic is used to
1209 determine if a function needs stack protectors or not.<br>
1210<br>
1211 If a function that has an <tt>ssp</tt> attribute is inlined into a
1212 function that doesn't have an <tt>ssp</tt> attribute, then the resulting
1213 function will have an <tt>ssp</tt> attribute.</dd>
1214
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1215 <dt><tt><b>sspreq</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1216 <dd>This attribute indicates that the function should <em>always</em> emit a
1217 stack smashing protector. This overrides
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1218 the <tt><a href="#ssp">ssp</a></tt> function attribute.<br>
1219<br>
1220 If a function that has an <tt>sspreq</tt> attribute is inlined into a
1221 function that doesn't have an <tt>sspreq</tt> attribute or which has
1222 an <tt>ssp</tt> attribute, then the resulting function will have
1223 an <tt>sspreq</tt> attribute.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1224</dl>
1225
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1226</div>
1227
1228<!-- ======================================================================= -->
1229<div class="doc_subsection">
Chris Lattner1eeeb0c2006-04-08 04:40:53 +0000[diff] [blame]1230 <a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1231</div>
1232
1233<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1234
1235<p>Modules may contain "module-level inline asm" blocks, which corresponds to
1236 the GCC "file scope inline asm" blocks. These blocks are internally
1237 concatenated by LLVM and treated as a single unit, but may be separated in
1238 the <tt>.ll</tt> file if desired. The syntax is very simple:</p>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1239
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]1240<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1241module asm "inline asm code goes here"
1242module asm "more can go here"
1243</pre>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1244
1245<p>The strings can contain any character by escaping non-printable characters.
1246 The escape sequence used is simply "\xx" where "xx" is the two digit hex code
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1247 for the number.</p>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1248
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1249<p>The inline asm code is simply printed to the machine code .s file when
1250 assembly code is generated.</p>
1251
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1252</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1253
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1254<!-- ======================================================================= -->
1255<div class="doc_subsection">
1256 <a name="datalayout">Data Layout</a>
1257</div>
1258
1259<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1260
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1261<p>A module may specify a target specific data layout string that specifies how
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1262 data is to be laid out in memory. The syntax for the data layout is
1263 simply:</p>
1264
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]1265<pre class="doc_code">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1266target datalayout = "<i>layout specification</i>"
1267</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1268
1269<p>The <i>layout specification</i> consists of a list of specifications
1270 separated by the minus sign character ('-'). Each specification starts with
1271 a letter and may include other information after the letter to define some
1272 aspect of the data layout. The specifications accepted are as follows:</p>
1273
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1274<dl>
1275 <dt><tt>E</tt></dt>
1276 <dd>Specifies that the target lays out data in big-endian form. That is, the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1277 bits with the most significance have the lowest address location.</dd>
1278
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1279 <dt><tt>e</tt></dt>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1280 <dd>Specifies that the target lays out data in little-endian form. That is,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1281 the bits with the least significance have the lowest address
1282 location.</dd>
1283
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1284 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1285 <dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1286 <i>preferred</i> alignments. All sizes are in bits. Specifying
1287 the <i>pref</i> alignment is optional. If omitted, the
1288 preceding <tt>:</tt> should be omitted too.</dd>
1289
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1290 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1291 <dd>This specifies the alignment for an integer type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1292 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
1293
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1294 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1295 <dd>This specifies the alignment for a vector type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1296 <i>size</i>.</dd>
1297
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1298 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1299 <dd>This specifies the alignment for a floating point type of a given bit
Dale Johannesen9d8d2212010-05-28 18:54:47 +0000[diff] [blame]1300 <i>size</i>. Only values of <i>size</i> that are supported by the target
1301 will work. 32 (float) and 64 (double) are supported on all targets;
1302 80 or 128 (different flavors of long double) are also supported on some
1303 targets.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1304
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1305 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1306 <dd>This specifies the alignment for an aggregate type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1307 <i>size</i>.</dd>
1308
Daniel Dunbar87bde0b2009-06-08 22:17:53 +0000[diff] [blame]1309 <dt><tt>s<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1310 <dd>This specifies the alignment for a stack object of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1311 <i>size</i>.</dd>
Chris Lattnere82bdc42009-11-07 09:35:34 +0000[diff] [blame]1312
1313 <dt><tt>n<i>size1</i>:<i>size2</i>:<i>size3</i>...</tt></dt>
1314 <dd>This specifies a set of native integer widths for the target CPU
1315 in bits. For example, it might contain "n32" for 32-bit PowerPC,
1316 "n32:64" for PowerPC 64, or "n8:16:32:64" for X86-64. Elements of
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1317 this set are considered to support most general arithmetic
Chris Lattnere82bdc42009-11-07 09:35:34 +0000[diff] [blame]1318 operations efficiently.</dd>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1319</dl>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1320
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1321<p>When constructing the data layout for a given target, LLVM starts with a
Dan Gohman1c70c002010-04-28 00:36:01 +0000[diff] [blame]1322 default set of specifications which are then (possibly) overridden by the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1323 specifications in the <tt>datalayout</tt> keyword. The default specifications
1324 are given in this list:</p>
1325
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1326<ul>
1327 <li><tt>E</tt> - big endian</li>
Dan Gohmanfdf2e8c2010-02-23 02:44:03 +0000[diff] [blame]1328 <li><tt>p:64:64:64</tt> - 64-bit pointers with 64-bit alignment</li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1329 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
1330 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
1331 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
1332 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1333 <li><tt>i64:32:64</tt> - i64 has ABI alignment of 32-bits but preferred
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1334 alignment of 64-bits</li>
1335 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
1336 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
1337 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
1338 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
1339 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
Daniel Dunbar87bde0b2009-06-08 22:17:53 +0000[diff] [blame]1340 <li><tt>s0:64:64</tt> - stack objects are 64-bit aligned</li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1341</ul>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1342
1343<p>When LLVM is determining the alignment for a given type, it uses the
1344 following rules:</p>
1345
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1346<ol>
1347 <li>If the type sought is an exact match for one of the specifications, that
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1348 specification is used.</li>
1349
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1350 <li>If no match is found, and the type sought is an integer type, then the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1351 smallest integer type that is larger than the bitwidth of the sought type
1352 is used. If none of the specifications are larger than the bitwidth then
1353 the the largest integer type is used. For example, given the default
1354 specifications above, the i7 type will use the alignment of i8 (next
1355 largest) while both i65 and i256 will use the alignment of i64 (largest
1356 specified).</li>
1357
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1358 <li>If no match is found, and the type sought is a vector type, then the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1359 largest vector type that is smaller than the sought vector type will be
1360 used as a fall back. This happens because &lt;128 x double&gt; can be
1361 implemented in terms of 64 &lt;2 x double&gt;, for example.</li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1362</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1363
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1364</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1365
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1366<!-- ======================================================================= -->
1367<div class="doc_subsection">
1368 <a name="pointeraliasing">Pointer Aliasing Rules</a>
1369</div>
1370
1371<div class="doc_text">
1372
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1373<p>Any memory access must be done through a pointer value associated
Andreas Bolka99a82052009-07-27 20:37:10 +0000[diff] [blame]1374with an address range of the memory access, otherwise the behavior
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1375is undefined. Pointer values are associated with address ranges
1376according to the following rules:</p>
1377
1378<ul>
Dan Gohman1e109622010-07-02 18:41:32 +0000[diff] [blame]1379 <li>A pointer value is associated with the addresses associated with
1380 any value it is <i>based</i> on.
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1381 <li>An address of a global variable is associated with the address
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1382 range of the variable's storage.</li>
1383 <li>The result value of an allocation instruction is associated with
1384 the address range of the allocated storage.</li>
1385 <li>A null pointer in the default address-space is associated with
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1386 no address.</li>
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1387 <li>An integer constant other than zero or a pointer value returned
1388 from a function not defined within LLVM may be associated with address
1389 ranges allocated through mechanisms other than those provided by
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1390 LLVM. Such ranges shall not overlap with any ranges of addresses
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1391 allocated by mechanisms provided by LLVM.</li>
Dan Gohman1e109622010-07-02 18:41:32 +0000[diff] [blame]1392</ul>
1393
1394<p>A pointer value is <i>based</i> on another pointer value according
1395 to the following rules:</p>
1396
1397<ul>
1398 <li>A pointer value formed from a
1399 <tt><a href="#i_getelementptr">getelementptr</a></tt> operation
1400 is <i>based</i> on the first operand of the <tt>getelementptr</tt>.</li>
1401 <li>The result value of a
1402 <tt><a href="#i_bitcast">bitcast</a></tt> is <i>based</i> on the operand
1403 of the <tt>bitcast</tt>.</li>
1404 <li>A pointer value formed by an
1405 <tt><a href="#i_inttoptr">inttoptr</a></tt> is <i>based</i> on all
1406 pointer values that contribute (directly or indirectly) to the
1407 computation of the pointer's value.</li>
1408 <li>The "<i>based</i> on" relationship is transitive.</li>
1409</ul>
1410
1411<p>Note that this definition of <i>"based"</i> is intentionally
1412 similar to the definition of <i>"based"</i> in C99, though it is
1413 slightly weaker.</p>
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1414
1415<p>LLVM IR does not associate types with memory. The result type of a
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1416<tt><a href="#i_load">load</a></tt> merely indicates the size and
1417alignment of the memory from which to load, as well as the
Dan Gohmanc22c0f32010-06-17 19:23:50 +0000[diff] [blame]1418interpretation of the value. The first operand type of a
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1419<tt><a href="#i_store">store</a></tt> similarly only indicates the size
1420and alignment of the store.</p>
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1421
1422<p>Consequently, type-based alias analysis, aka TBAA, aka
1423<tt>-fstrict-aliasing</tt>, is not applicable to general unadorned
1424LLVM IR. <a href="#metadata">Metadata</a> may be used to encode
1425additional information which specialized optimization passes may use
1426to implement type-based alias analysis.</p>
1427
1428</div>
1429
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]1430<!-- ======================================================================= -->
1431<div class="doc_subsection">
1432 <a name="volatile">Volatile Memory Accesses</a>
1433</div>
1434
1435<div class="doc_text">
1436
1437<p>Certain memory accesses, such as <a href="#i_load"><tt>load</tt></a>s, <a
1438href="#i_store"><tt>store</tt></a>s, and <a
1439href="#int_memcpy"><tt>llvm.memcpy</tt></a>s may be marked <tt>volatile</tt>.
1440The optimizers must not change the number of volatile operations or change their
1441order of execution relative to other volatile operations. The optimizers
1442<i>may</i> change the order of volatile operations relative to non-volatile
1443operations. This is not Java's "volatile" and has no cross-thread
1444synchronization behavior.</p>
1445
1446</div>
1447
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1448<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1449<div class="doc_section"> <a name="typesystem">Type System</a> </div>
1450<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1451
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1452<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1453
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1454<p>The LLVM type system is one of the most important features of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1455 intermediate representation. Being typed enables a number of optimizations
1456 to be performed on the intermediate representation directly, without having
1457 to do extra analyses on the side before the transformation. A strong type
1458 system makes it easier to read the generated code and enables novel analyses
1459 and transformations that are not feasible to perform on normal three address
1460 code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1461
1462</div>
1463
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1464<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1465<div class="doc_subsection"> <a name="t_classifications">Type
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1466Classifications</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1467
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1468<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1469
1470<p>The types fall into a few useful classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1471
1472<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1473 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1474 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1475 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1476 <td><a href="#t_integer">integer</a></td>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1477 <td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1478 </tr>
1479 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1480 <td><a href="#t_floating">floating point</a></td>
1481 <td><tt>float, double, x86_fp80, fp128, ppc_fp128</tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1482 </tr>
1483 <tr>
1484 <td><a name="t_firstclass">first class</a></td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1485 <td><a href="#t_integer">integer</a>,
1486 <a href="#t_floating">floating point</a>,
1487 <a href="#t_pointer">pointer</a>,
Dan Gohman0066db62008-06-18 18:42:13 +0000[diff] [blame]1488 <a href="#t_vector">vector</a>,
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]1489 <a href="#t_struct">structure</a>,
1490 <a href="#t_array">array</a>,
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1491 <a href="#t_label">label</a>,
1492 <a href="#t_metadata">metadata</a>.
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1493 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1494 </tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1495 <tr>
1496 <td><a href="#t_primitive">primitive</a></td>
1497 <td><a href="#t_label">label</a>,
1498 <a href="#t_void">void</a>,
Tobias Grosser05387292010-12-28 20:29:31 +0000[diff] [blame^]1499 <a href="#t_integer">integer</a>,
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1500 <a href="#t_floating">floating point</a>,
Dale Johannesen21fe99b2010-10-01 00:48:59 +0000[diff] [blame]1501 <a href="#t_x86mmx">x86mmx</a>,
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1502 <a href="#t_metadata">metadata</a>.</td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1503 </tr>
1504 <tr>
1505 <td><a href="#t_derived">derived</a></td>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]1506 <td><a href="#t_array">array</a>,
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1507 <a href="#t_function">function</a>,
1508 <a href="#t_pointer">pointer</a>,
1509 <a href="#t_struct">structure</a>,
1510 <a href="#t_pstruct">packed structure</a>,
1511 <a href="#t_vector">vector</a>,
1512 <a href="#t_opaque">opaque</a>.
Dan Gohman01ac1012008-10-14 16:32:04 +0000[diff] [blame]1513 </td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1514 </tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1515 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1516</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1517
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1518<p>The <a href="#t_firstclass">first class</a> types are perhaps the most
1519 important. Values of these types are the only ones which can be produced by
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1520 instructions.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1521
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1522</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1523
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1524<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1525<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1526
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1527<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1528
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1529<p>The primitive types are the fundamental building blocks of the LLVM
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1530 system.</p>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1531
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1532</div>
1533
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1534<!-- _______________________________________________________________________ -->
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1535<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
1536
1537<div class="doc_text">
1538
1539<h5>Overview:</h5>
1540<p>The integer type is a very simple type that simply specifies an arbitrary
1541 bit width for the integer type desired. Any bit width from 1 bit to
1542 2<sup>23</sup>-1 (about 8 million) can be specified.</p>
1543
1544<h5>Syntax:</h5>
1545<pre>
1546 iN
1547</pre>
1548
1549<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
1550 value.</p>
1551
1552<h5>Examples:</h5>
1553<table class="layout">
1554 <tr class="layout">
1555 <td class="left"><tt>i1</tt></td>
1556 <td class="left">a single-bit integer.</td>
1557 </tr>
1558 <tr class="layout">
1559 <td class="left"><tt>i32</tt></td>
1560 <td class="left">a 32-bit integer.</td>
1561 </tr>
1562 <tr class="layout">
1563 <td class="left"><tt>i1942652</tt></td>
1564 <td class="left">a really big integer of over 1 million bits.</td>
1565 </tr>
1566</table>
1567
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1568</div>
1569
1570<!-- _______________________________________________________________________ -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1571<div class="doc_subsubsection"> <a name="t_floating">Floating Point Types</a> </div>
1572
1573<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1574
1575<table>
1576 <tbody>
1577 <tr><th>Type</th><th>Description</th></tr>
1578 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
1579 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
1580 <tr><td><tt>fp128</tt></td><td>128-bit floating point value (112-bit mantissa)</td></tr>
1581 <tr><td><tt>x86_fp80</tt></td><td>80-bit floating point value (X87)</td></tr>
1582 <tr><td><tt>ppc_fp128</tt></td><td>128-bit floating point value (two 64-bits)</td></tr>
1583 </tbody>
1584</table>
1585
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1586</div>
1587
1588<!-- _______________________________________________________________________ -->
Dale Johannesen21fe99b2010-10-01 00:48:59 +0000[diff] [blame]1589<div class="doc_subsubsection"> <a name="t_x86mmx">X86mmx Type</a> </div>
1590
1591<div class="doc_text">
1592
1593<h5>Overview:</h5>
1594<p>The x86mmx type represents a value held in an MMX register on an x86 machine. The operations allowed on it are quite limited: parameters and return values, load and store, and bitcast. User-specified MMX instructions are represented as intrinsic or asm calls with arguments and/or results of this type. There are no arrays, vectors or constants of this type.</p>
1595
1596<h5>Syntax:</h5>
1597<pre>
Dale Johannesen473a8c82010-10-01 01:07:02 +0000[diff] [blame]1598 x86mmx
Dale Johannesen21fe99b2010-10-01 00:48:59 +0000[diff] [blame]1599</pre>
1600
1601</div>
1602
1603<!-- _______________________________________________________________________ -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1604<div class="doc_subsubsection"> <a name="t_void">Void Type</a> </div>
1605
1606<div class="doc_text">
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1607
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1608<h5>Overview:</h5>
1609<p>The void type does not represent any value and has no size.</p>
1610
1611<h5>Syntax:</h5>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1612<pre>
1613 void
1614</pre>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1615
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1616</div>
1617
1618<!-- _______________________________________________________________________ -->
1619<div class="doc_subsubsection"> <a name="t_label">Label Type</a> </div>
1620
1621<div class="doc_text">
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1622
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1623<h5>Overview:</h5>
1624<p>The label type represents code labels.</p>
1625
1626<h5>Syntax:</h5>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1627<pre>
1628 label
1629</pre>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1630
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1631</div>
1632
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1633<!-- _______________________________________________________________________ -->
1634<div class="doc_subsubsection"> <a name="t_metadata">Metadata Type</a> </div>
1635
1636<div class="doc_text">
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1637
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1638<h5>Overview:</h5>
Nick Lewyckyc261df92009-09-27 23:27:42 +0000[diff] [blame]1639<p>The metadata type represents embedded metadata. No derived types may be
1640 created from metadata except for <a href="#t_function">function</a>
1641 arguments.
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1642
1643<h5>Syntax:</h5>
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1644<pre>
1645 metadata
1646</pre>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1647
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1648</div>
1649
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1650
1651<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1652<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1653
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1654<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1655
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1656<p>The real power in LLVM comes from the derived types in the system. This is
1657 what allows a programmer to represent arrays, functions, pointers, and other
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1658 useful types. Each of these types contain one or more element types which
1659 may be a primitive type, or another derived type. For example, it is
1660 possible to have a two dimensional array, using an array as the element type
1661 of another array.</p>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1662
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]1663
1664</div>
1665
1666<!-- _______________________________________________________________________ -->
1667<div class="doc_subsubsection"> <a name="t_aggregate">Aggregate Types</a> </div>
1668
1669<div class="doc_text">
1670
1671<p>Aggregate Types are a subset of derived types that can contain multiple
1672 member types. <a href="#t_array">Arrays</a>,
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]1673 <a href="#t_struct">structs</a>, and <a href="#t_vector">vectors</a> are
1674 aggregate types.</p>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]1675
1676</div>
1677
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1678<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1679<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1680
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1681<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1682
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1683<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1684<p>The array type is a very simple derived type that arranges elements
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1685 sequentially in memory. The array type requires a size (number of elements)
1686 and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1687
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1688<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1689<pre>
1690 [&lt;# elements&gt; x &lt;elementtype&gt;]
1691</pre>
1692
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1693<p>The number of elements is a constant integer value; <tt>elementtype</tt> may
1694 be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1695
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1696<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1697<table class="layout">
1698 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1699 <td class="left"><tt>[40 x i32]</tt></td>
1700 <td class="left">Array of 40 32-bit integer values.</td>
1701 </tr>
1702 <tr class="layout">
1703 <td class="left"><tt>[41 x i32]</tt></td>
1704 <td class="left">Array of 41 32-bit integer values.</td>
1705 </tr>
1706 <tr class="layout">
1707 <td class="left"><tt>[4 x i8]</tt></td>
1708 <td class="left">Array of 4 8-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1709 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1710</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1711<p>Here are some examples of multidimensional arrays:</p>
1712<table class="layout">
1713 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1714 <td class="left"><tt>[3 x [4 x i32]]</tt></td>
1715 <td class="left">3x4 array of 32-bit integer values.</td>
1716 </tr>
1717 <tr class="layout">
1718 <td class="left"><tt>[12 x [10 x float]]</tt></td>
1719 <td class="left">12x10 array of single precision floating point values.</td>
1720 </tr>
1721 <tr class="layout">
1722 <td class="left"><tt>[2 x [3 x [4 x i16]]]</tt></td>
1723 <td class="left">2x3x4 array of 16-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1724 </tr>
1725</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1726
Dan Gohman7657f6b2009-11-09 19:01:53 +0000[diff] [blame]1727<p>There is no restriction on indexing beyond the end of the array implied by
1728 a static type (though there are restrictions on indexing beyond the bounds
1729 of an allocated object in some cases). This means that single-dimension
1730 'variable sized array' addressing can be implemented in LLVM with a zero
1731 length array type. An implementation of 'pascal style arrays' in LLVM could
1732 use the type "<tt>{ i32, [0 x float]}</tt>", for example.</p>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1733
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1734</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1735
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1736<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1737<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1738
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1739<div class="doc_text">
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1740
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1741<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1742<p>The function type can be thought of as a function signature. It consists of
1743 a return type and a list of formal parameter types. The return type of a
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]1744 function type is a first class type or a void type.</p>
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]1745
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1746<h5>Syntax:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1747<pre>
Nick Lewycky51386942009-09-27 07:55:32 +0000[diff] [blame]1748 &lt;returntype&gt; (&lt;parameter list&gt;)
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1749</pre>
1750
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1751<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1752 specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
1753 which indicates that the function takes a variable number of arguments.
1754 Variable argument functions can access their arguments with
1755 the <a href="#int_varargs">variable argument handling intrinsic</a>
Chris Lattner0724fbd2010-03-02 06:36:51 +0000[diff] [blame]1756 functions. '<tt>&lt;returntype&gt;</tt>' is any type except
Nick Lewyckyc261df92009-09-27 23:27:42 +0000[diff] [blame]1757 <a href="#t_label">label</a>.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1758
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1759<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1760<table class="layout">
1761 <tr class="layout">
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1762 <td class="left"><tt>i32 (i32)</tt></td>
1763 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1764 </td>
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1765 </tr><tr class="layout">
Chris Lattner0724fbd2010-03-02 06:36:51 +0000[diff] [blame]1766 <td class="left"><tt>float&nbsp;(i16,&nbsp;i32&nbsp;*)&nbsp;*
Reid Spencerf17a0b72006-12-31 07:20:23 +0000[diff] [blame]1767 </tt></td>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1768 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
Chris Lattner0724fbd2010-03-02 06:36:51 +0000[diff] [blame]1769 an <tt>i16</tt> and a <a href="#t_pointer">pointer</a> to <tt>i32</tt>,
1770 returning <tt>float</tt>.
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1771 </td>
1772 </tr><tr class="layout">
1773 <td class="left"><tt>i32 (i8*, ...)</tt></td>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1774 <td class="left">A vararg function that takes at least one
1775 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
1776 which returns an integer. This is the signature for <tt>printf</tt> in
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1777 LLVM.
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1778 </td>
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1779 </tr><tr class="layout">
1780 <td class="left"><tt>{i32, i32} (i32)</tt></td>
Nick Lewycky51386942009-09-27 07:55:32 +0000[diff] [blame]1781 <td class="left">A function taking an <tt>i32</tt>, returning a
1782 <a href="#t_struct">structure</a> containing two <tt>i32</tt> values
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1783 </td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1784 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1785</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1786
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1787</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1788
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1789<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1790<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1791
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1792<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1793
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1794<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1795<p>The structure type is used to represent a collection of data members together
1796 in memory. The packing of the field types is defined to match the ABI of the
1797 underlying processor. The elements of a structure may be any type that has a
1798 size.</p>
1799
Jeffrey Yasskin7a088cf2010-01-11 19:19:26 +0000[diff] [blame]1800<p>Structures in memory are accessed using '<tt><a href="#i_load">load</a></tt>'
1801 and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a field
1802 with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.
1803 Structures in registers are accessed using the
1804 '<tt><a href="#i_extractvalue">extractvalue</a></tt>' and
1805 '<tt><a href="#i_insertvalue">insertvalue</a></tt>' instructions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1806<h5>Syntax:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1807<pre>
1808 { &lt;type list&gt; }
1809</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1810
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1811<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1812<table class="layout">
1813 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1814 <td class="left"><tt>{ i32, i32, i32 }</tt></td>
1815 <td class="left">A triple of three <tt>i32</tt> values</td>
1816 </tr><tr class="layout">
1817 <td class="left"><tt>{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}</tt></td>
1818 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1819 second element is a <a href="#t_pointer">pointer</a> to a
1820 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1821 an <tt>i32</tt>.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1822 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1823</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1824
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1825</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1826
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1827<!-- _______________________________________________________________________ -->
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1828<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1829</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1830
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1831<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1832
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1833<h5>Overview:</h5>
1834<p>The packed structure type is used to represent a collection of data members
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1835 together in memory. There is no padding between fields. Further, the
1836 alignment of a packed structure is 1 byte. The elements of a packed
1837 structure may be any type that has a size.</p>
1838
1839<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt> and
1840 '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a field with
1841 the '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.</p>
1842
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1843<h5>Syntax:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1844<pre>
1845 &lt; { &lt;type list&gt; } &gt;
1846</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1847
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1848<h5>Examples:</h5>
1849<table class="layout">
1850 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1851 <td class="left"><tt>&lt; { i32, i32, i32 } &gt;</tt></td>
1852 <td class="left">A triple of three <tt>i32</tt> values</td>
1853 </tr><tr class="layout">
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1854 <td class="left">
1855<tt>&lt;&nbsp;{&nbsp;float,&nbsp;i32&nbsp;(i32)*&nbsp;}&nbsp;&gt;</tt></td>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1856 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1857 second element is a <a href="#t_pointer">pointer</a> to a
1858 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1859 an <tt>i32</tt>.</td>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1860 </tr>
1861</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1862
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1863</div>
1864
1865<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1866<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Chris Lattner0fd4a272009-02-08 19:53:29 +0000[diff] [blame]1867
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1868<div class="doc_text">
1869
1870<h5>Overview:</h5>
Dan Gohmanff3ef322010-02-25 16:50:07 +0000[diff] [blame]1871<p>The pointer type is used to specify memory locations.
1872 Pointers are commonly used to reference objects in memory.</p>
1873
1874<p>Pointer types may have an optional address space attribute defining the
1875 numbered address space where the pointed-to object resides. The default
1876 address space is number zero. The semantics of non-zero address
1877 spaces are target-specific.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1878
1879<p>Note that LLVM does not permit pointers to void (<tt>void*</tt>) nor does it
1880 permit pointers to labels (<tt>label*</tt>). Use <tt>i8*</tt> instead.</p>
Chris Lattner0fd4a272009-02-08 19:53:29 +0000[diff] [blame]1881
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1882<h5>Syntax:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1883<pre>
1884 &lt;type&gt; *
1885</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1886
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1887<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1888<table class="layout">
1889 <tr class="layout">
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]1890 <td class="left"><tt>[4 x i32]*</tt></td>
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1891 <td class="left">A <a href="#t_pointer">pointer</a> to <a
1892 href="#t_array">array</a> of four <tt>i32</tt> values.</td>
1893 </tr>
1894 <tr class="layout">
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]1895 <td class="left"><tt>i32 (i32*) *</tt></td>
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1896 <td class="left"> A <a href="#t_pointer">pointer</a> to a <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1897 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1898 <tt>i32</tt>.</td>
1899 </tr>
1900 <tr class="layout">
1901 <td class="left"><tt>i32 addrspace(5)*</tt></td>
1902 <td class="left">A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value
1903 that resides in address space #5.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1904 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1905</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1906
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1907</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1908
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1909<!-- _______________________________________________________________________ -->
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1910<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1911
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1912<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1913
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1914<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1915<p>A vector type is a simple derived type that represents a vector of elements.
1916 Vector types are used when multiple primitive data are operated in parallel
1917 using a single instruction (SIMD). A vector type requires a size (number of
Duncan Sandsd40d14e2009-11-27 13:38:03 +0000[diff] [blame]1918 elements) and an underlying primitive data type. Vector types are considered
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1919 <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1920
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1921<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1922<pre>
1923 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1924</pre>
1925
Chris Lattner7d2e7be2010-10-10 18:20:35 +0000[diff] [blame]1926<p>The number of elements is a constant integer value larger than 0; elementtype
1927 may be any integer or floating point type. Vectors of size zero are not
1928 allowed, and pointers are not allowed as the element type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1929
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1930<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1931<table class="layout">
1932 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1933 <td class="left"><tt>&lt;4 x i32&gt;</tt></td>
1934 <td class="left">Vector of 4 32-bit integer values.</td>
1935 </tr>
1936 <tr class="layout">
1937 <td class="left"><tt>&lt;8 x float&gt;</tt></td>
1938 <td class="left">Vector of 8 32-bit floating-point values.</td>
1939 </tr>
1940 <tr class="layout">
1941 <td class="left"><tt>&lt;2 x i64&gt;</tt></td>
1942 <td class="left">Vector of 2 64-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1943 </tr>
1944</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1945
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1946</div>
1947
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1948<!-- _______________________________________________________________________ -->
1949<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1950<div class="doc_text">
1951
1952<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1953<p>Opaque types are used to represent unknown types in the system. This
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1954 corresponds (for example) to the C notion of a forward declared structure
1955 type. In LLVM, opaque types can eventually be resolved to any type (not just
1956 a structure type).</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1957
1958<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1959<pre>
1960 opaque
1961</pre>
1962
1963<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1964<table class="layout">
1965 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1966 <td class="left"><tt>opaque</tt></td>
1967 <td class="left">An opaque type.</td>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1968 </tr>
1969</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1970
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1971</div>
1972
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1973<!-- ======================================================================= -->
1974<div class="doc_subsection">
1975 <a name="t_uprefs">Type Up-references</a>
1976</div>
1977
1978<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1979
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1980<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1981<p>An "up reference" allows you to refer to a lexically enclosing type without
1982 requiring it to have a name. For instance, a structure declaration may
1983 contain a pointer to any of the types it is lexically a member of. Example
1984 of up references (with their equivalent as named type declarations)
1985 include:</p>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1986
1987<pre>
Chris Lattner3060f5b2009-02-09 10:00:56 +0000[diff] [blame]1988 { \2 * } %x = type { %x* }
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1989 { \2 }* %y = type { %y }*
1990 \1* %z = type %z*
1991</pre>
1992
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1993<p>An up reference is needed by the asmprinter for printing out cyclic types
1994 when there is no declared name for a type in the cycle. Because the
1995 asmprinter does not want to print out an infinite type string, it needs a
1996 syntax to handle recursive types that have no names (all names are optional
1997 in llvm IR).</p>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1998
1999<h5>Syntax:</h5>
2000<pre>
2001 \&lt;level&gt;
2002</pre>
2003
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2004<p>The level is the count of the lexical type that is being referred to.</p>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]2005
2006<h5>Examples:</h5>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]2007<table class="layout">
2008 <tr class="layout">
2009 <td class="left"><tt>\1*</tt></td>
2010 <td class="left">Self-referential pointer.</td>
2011 </tr>
2012 <tr class="layout">
2013 <td class="left"><tt>{ { \3*, i8 }, i32 }</tt></td>
2014 <td class="left">Recursive structure where the upref refers to the out-most
2015 structure.</td>
2016 </tr>
2017</table>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]2018
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2019</div>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]2020
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2021<!-- *********************************************************************** -->
2022<div class="doc_section"> <a name="constants">Constants</a> </div>
2023<!-- *********************************************************************** -->
2024
2025<div class="doc_text">
2026
2027<p>LLVM has several different basic types of constants. This section describes
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2028 them all and their syntax.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2029
2030</div>
2031
2032<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]2033<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2034
2035<div class="doc_text">
2036
2037<dl>
2038 <dt><b>Boolean constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2039 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]2040 constants of the <tt><a href="#t_integer">i1</a></tt> type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2041
2042 <dt><b>Integer constants</b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2043 <dd>Standard integers (such as '4') are constants of
2044 the <a href="#t_integer">integer</a> type. Negative numbers may be used
2045 with integer types.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2046
2047 <dt><b>Floating point constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2048 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2049 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
2050 notation (see below). The assembler requires the exact decimal value of a
2051 floating-point constant. For example, the assembler accepts 1.25 but
2052 rejects 1.3 because 1.3 is a repeating decimal in binary. Floating point
2053 constants must have a <a href="#t_floating">floating point</a> type. </dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2054
2055 <dt><b>Null pointer constants</b></dt>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2056 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2057 and must be of <a href="#t_pointer">pointer type</a>.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2058</dl>
2059
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2060<p>The one non-intuitive notation for constants is the hexadecimal form of
2061 floating point constants. For example, the form '<tt>double
2062 0x432ff973cafa8000</tt>' is equivalent to (but harder to read than)
2063 '<tt>double 4.5e+15</tt>'. The only time hexadecimal floating point
2064 constants are required (and the only time that they are generated by the
2065 disassembler) is when a floating point constant must be emitted but it cannot
2066 be represented as a decimal floating point number in a reasonable number of
2067 digits. For example, NaN's, infinities, and other special values are
2068 represented in their IEEE hexadecimal format so that assembly and disassembly
2069 do not cause any bits to change in the constants.</p>
2070
Dale Johannesenbd5e5a82009-02-11 22:14:51 +0000[diff] [blame]2071<p>When using the hexadecimal form, constants of types float and double are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2072 represented using the 16-digit form shown above (which matches the IEEE754
2073 representation for double); float values must, however, be exactly
2074 representable as IEE754 single precision. Hexadecimal format is always used
2075 for long double, and there are three forms of long double. The 80-bit format
2076 used by x86 is represented as <tt>0xK</tt> followed by 20 hexadecimal digits.
2077 The 128-bit format used by PowerPC (two adjacent doubles) is represented
2078 by <tt>0xM</tt> followed by 32 hexadecimal digits. The IEEE 128-bit format
2079 is represented by <tt>0xL</tt> followed by 32 hexadecimal digits; no
2080 currently supported target uses this format. Long doubles will only work if
2081 they match the long double format on your target. All hexadecimal formats
2082 are big-endian (sign bit at the left).</p>
2083
Dale Johannesen21fe99b2010-10-01 00:48:59 +0000[diff] [blame]2084<p>There are no constants of type x86mmx.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2085</div>
2086
2087<!-- ======================================================================= -->
Chris Lattner70882792009-02-28 18:32:25 +0000[diff] [blame]2088<div class="doc_subsection">
Bill Wendlingd9fe2982009-07-20 02:32:41 +0000[diff] [blame]2089<a name="aggregateconstants"></a> <!-- old anchor -->
2090<a name="complexconstants">Complex Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2091</div>
2092
2093<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2094
Chris Lattner70882792009-02-28 18:32:25 +0000[diff] [blame]2095<p>Complex constants are a (potentially recursive) combination of simple
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2096 constants and smaller complex constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2097
2098<dl>
2099 <dt><b>Structure constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2100 <dd>Structure constants are represented with notation similar to structure
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2101 type definitions (a comma separated list of elements, surrounded by braces
2102 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* @G }</tt>",
2103 where "<tt>@G</tt>" is declared as "<tt>@G = external global i32</tt>".
2104 Structure constants must have <a href="#t_struct">structure type</a>, and
2105 the number and types of elements must match those specified by the
2106 type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2107
2108 <dt><b>Array constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2109 <dd>Array constants are represented with notation similar to array type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2110 definitions (a comma separated list of elements, surrounded by square
2111 brackets (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74
2112 ]</tt>". Array constants must have <a href="#t_array">array type</a>, and
2113 the number and types of elements must match those specified by the
2114 type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2115
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2116 <dt><b>Vector constants</b></dt>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2117 <dd>Vector constants are represented with notation similar to vector type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2118 definitions (a comma separated list of elements, surrounded by
2119 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32
2120 42, i32 11, i32 74, i32 100 &gt;</tt>". Vector constants must
2121 have <a href="#t_vector">vector type</a>, and the number and types of
2122 elements must match those specified by the type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2123
2124 <dt><b>Zero initialization</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2125 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]2126 value to zero of <em>any</em> type, including scalar and
2127 <a href="#t_aggregate">aggregate</a> types.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2128 This is often used to avoid having to print large zero initializers
2129 (e.g. for large arrays) and is always exactly equivalent to using explicit
2130 zero initializers.</dd>
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2131
2132 <dt><b>Metadata node</b></dt>
Nick Lewycky1e8c7a62009-05-30 16:08:30 +0000[diff] [blame]2133 <dd>A metadata node is a structure-like constant with
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2134 <a href="#t_metadata">metadata type</a>. For example: "<tt>metadata !{
2135 i32 0, metadata !"test" }</tt>". Unlike other constants that are meant to
2136 be interpreted as part of the instruction stream, metadata is a place to
2137 attach additional information such as debug info.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2138</dl>
2139
2140</div>
2141
2142<!-- ======================================================================= -->
2143<div class="doc_subsection">
2144 <a name="globalconstants">Global Variable and Function Addresses</a>
2145</div>
2146
2147<div class="doc_text">
2148
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2149<p>The addresses of <a href="#globalvars">global variables</a>
2150 and <a href="#functionstructure">functions</a> are always implicitly valid
2151 (link-time) constants. These constants are explicitly referenced when
2152 the <a href="#identifiers">identifier for the global</a> is used and always
2153 have <a href="#t_pointer">pointer</a> type. For example, the following is a
2154 legal LLVM file:</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2155
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2156<pre class="doc_code">
Chris Lattnera18a4242007-06-06 18:28:13 +0000[diff] [blame]2157@X = global i32 17
2158@Y = global i32 42
2159@Z = global [2 x i32*] [ i32* @X, i32* @Y ]
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2160</pre>
2161
2162</div>
2163
2164<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]2165<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2166<div class="doc_text">
2167
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2168<p>The string '<tt>undef</tt>' can be used anywhere a constant is expected, and
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]2169 indicates that the user of the value may receive an unspecified bit-pattern.
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2170 Undefined values may be of any type (other than '<tt>label</tt>'
2171 or '<tt>void</tt>') and be used anywhere a constant is permitted.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2172
Chris Lattnerc608cb12009-09-11 01:49:31 +0000[diff] [blame]2173<p>Undefined values are useful because they indicate to the compiler that the
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2174 program is well defined no matter what value is used. This gives the
2175 compiler more freedom to optimize. Here are some examples of (potentially
2176 surprising) transformations that are valid (in pseudo IR):</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2177
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2178
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2179<pre class="doc_code">
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2180 %A = add %X, undef
2181 %B = sub %X, undef
2182 %C = xor %X, undef
2183Safe:
2184 %A = undef
2185 %B = undef
2186 %C = undef
2187</pre>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2188
2189<p>This is safe because all of the output bits are affected by the undef bits.
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2190 Any output bit can have a zero or one depending on the input bits.</p>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2191
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2192<pre class="doc_code">
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2193 %A = or %X, undef
2194 %B = and %X, undef
2195Safe:
2196 %A = -1
2197 %B = 0
2198Unsafe:
2199 %A = undef
2200 %B = undef
2201</pre>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2202
2203<p>These logical operations have bits that are not always affected by the input.
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2204 For example, if <tt>%X</tt> has a zero bit, then the output of the
2205 '<tt>and</tt>' operation will always be a zero for that bit, no matter what
2206 the corresponding bit from the '<tt>undef</tt>' is. As such, it is unsafe to
2207 optimize or assume that the result of the '<tt>and</tt>' is '<tt>undef</tt>'.
2208 However, it is safe to assume that all bits of the '<tt>undef</tt>' could be
2209 0, and optimize the '<tt>and</tt>' to 0. Likewise, it is safe to assume that
2210 all the bits of the '<tt>undef</tt>' operand to the '<tt>or</tt>' could be
2211 set, allowing the '<tt>or</tt>' to be folded to -1.</p>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2212
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2213<pre class="doc_code">
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2214 %A = select undef, %X, %Y
2215 %B = select undef, 42, %Y
2216 %C = select %X, %Y, undef
2217Safe:
2218 %A = %X (or %Y)
2219 %B = 42 (or %Y)
2220 %C = %Y
2221Unsafe:
2222 %A = undef
2223 %B = undef
2224 %C = undef
2225</pre>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2226
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2227<p>This set of examples shows that undefined '<tt>select</tt>' (and conditional
2228 branch) conditions can go <em>either way</em>, but they have to come from one
2229 of the two operands. In the <tt>%A</tt> example, if <tt>%X</tt> and
2230 <tt>%Y</tt> were both known to have a clear low bit, then <tt>%A</tt> would
2231 have to have a cleared low bit. However, in the <tt>%C</tt> example, the
2232 optimizer is allowed to assume that the '<tt>undef</tt>' operand could be the
2233 same as <tt>%Y</tt>, allowing the whole '<tt>select</tt>' to be
2234 eliminated.</p>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2235
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2236<pre class="doc_code">
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2237 %A = xor undef, undef
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2238
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2239 %B = undef
2240 %C = xor %B, %B
2241
2242 %D = undef
2243 %E = icmp lt %D, 4
2244 %F = icmp gte %D, 4
2245
2246Safe:
2247 %A = undef
2248 %B = undef
2249 %C = undef
2250 %D = undef
2251 %E = undef
2252 %F = undef
2253</pre>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2254
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2255<p>This example points out that two '<tt>undef</tt>' operands are not
2256 necessarily the same. This can be surprising to people (and also matches C
2257 semantics) where they assume that "<tt>X^X</tt>" is always zero, even
2258 if <tt>X</tt> is undefined. This isn't true for a number of reasons, but the
2259 short answer is that an '<tt>undef</tt>' "variable" can arbitrarily change
2260 its value over its "live range". This is true because the variable doesn't
2261 actually <em>have a live range</em>. Instead, the value is logically read
2262 from arbitrary registers that happen to be around when needed, so the value
2263 is not necessarily consistent over time. In fact, <tt>%A</tt> and <tt>%C</tt>
2264 need to have the same semantics or the core LLVM "replace all uses with"
2265 concept would not hold.</p>
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2266
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2267<pre class="doc_code">
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2268 %A = fdiv undef, %X
2269 %B = fdiv %X, undef
2270Safe:
2271 %A = undef
2272b: unreachable
2273</pre>
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2274
2275<p>These examples show the crucial difference between an <em>undefined
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2276 value</em> and <em>undefined behavior</em>. An undefined value (like
2277 '<tt>undef</tt>') is allowed to have an arbitrary bit-pattern. This means that
2278 the <tt>%A</tt> operation can be constant folded to '<tt>undef</tt>', because
2279 the '<tt>undef</tt>' could be an SNaN, and <tt>fdiv</tt> is not (currently)
2280 defined on SNaN's. However, in the second example, we can make a more
2281 aggressive assumption: because the <tt>undef</tt> is allowed to be an
2282 arbitrary value, we are allowed to assume that it could be zero. Since a
2283 divide by zero has <em>undefined behavior</em>, we are allowed to assume that
2284 the operation does not execute at all. This allows us to delete the divide and
2285 all code after it. Because the undefined operation "can't happen", the
2286 optimizer can assume that it occurs in dead code.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2287
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2288<pre class="doc_code">
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2289a: store undef -> %X
2290b: store %X -> undef
2291Safe:
2292a: &lt;deleted&gt;
2293b: unreachable
2294</pre>
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2295
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2296<p>These examples reiterate the <tt>fdiv</tt> example: a store <em>of</em> an
2297 undefined value can be assumed to not have any effect; we can assume that the
2298 value is overwritten with bits that happen to match what was already there.
2299 However, a store <em>to</em> an undefined location could clobber arbitrary
2300 memory, therefore, it has undefined behavior.</p>
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2301
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2302</div>
2303
2304<!-- ======================================================================= -->
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]2305<div class="doc_subsection"><a name="trapvalues">Trap Values</a></div>
2306<div class="doc_text">
2307
Dan Gohmanc68ce062010-04-26 20:21:21 +0000[diff] [blame]2308<p>Trap values are similar to <a href="#undefvalues">undef values</a>, however
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]2309 instead of representing an unspecified bit pattern, they represent the
2310 fact that an instruction or constant expression which cannot evoke side
2311 effects has nevertheless detected a condition which results in undefined
Dan Gohmanc68ce062010-04-26 20:21:21 +0000[diff] [blame]2312 behavior.</p>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]2313
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2314<p>There is currently no way of representing a trap value in the IR; they
Dan Gohman855abed2010-05-03 14:51:43 +0000[diff] [blame]2315 only exist when produced by operations such as
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2316 <a href="#i_add"><tt>add</tt></a> with the <tt>nsw</tt> flag.</p>
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2317
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2318<p>Trap value behavior is defined in terms of value <i>dependence</i>:</p>
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2319
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2320<ul>
2321<li>Values other than <a href="#i_phi"><tt>phi</tt></a> nodes depend on
2322 their operands.</li>
2323
2324<li><a href="#i_phi"><tt>Phi</tt></a> nodes depend on the operand corresponding
2325 to their dynamic predecessor basic block.</li>
2326
2327<li>Function arguments depend on the corresponding actual argument values in
2328 the dynamic callers of their functions.</li>
2329
2330<li><a href="#i_call"><tt>Call</tt></a> instructions depend on the
2331 <a href="#i_ret"><tt>ret</tt></a> instructions that dynamically transfer
2332 control back to them.</li>
2333
Dan Gohmanb5328162010-05-03 14:55:22 +0000[diff] [blame]2334<li><a href="#i_invoke"><tt>Invoke</tt></a> instructions depend on the
2335 <a href="#i_ret"><tt>ret</tt></a>, <a href="#i_unwind"><tt>unwind</tt></a>,
2336 or exception-throwing call instructions that dynamically transfer control
2337 back to them.</li>
2338
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2339<li>Non-volatile loads and stores depend on the most recent stores to all of the
2340 referenced memory addresses, following the order in the IR
2341 (including loads and stores implied by intrinsics such as
2342 <a href="#int_memcpy"><tt>@llvm.memcpy</tt></a>.)</li>
2343
Dan Gohman7c24ff12010-05-03 14:59:34 +0000[diff] [blame]2344<!-- TODO: In the case of multiple threads, this only applies if the store
2345 "happens-before" the load or store. -->
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2346
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2347<!-- TODO: floating-point exception state -->
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2348
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2349<li>An instruction with externally visible side effects depends on the most
2350 recent preceding instruction with externally visible side effects, following
Dan Gohmanff70fe42010-07-06 15:26:33 +0000[diff] [blame]2351 the order in the IR. (This includes
2352 <a href="#volatile">volatile operations</a>.)</li>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]2353
Dan Gohmanb5328162010-05-03 14:55:22 +0000[diff] [blame]2354<li>An instruction <i>control-depends</i> on a
2355 <a href="#terminators">terminator instruction</a>
2356 if the terminator instruction has multiple successors and the instruction
2357 is always executed when control transfers to one of the successors, and
2358 may not be executed when control is transfered to another.</li>
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2359
2360<li>Dependence is transitive.</li>
2361
2362</ul>
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2363
2364<p>Whenever a trap value is generated, all values which depend on it evaluate
2365 to trap. If they have side effects, the evoke their side effects as if each
2366 operand with a trap value were undef. If they have externally-visible side
2367 effects, the behavior is undefined.</p>
2368
2369<p>Here are some examples:</p>
Dan Gohmanc30f6e12010-04-26 20:54:53 +0000[diff] [blame]2370
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2371<pre class="doc_code">
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2372entry:
2373 %trap = sub nuw i32 0, 1 ; Results in a trap value.
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2374 %still_trap = and i32 %trap, 0 ; Whereas (and i32 undef, 0) would return 0.
2375 %trap_yet_again = getelementptr i32* @h, i32 %still_trap
2376 store i32 0, i32* %trap_yet_again ; undefined behavior
2377
2378 store i32 %trap, i32* @g ; Trap value conceptually stored to memory.
2379 %trap2 = load i32* @g ; Returns a trap value, not just undef.
2380
2381 volatile store i32 %trap, i32* @g ; External observation; undefined behavior.
2382
2383 %narrowaddr = bitcast i32* @g to i16*
2384 %wideaddr = bitcast i32* @g to i64*
2385 %trap3 = load 16* %narrowaddr ; Returns a trap value.
2386 %trap4 = load i64* %widaddr ; Returns a trap value.
2387
2388 %cmp = icmp i32 slt %trap, 0 ; Returns a trap value.
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2389 %br i1 %cmp, %true, %end ; Branch to either destination.
2390
2391true:
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2392 volatile store i32 0, i32* @g ; This is control-dependent on %cmp, so
2393 ; it has undefined behavior.
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2394 br label %end
2395
2396end:
2397 %p = phi i32 [ 0, %entry ], [ 1, %true ]
2398 ; Both edges into this PHI are
2399 ; control-dependent on %cmp, so this
Dan Gohman34b3d992010-04-28 00:49:41 +0000[diff] [blame]2400 ; always results in a trap value.
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2401
2402 volatile store i32 0, i32* @g ; %end is control-equivalent to %entry
2403 ; so this is defined (ignoring earlier
2404 ; undefined behavior in this example).
Dan Gohmanae11c3f2010-04-26 23:36:52 +0000[diff] [blame]2405</pre>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]2406
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]2407</div>
2408
2409<!-- ======================================================================= -->
Chris Lattnerf9d078e2009-10-27 21:19:13 +0000[diff] [blame]2410<div class="doc_subsection"><a name="blockaddress">Addresses of Basic
2411 Blocks</a></div>
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2412<div class="doc_text">
2413
Chris Lattnercdfc9402009-11-01 01:27:45 +0000[diff] [blame]2414<p><b><tt>blockaddress(@function, %block)</tt></b></p>
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2415
2416<p>The '<tt>blockaddress</tt>' constant computes the address of the specified
Chris Lattner2dfdf2a2009-10-27 21:49:40 +0000[diff] [blame]2417 basic block in the specified function, and always has an i8* type. Taking
Chris Lattnercdfc9402009-11-01 01:27:45 +0000[diff] [blame]2418 the address of the entry block is illegal.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2419
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2420<p>This value only has defined behavior when used as an operand to the
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2421 '<a href="#i_indirectbr"><tt>indirectbr</tt></a>' instruction, or for
2422 comparisons against null. Pointer equality tests between labels addresses
2423 results in undefined behavior &mdash; though, again, comparison against null
2424 is ok, and no label is equal to the null pointer. This may be passed around
2425 as an opaque pointer sized value as long as the bits are not inspected. This
2426 allows <tt>ptrtoint</tt> and arithmetic to be performed on these values so
2427 long as the original value is reconstituted before the <tt>indirectbr</tt>
2428 instruction.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2429
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2430<p>Finally, some targets may provide defined semantics when using the value as
2431 the operand to an inline assembly, but that is target specific.</p>
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2432
2433</div>
2434
2435
2436<!-- ======================================================================= -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2437<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
2438</div>
2439
2440<div class="doc_text">
2441
2442<p>Constant expressions are used to allow expressions involving other constants
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2443 to be used as constants. Constant expressions may be of
2444 any <a href="#t_firstclass">first class</a> type and may involve any LLVM
2445 operation that does not have side effects (e.g. load and call are not
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]2446 supported). The following is the syntax for constant expressions:</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2447
2448<dl>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2449 <dt><b><tt>trunc (CST to TYPE)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2450 <dd>Truncate a constant to another type. The bit size of CST must be larger
2451 than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2452
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2453 <dt><b><tt>zext (CST to TYPE)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2454 <dd>Zero extend a constant to another type. The bit size of CST must be
Duncan Sands28afd432010-07-13 12:06:14 +0000[diff] [blame]2455 smaller than the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2456
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2457 <dt><b><tt>sext (CST to TYPE)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2458 <dd>Sign extend a constant to another type. The bit size of CST must be
Duncan Sands28afd432010-07-13 12:06:14 +0000[diff] [blame]2459 smaller than the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2460
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2461 <dt><b><tt>fptrunc (CST to TYPE)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2462 <dd>Truncate a floating point constant to another floating point type. The
2463 size of CST must be larger than the size of TYPE. Both types must be
2464 floating point.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2465
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2466 <dt><b><tt>fpext (CST to TYPE)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2467 <dd>Floating point extend a constant to another type. The size of CST must be
2468 smaller or equal to the size of TYPE. Both types must be floating
2469 point.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2470
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2471 <dt><b><tt>fptoui (CST to TYPE)</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2472 <dd>Convert a floating point constant to the corresponding unsigned integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2473 constant. TYPE must be a scalar or vector integer type. CST must be of
2474 scalar or vector floating point type. Both CST and TYPE must be scalars,
2475 or vectors of the same number of elements. If the value won't fit in the
2476 integer type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2477
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2478 <dt><b><tt>fptosi (CST to TYPE)</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2479 <dd>Convert a floating point constant to the corresponding signed integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2480 constant. TYPE must be a scalar or vector integer type. CST must be of
2481 scalar or vector floating point type. Both CST and TYPE must be scalars,
2482 or vectors of the same number of elements. If the value won't fit in the
2483 integer type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2484
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2485 <dt><b><tt>uitofp (CST to TYPE)</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2486 <dd>Convert an unsigned integer constant to the corresponding floating point
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2487 constant. TYPE must be a scalar or vector floating point type. CST must be
2488 of scalar or vector integer type. Both CST and TYPE must be scalars, or
2489 vectors of the same number of elements. If the value won't fit in the
2490 floating point type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2491
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2492 <dt><b><tt>sitofp (CST to TYPE)</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2493 <dd>Convert a signed integer constant to the corresponding floating point
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2494 constant. TYPE must be a scalar or vector floating point type. CST must be
2495 of scalar or vector integer type. Both CST and TYPE must be scalars, or
2496 vectors of the same number of elements. If the value won't fit in the
2497 floating point type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2498
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2499 <dt><b><tt>ptrtoint (CST to TYPE)</tt></b></dt>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]2500 <dd>Convert a pointer typed constant to the corresponding integer constant
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2501 <tt>TYPE</tt> must be an integer type. <tt>CST</tt> must be of pointer
2502 type. The <tt>CST</tt> value is zero extended, truncated, or unchanged to
2503 make it fit in <tt>TYPE</tt>.</dd>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]2504
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2505 <dt><b><tt>inttoptr (CST to TYPE)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2506 <dd>Convert a integer constant to a pointer constant. TYPE must be a pointer
2507 type. CST must be of integer type. The CST value is zero extended,
2508 truncated, or unchanged to make it fit in a pointer size. This one is
2509 <i>really</i> dangerous!</dd>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]2510
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2511 <dt><b><tt>bitcast (CST to TYPE)</tt></b></dt>
Chris Lattner03bbad62009-02-28 18:27:03 +0000[diff] [blame]2512 <dd>Convert a constant, CST, to another TYPE. The constraints of the operands
2513 are the same as those for the <a href="#i_bitcast">bitcast
2514 instruction</a>.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2515
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2516 <dt><b><tt>getelementptr (CSTPTR, IDX0, IDX1, ...)</tt></b></dt>
2517 <dt><b><tt>getelementptr inbounds (CSTPTR, IDX0, IDX1, ...)</tt></b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2518 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2519 constants. As with the <a href="#i_getelementptr">getelementptr</a>
2520 instruction, the index list may have zero or more indexes, which are
2521 required to make sense for the type of "CSTPTR".</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2522
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2523 <dt><b><tt>select (COND, VAL1, VAL2)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2524 <dd>Perform the <a href="#i_select">select operation</a> on constants.</dd>
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]2525
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2526 <dt><b><tt>icmp COND (VAL1, VAL2)</tt></b></dt>
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]2527 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
2528
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2529 <dt><b><tt>fcmp COND (VAL1, VAL2)</tt></b></dt>
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]2530 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]2531
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2532 <dt><b><tt>extractelement (VAL, IDX)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2533 <dd>Perform the <a href="#i_extractelement">extractelement operation</a> on
2534 constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]2535
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2536 <dt><b><tt>insertelement (VAL, ELT, IDX)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2537 <dd>Perform the <a href="#i_insertelement">insertelement operation</a> on
2538 constants.</dd>
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]2539
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2540 <dt><b><tt>shufflevector (VEC1, VEC2, IDXMASK)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2541 <dd>Perform the <a href="#i_shufflevector">shufflevector operation</a> on
2542 constants.</dd>
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]2543
Nick Lewycky9e130ce2010-05-29 06:44:15 +0000[diff] [blame]2544 <dt><b><tt>extractvalue (VAL, IDX0, IDX1, ...)</tt></b></dt>
2545 <dd>Perform the <a href="#i_extractvalue">extractvalue operation</a> on
2546 constants. The index list is interpreted in a similar manner as indices in
2547 a '<a href="#i_getelementptr">getelementptr</a>' operation. At least one
2548 index value must be specified.</dd>
2549
2550 <dt><b><tt>insertvalue (VAL, ELT, IDX0, IDX1, ...)</tt></b></dt>
2551 <dd>Perform the <a href="#i_insertvalue">insertvalue operation</a> on
2552 constants. The index list is interpreted in a similar manner as indices in
2553 a '<a href="#i_getelementptr">getelementptr</a>' operation. At least one
2554 index value must be specified.</dd>
2555
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]2556 <dt><b><tt>OPCODE (LHS, RHS)</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2557 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
2558 be any of the <a href="#binaryops">binary</a>
2559 or <a href="#bitwiseops">bitwise binary</a> operations. The constraints
2560 on operands are the same as those for the corresponding instruction
2561 (e.g. no bitwise operations on floating point values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2562</dl>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2563
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2564</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]2565
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2566<!-- *********************************************************************** -->
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2567<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
2568<!-- *********************************************************************** -->
2569
2570<!-- ======================================================================= -->
2571<div class="doc_subsection">
2572<a name="inlineasm">Inline Assembler Expressions</a>
2573</div>
2574
2575<div class="doc_text">
2576
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2577<p>LLVM supports inline assembler expressions (as opposed
2578 to <a href="#moduleasm"> Module-Level Inline Assembly</a>) through the use of
2579 a special value. This value represents the inline assembler as a string
2580 (containing the instructions to emit), a list of operand constraints (stored
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2581 as a string), a flag that indicates whether or not the inline asm
Dale Johannesen8ba2d5b2009-10-21 23:28:00 +0000[diff] [blame]2582 expression has side effects, and a flag indicating whether the function
2583 containing the asm needs to align its stack conservatively. An example
2584 inline assembler expression is:</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2585
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2586<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2587i32 (i32) asm "bswap $0", "=r,r"
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2588</pre>
2589
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2590<p>Inline assembler expressions may <b>only</b> be used as the callee operand of
2591 a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we
2592 have:</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2593
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2594<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2595%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]2596</pre>
2597
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2598<p>Inline asms with side effects not visible in the constraint list must be
2599 marked as having side effects. This is done through the use of the
2600 '<tt>sideeffect</tt>' keyword, like so:</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2601
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2602<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2603call void asm sideeffect "eieio", ""()
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2604</pre>
2605
Dale Johannesen8ba2d5b2009-10-21 23:28:00 +0000[diff] [blame]2606<p>In some cases inline asms will contain code that will not work unless the
2607 stack is aligned in some way, such as calls or SSE instructions on x86,
2608 yet will not contain code that does that alignment within the asm.
2609 The compiler should make conservative assumptions about what the asm might
2610 contain and should generate its usual stack alignment code in the prologue
2611 if the '<tt>alignstack</tt>' keyword is present:</p>
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2612
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2613<pre class="doc_code">
Dale Johannesen8ba2d5b2009-10-21 23:28:00 +0000[diff] [blame]2614call void asm alignstack "eieio", ""()
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2615</pre>
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2616
2617<p>If both keywords appear the '<tt>sideeffect</tt>' keyword must come
2618 first.</p>
2619
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2620<p>TODO: The format of the asm and constraints string still need to be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2621 documented here. Constraints on what can be done (e.g. duplication, moving,
2622 etc need to be documented). This is probably best done by reference to
2623 another document that covers inline asm from a holistic perspective.</p>
Chris Lattnercf9a4152010-04-07 05:38:05 +0000[diff] [blame]2624</div>
2625
2626<div class="doc_subsubsection">
2627<a name="inlineasm_md">Inline Asm Metadata</a>
2628</div>
2629
2630<div class="doc_text">
2631
2632<p>The call instructions that wrap inline asm nodes may have a "!srcloc" MDNode
Chris Lattnerce1b9ad2010-11-17 08:20:42 +0000[diff] [blame]2633 attached to it that contains a list of constant integers. If present, the
2634 code generator will use the integer as the location cookie value when report
Chris Lattnercf9a4152010-04-07 05:38:05 +0000[diff] [blame]2635 errors through the LLVMContext error reporting mechanisms. This allows a
Dan Gohman1c70c002010-04-28 00:36:01 +0000[diff] [blame]2636 front-end to correlate backend errors that occur with inline asm back to the
Chris Lattnercf9a4152010-04-07 05:38:05 +0000[diff] [blame]2637 source code that produced it. For example:</p>
2638
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2639<pre class="doc_code">
Chris Lattnercf9a4152010-04-07 05:38:05 +0000[diff] [blame]2640call void asm sideeffect "something bad", ""()<b>, !srcloc !42</b>
2641...
2642!42 = !{ i32 1234567 }
2643</pre>
Chris Lattnercf9a4152010-04-07 05:38:05 +0000[diff] [blame]2644
2645<p>It is up to the front-end to make sense of the magic numbers it places in the
Chris Lattnerce1b9ad2010-11-17 08:20:42 +0000[diff] [blame]2646 IR. If the MDNode contains multiple constants, the code generator will use
2647 the one that corresponds to the line of the asm that the error occurs on.</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2648
2649</div>
2650
Chris Lattnere6a5ddd2010-01-15 21:50:19 +0000[diff] [blame]2651<!-- ======================================================================= -->
2652<div class="doc_subsection"><a name="metadata">Metadata Nodes and Metadata
2653 Strings</a>
2654</div>
2655
2656<div class="doc_text">
2657
2658<p>LLVM IR allows metadata to be attached to instructions in the program that
2659 can convey extra information about the code to the optimizers and code
2660 generator. One example application of metadata is source-level debug
2661 information. There are two metadata primitives: strings and nodes. All
2662 metadata has the <tt>metadata</tt> type and is identified in syntax by a
2663 preceding exclamation point ('<tt>!</tt>').</p>
2664
2665<p>A metadata string is a string surrounded by double quotes. It can contain
2666 any character by escaping non-printable characters with "\xx" where "xx" is
2667 the two digit hex code. For example: "<tt>!"test\00"</tt>".</p>
2668
2669<p>Metadata nodes are represented with notation similar to structure constants
2670 (a comma separated list of elements, surrounded by braces and preceded by an
2671 exclamation point). For example: "<tt>!{ metadata !"test\00", i32
2672 10}</tt>". Metadata nodes can have any values as their operand.</p>
2673
2674<p>A <a href="#namedmetadatastructure">named metadata</a> is a collection of
2675 metadata nodes, which can be looked up in the module symbol table. For
2676 example: "<tt>!foo = metadata !{!4, !3}</tt>".
2677
Devang Patele1d50cd2010-03-04 23:44:48 +0000[diff] [blame]2678<p>Metadata can be used as function arguments. Here <tt>llvm.dbg.value</tt>
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2679 function is using two metadata arguments.</p>
Devang Patele1d50cd2010-03-04 23:44:48 +0000[diff] [blame]2680
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2681 <pre class="doc_code">
Devang Patele1d50cd2010-03-04 23:44:48 +0000[diff] [blame]2682 call void @llvm.dbg.value(metadata !24, i64 0, metadata !25)
2683 </pre>
Devang Patele1d50cd2010-03-04 23:44:48 +0000[diff] [blame]2684
2685<p>Metadata can be attached with an instruction. Here metadata <tt>!21</tt> is
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2686 attached with <tt>add</tt> instruction using <tt>!dbg</tt> identifier.</p>
Devang Patele1d50cd2010-03-04 23:44:48 +0000[diff] [blame]2687
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]2688 <pre class="doc_code">
Devang Patele1d50cd2010-03-04 23:44:48 +0000[diff] [blame]2689 %indvar.next = add i64 %indvar, 1, !dbg !21
2690 </pre>
Chris Lattnere6a5ddd2010-01-15 21:50:19 +0000[diff] [blame]2691</div>
2692
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2693
2694<!-- *********************************************************************** -->
2695<div class="doc_section">
2696 <a name="intrinsic_globals">Intrinsic Global Variables</a>
2697</div>
2698<!-- *********************************************************************** -->
2699
2700<p>LLVM has a number of "magic" global variables that contain data that affect
2701code generation or other IR semantics. These are documented here. All globals
Chris Lattner401e10c2009-07-20 06:14:25 +0000[diff] [blame]2702of this sort should have a section specified as "<tt>llvm.metadata</tt>". This
2703section and all globals that start with "<tt>llvm.</tt>" are reserved for use
2704by LLVM.</p>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2705
2706<!-- ======================================================================= -->
2707<div class="doc_subsection">
2708<a name="intg_used">The '<tt>llvm.used</tt>' Global Variable</a>
2709</div>
2710
2711<div class="doc_text">
2712
2713<p>The <tt>@llvm.used</tt> global is an array with i8* element type which has <a
2714href="#linkage_appending">appending linkage</a>. This array contains a list of
2715pointers to global variables and functions which may optionally have a pointer
2716cast formed of bitcast or getelementptr. For example, a legal use of it is:</p>
2717
2718<pre>
2719 @X = global i8 4
2720 @Y = global i32 123
2721
2722 @llvm.used = appending global [2 x i8*] [
2723 i8* @X,
2724 i8* bitcast (i32* @Y to i8*)
2725 ], section "llvm.metadata"
2726</pre>
2727
2728<p>If a global variable appears in the <tt>@llvm.used</tt> list, then the
2729compiler, assembler, and linker are required to treat the symbol as if there is
2730a reference to the global that it cannot see. For example, if a variable has
2731internal linkage and no references other than that from the <tt>@llvm.used</tt>
2732list, it cannot be deleted. This is commonly used to represent references from
2733inline asms and other things the compiler cannot "see", and corresponds to
2734"attribute((used))" in GNU C.</p>
2735
2736<p>On some targets, the code generator must emit a directive to the assembler or
2737object file to prevent the assembler and linker from molesting the symbol.</p>
2738
2739</div>
2740
2741<!-- ======================================================================= -->
2742<div class="doc_subsection">
Chris Lattner401e10c2009-07-20 06:14:25 +0000[diff] [blame]2743<a name="intg_compiler_used">The '<tt>llvm.compiler.used</tt>' Global Variable</a>
2744</div>
2745
2746<div class="doc_text">
2747
2748<p>The <tt>@llvm.compiler.used</tt> directive is the same as the
2749<tt>@llvm.used</tt> directive, except that it only prevents the compiler from
2750touching the symbol. On targets that support it, this allows an intelligent
2751linker to optimize references to the symbol without being impeded as it would be
2752by <tt>@llvm.used</tt>.</p>
2753
2754<p>This is a rare construct that should only be used in rare circumstances, and
2755should not be exposed to source languages.</p>
2756
2757</div>
2758
2759<!-- ======================================================================= -->
2760<div class="doc_subsection">
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2761<a name="intg_global_ctors">The '<tt>llvm.global_ctors</tt>' Global Variable</a>
2762</div>
2763
2764<div class="doc_text">
David Chisnalle31e9962010-04-30 19:23:49 +0000[diff] [blame]2765<pre>
2766%0 = type { i32, void ()* }
David Chisnall27195a52010-04-30 19:27:35 +0000[diff] [blame]2767@llvm.global_ctors = appending global [1 x %0] [%0 { i32 65535, void ()* @ctor }]
David Chisnalle31e9962010-04-30 19:23:49 +0000[diff] [blame]2768</pre>
2769<p>The <tt>@llvm.global_ctors</tt> array contains a list of constructor functions and associated priorities. The functions referenced by this array will be called in ascending order of priority (i.e. lowest first) when the module is loaded. The order of functions with the same priority is not defined.
2770</p>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2771
2772</div>
2773
2774<!-- ======================================================================= -->
2775<div class="doc_subsection">
2776<a name="intg_global_dtors">The '<tt>llvm.global_dtors</tt>' Global Variable</a>
2777</div>
2778
2779<div class="doc_text">
David Chisnalle31e9962010-04-30 19:23:49 +0000[diff] [blame]2780<pre>
2781%0 = type { i32, void ()* }
David Chisnall27195a52010-04-30 19:27:35 +0000[diff] [blame]2782@llvm.global_dtors = appending global [1 x %0] [%0 { i32 65535, void ()* @dtor }]
David Chisnalle31e9962010-04-30 19:23:49 +0000[diff] [blame]2783</pre>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2784
David Chisnalle31e9962010-04-30 19:23:49 +0000[diff] [blame]2785<p>The <tt>@llvm.global_dtors</tt> array contains a list of destructor functions and associated priorities. The functions referenced by this array will be called in descending order of priority (i.e. highest first) when the module is loaded. The order of functions with the same priority is not defined.
2786</p>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2787
2788</div>
2789
2790
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2791<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2792<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
2793<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2794
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2795<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2796
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2797<p>The LLVM instruction set consists of several different classifications of
2798 instructions: <a href="#terminators">terminator
2799 instructions</a>, <a href="#binaryops">binary instructions</a>,
2800 <a href="#bitwiseops">bitwise binary instructions</a>,
2801 <a href="#memoryops">memory instructions</a>, and
2802 <a href="#otherops">other instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2803
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2804</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2805
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2806<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2807<div class="doc_subsection"> <a name="terminators">Terminator
2808Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2809
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2810<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2811
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2812<p>As mentioned <a href="#functionstructure">previously</a>, every basic block
2813 in a program ends with a "Terminator" instruction, which indicates which
2814 block should be executed after the current block is finished. These
2815 terminator instructions typically yield a '<tt>void</tt>' value: they produce
2816 control flow, not values (the one exception being the
2817 '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
2818
Duncan Sands83821c82010-04-15 20:35:54 +0000[diff] [blame]2819<p>There are seven different terminator instructions: the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2820 '<a href="#i_ret"><tt>ret</tt></a>' instruction, the
2821 '<a href="#i_br"><tt>br</tt></a>' instruction, the
2822 '<a href="#i_switch"><tt>switch</tt></a>' instruction, the
Bill Wendling21c346e2009-11-02 00:25:26 +0000[diff] [blame]2823 '<a href="#i_indirectbr">'<tt>indirectbr</tt></a>' Instruction, the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2824 '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the
2825 '<a href="#i_unwind"><tt>unwind</tt></a>' instruction, and the
2826 '<a href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2827
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2828</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2829
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2830<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2831<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
2832Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2833
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2834<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2835
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2836<h5>Syntax:</h5>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2837<pre>
2838 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]2839 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2840</pre>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2841
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2842<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2843<p>The '<tt>ret</tt>' instruction is used to return control flow (and optionally
2844 a value) from a function back to the caller.</p>
2845
2846<p>There are two forms of the '<tt>ret</tt>' instruction: one that returns a
2847 value and then causes control flow, and one that just causes control flow to
2848 occur.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2849
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2850<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2851<p>The '<tt>ret</tt>' instruction optionally accepts a single argument, the
2852 return value. The type of the return value must be a
2853 '<a href="#t_firstclass">first class</a>' type.</p>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2854
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2855<p>A function is not <a href="#wellformed">well formed</a> if it it has a
2856 non-void return type and contains a '<tt>ret</tt>' instruction with no return
2857 value or a return value with a type that does not match its type, or if it
2858 has a void return type and contains a '<tt>ret</tt>' instruction with a
2859 return value.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2860
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2861<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2862<p>When the '<tt>ret</tt>' instruction is executed, control flow returns back to
2863 the calling function's context. If the caller is a
2864 "<a href="#i_call"><tt>call</tt></a>" instruction, execution continues at the
2865 instruction after the call. If the caller was an
2866 "<a href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues at
2867 the beginning of the "normal" destination block. If the instruction returns
2868 a value, that value shall set the call or invoke instruction's return
2869 value.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2870
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2871<h5>Example:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2872<pre>
2873 ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2874 ret void <i>; Return from a void function</i>
Bill Wendling0a4bbbf2009-02-28 22:12:54 +0000[diff] [blame]2875 ret { i32, i8 } { i32 4, i8 2 } <i>; Return a struct of values 4 and 2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2876</pre>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]2877
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2878</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2879<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2880<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2881
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2882<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2883
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2884<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2885<pre>
2886 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]2887</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2888
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2889<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2890<p>The '<tt>br</tt>' instruction is used to cause control flow to transfer to a
2891 different basic block in the current function. There are two forms of this
2892 instruction, corresponding to a conditional branch and an unconditional
2893 branch.</p>
2894
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2895<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2896<p>The conditional branch form of the '<tt>br</tt>' instruction takes a single
2897 '<tt>i1</tt>' value and two '<tt>label</tt>' values. The unconditional form
2898 of the '<tt>br</tt>' instruction takes a single '<tt>label</tt>' value as a
2899 target.</p>
2900
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2901<h5>Semantics:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2902<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2903 argument is evaluated. If the value is <tt>true</tt>, control flows to the
2904 '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
2905 control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
2906
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2907<h5>Example:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]2908<pre>
2909Test:
2910 %cond = <a href="#i_icmp">icmp</a> eq i32 %a, %b
2911 br i1 %cond, label %IfEqual, label %IfUnequal
2912IfEqual:
2913 <a href="#i_ret">ret</a> i32 1
2914IfUnequal:
2915 <a href="#i_ret">ret</a> i32 0
2916</pre>
2917
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2918</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2919
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2920<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2921<div class="doc_subsubsection">
2922 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
2923</div>
2924
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2925<div class="doc_text">
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2926
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2927<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2928<pre>
2929 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
2930</pre>
2931
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2932<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2933<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2934 several different places. It is a generalization of the '<tt>br</tt>'
2935 instruction, allowing a branch to occur to one of many possible
2936 destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2937
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2938<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2939<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2940 comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination,
2941 and an array of pairs of comparison value constants and '<tt>label</tt>'s.
2942 The table is not allowed to contain duplicate constant entries.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2943
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2944<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2945<p>The <tt>switch</tt> instruction specifies a table of values and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2946 destinations. When the '<tt>switch</tt>' instruction is executed, this table
2947 is searched for the given value. If the value is found, control flow is
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]2948 transferred to the corresponding destination; otherwise, control flow is
2949 transferred to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2950
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2951<h5>Implementation:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2952<p>Depending on properties of the target machine and the particular
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2953 <tt>switch</tt> instruction, this instruction may be code generated in
2954 different ways. For example, it could be generated as a series of chained
2955 conditional branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2956
2957<h5>Example:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2958<pre>
2959 <i>; Emulate a conditional br instruction</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2960 %Val = <a href="#i_zext">zext</a> i1 %value to i32
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]2961 switch i32 %Val, label %truedest [ i32 0, label %falsedest ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2962
2963 <i>; Emulate an unconditional br instruction</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2964 switch i32 0, label %dest [ ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2965
2966 <i>; Implement a jump table:</i>
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]2967 switch i32 %val, label %otherwise [ i32 0, label %onzero
2968 i32 1, label %onone
2969 i32 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2970</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2971
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2972</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2973
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2974
2975<!-- _______________________________________________________________________ -->
2976<div class="doc_subsubsection">
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2977 <a name="i_indirectbr">'<tt>indirectbr</tt>' Instruction</a>
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2978</div>
2979
2980<div class="doc_text">
2981
2982<h5>Syntax:</h5>
2983<pre>
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2984 indirectbr &lt;somety&gt;* &lt;address&gt;, [ label &lt;dest1&gt;, label &lt;dest2&gt;, ... ]
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2985</pre>
2986
2987<h5>Overview:</h5>
2988
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2989<p>The '<tt>indirectbr</tt>' instruction implements an indirect branch to a label
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2990 within the current function, whose address is specified by
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2991 "<tt>address</tt>". Address must be derived from a <a
2992 href="#blockaddress">blockaddress</a> constant.</p>
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2993
2994<h5>Arguments:</h5>
2995
2996<p>The '<tt>address</tt>' argument is the address of the label to jump to. The
2997 rest of the arguments indicate the full set of possible destinations that the
2998 address may point to. Blocks are allowed to occur multiple times in the
2999 destination list, though this isn't particularly useful.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3000
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]3001<p>This destination list is required so that dataflow analysis has an accurate
3002 understanding of the CFG.</p>
3003
3004<h5>Semantics:</h5>
3005
3006<p>Control transfers to the block specified in the address argument. All
3007 possible destination blocks must be listed in the label list, otherwise this
3008 instruction has undefined behavior. This implies that jumps to labels
3009 defined in other functions have undefined behavior as well.</p>
3010
3011<h5>Implementation:</h5>
3012
3013<p>This is typically implemented with a jump through a register.</p>
3014
3015<h5>Example:</h5>
3016<pre>
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]3017 indirectbr i8* %Addr, [ label %bb1, label %bb2, label %bb3 ]
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]3018</pre>
3019
3020</div>
3021
3022
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3023<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3024<div class="doc_subsubsection">
3025 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
3026</div>
3027
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3028<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3029
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3030<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3031<pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]3032 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>] &lt;ptr to function ty&gt; &lt;function ptr val&gt;(&lt;function args&gt;) [<a href="#fnattrs">fn attrs</a>]
Chris Lattner76b8a332006-05-14 18:23:06 +0000[diff] [blame]3033 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3034</pre>
3035
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3036<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3037<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3038 function, with the possibility of control flow transfer to either the
3039 '<tt>normal</tt>' label or the '<tt>exception</tt>' label. If the callee
3040 function returns with the "<tt><a href="#i_ret">ret</a></tt>" instruction,
3041 control flow will return to the "normal" label. If the callee (or any
3042 indirect callees) returns with the "<a href="#i_unwind"><tt>unwind</tt></a>"
3043 instruction, control is interrupted and continued at the dynamically nearest
3044 "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3045
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3046<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3047<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3048
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3049<ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3050 <li>The optional "cconv" marker indicates which <a href="#callingconv">calling
3051 convention</a> the call should use. If none is specified, the call
3052 defaults to using C calling conventions.</li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]3053
3054 <li>The optional <a href="#paramattrs">Parameter Attributes</a> list for
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3055 return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>', and
3056 '<tt>inreg</tt>' attributes are valid here.</li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]3057
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3058 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3059 function value being invoked. In most cases, this is a direct function
3060 invocation, but indirect <tt>invoke</tt>s are just as possible, branching
3061 off an arbitrary pointer to function value.</li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3062
3063 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3064 function to be invoked. </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3065
3066 <li>'<tt>function args</tt>': argument list whose types match the function
Chris Lattner0724fbd2010-03-02 06:36:51 +0000[diff] [blame]3067 signature argument types and parameter attributes. All arguments must be
3068 of <a href="#t_firstclass">first class</a> type. If the function
3069 signature indicates the function accepts a variable number of arguments,
3070 the extra arguments can be specified.</li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3071
3072 <li>'<tt>normal label</tt>': the label reached when the called function
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3073 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3074
3075 <li>'<tt>exception label</tt>': the label reached when a callee returns with
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3076 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3077
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]3078 <li>The optional <a href="#fnattrs">function attributes</a> list. Only
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3079 '<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
3080 '<tt>readnone</tt>' attributes are valid here.</li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3081</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3082
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3083<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3084<p>This instruction is designed to operate as a standard
3085 '<tt><a href="#i_call">call</a></tt>' instruction in most regards. The
3086 primary difference is that it establishes an association with a label, which
3087 is used by the runtime library to unwind the stack.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3088
3089<p>This instruction is used in languages with destructors to ensure that proper
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3090 cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
3091 exception. Additionally, this is important for implementation of
3092 '<tt>catch</tt>' clauses in high-level languages that support them.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3093
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3094<p>For the purposes of the SSA form, the definition of the value returned by the
3095 '<tt>invoke</tt>' instruction is deemed to occur on the edge from the current
3096 block to the "normal" label. If the callee unwinds then no return value is
3097 available.</p>
Dan Gohmanf96a4992009-05-22 21:47:08 +0000[diff] [blame]3098
Chris Lattnerdf7a6802010-01-15 18:08:37 +0000[diff] [blame]3099<p>Note that the code generator does not yet completely support unwind, and
3100that the invoke/unwind semantics are likely to change in future versions.</p>
3101
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3102<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]3103<pre>
Nick Lewyckyd703f652008-03-16 07:18:12 +0000[diff] [blame]3104 %retval = invoke i32 @Test(i32 15) to label %Continue
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3105 unwind label %TestCleanup <i>; {i32}:retval set</i>
Nick Lewyckyd703f652008-03-16 07:18:12 +0000[diff] [blame]3106 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Testfnptr(i32 15) to label %Continue
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3107 unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3108</pre>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3109
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3110</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3111
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]3112<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3113
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3114<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
3115Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3116
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3117<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3118
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]3119<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3120<pre>
3121 unwind
3122</pre>
3123
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]3124<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3125<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3126 at the first callee in the dynamic call stack which used
3127 an <a href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call.
3128 This is primarily used to implement exception handling.</p>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3129
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]3130<h5>Semantics:</h5>
Chris Lattner72ed2002008-04-19 21:01:16 +0000[diff] [blame]3131<p>The '<tt>unwind</tt>' instruction causes execution of the current function to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3132 immediately halt. The dynamic call stack is then searched for the
3133 first <a href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack.
3134 Once found, execution continues at the "exceptional" destination block
3135 specified by the <tt>invoke</tt> instruction. If there is no <tt>invoke</tt>
3136 instruction in the dynamic call chain, undefined behavior results.</p>
3137
Chris Lattnerdf7a6802010-01-15 18:08:37 +0000[diff] [blame]3138<p>Note that the code generator does not yet completely support unwind, and
3139that the invoke/unwind semantics are likely to change in future versions.</p>
3140
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3141</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3142
3143<!-- _______________________________________________________________________ -->
3144
3145<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
3146Instruction</a> </div>
3147
3148<div class="doc_text">
3149
3150<h5>Syntax:</h5>
3151<pre>
3152 unreachable
3153</pre>
3154
3155<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3156<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3157 instruction is used to inform the optimizer that a particular portion of the
3158 code is not reachable. This can be used to indicate that the code after a
3159 no-return function cannot be reached, and other facts.</p>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3160
3161<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3162<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3163
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]3164</div>
3165
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3166<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3167<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3168
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3169<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3170
3171<p>Binary operators are used to do most of the computation in a program. They
3172 require two operands of the same type, execute an operation on them, and
3173 produce a single value. The operands might represent multiple data, as is
3174 the case with the <a href="#t_vector">vector</a> data type. The result value
3175 has the same type as its operands.</p>
3176
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3177<p>There are several different binary operators:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3178
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3179</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3180
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3181<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3182<div class="doc_subsubsection">
3183 <a name="i_add">'<tt>add</tt>' Instruction</a>
3184</div>
3185
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3186<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3187
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3188<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3189<pre>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3190 &lt;result&gt; = add &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Dan Gohmanfdfca792009-09-02 17:31:42 +0000[diff] [blame]3191 &lt;result&gt; = add nuw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3192 &lt;result&gt; = add nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3193 &lt;result&gt; = add nuw nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3194</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3195
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3196<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3197<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3198
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3199<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3200<p>The two arguments to the '<tt>add</tt>' instruction must
3201 be <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3202 integer values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3203
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3204<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3205<p>The value produced is the integer sum of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3206
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3207<p>If the sum has unsigned overflow, the result returned is the mathematical
3208 result modulo 2<sup>n</sup>, where n is the bit width of the result.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3209
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3210<p>Because LLVM integers use a two's complement representation, this instruction
3211 is appropriate for both signed and unsigned integers.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3212
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3213<p><tt>nuw</tt> and <tt>nsw</tt> stand for &quot;No Unsigned Wrap&quot;
3214 and &quot;No Signed Wrap&quot;, respectively. If the <tt>nuw</tt> and/or
3215 <tt>nsw</tt> keywords are present, the result value of the <tt>add</tt>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]3216 is a <a href="#trapvalues">trap value</a> if unsigned and/or signed overflow,
3217 respectively, occurs.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3218
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3219<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3220<pre>
3221 &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3222</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3223
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3224</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3225
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3226<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3227<div class="doc_subsubsection">
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3228 <a name="i_fadd">'<tt>fadd</tt>' Instruction</a>
3229</div>
3230
3231<div class="doc_text">
3232
3233<h5>Syntax:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3234<pre>
3235 &lt;result&gt; = fadd &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3236</pre>
3237
3238<h5>Overview:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3239<p>The '<tt>fadd</tt>' instruction returns the sum of its two operands.</p>
3240
3241<h5>Arguments:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3242<p>The two arguments to the '<tt>fadd</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3243 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3244 floating point values. Both arguments must have identical types.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3245
3246<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3247<p>The value produced is the floating point sum of the two operands.</p>
3248
3249<h5>Example:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3250<pre>
3251 &lt;result&gt; = fadd float 4.0, %var <i>; yields {float}:result = 4.0 + %var</i>
3252</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3253
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3254</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3255
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3256<!-- _______________________________________________________________________ -->
3257<div class="doc_subsubsection">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3258 <a name="i_sub">'<tt>sub</tt>' Instruction</a>
3259</div>
3260
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3261<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3262
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3263<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3264<pre>
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3265 &lt;result&gt; = sub &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Dan Gohmanfdfca792009-09-02 17:31:42 +0000[diff] [blame]3266 &lt;result&gt; = sub nuw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3267 &lt;result&gt; = sub nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3268 &lt;result&gt; = sub nuw nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3269</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3270
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3271<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3272<p>The '<tt>sub</tt>' instruction returns the difference of its two
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3273 operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3274
3275<p>Note that the '<tt>sub</tt>' instruction is used to represent the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3276 '<tt>neg</tt>' instruction present in most other intermediate
3277 representations.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3278
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3279<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3280<p>The two arguments to the '<tt>sub</tt>' instruction must
3281 be <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3282 integer values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3283
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3284<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3285<p>The value produced is the integer difference of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3286
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3287<p>If the difference has unsigned overflow, the result returned is the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3288 mathematical result modulo 2<sup>n</sup>, where n is the bit width of the
3289 result.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3290
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3291<p>Because LLVM integers use a two's complement representation, this instruction
3292 is appropriate for both signed and unsigned integers.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3293
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3294<p><tt>nuw</tt> and <tt>nsw</tt> stand for &quot;No Unsigned Wrap&quot;
3295 and &quot;No Signed Wrap&quot;, respectively. If the <tt>nuw</tt> and/or
3296 <tt>nsw</tt> keywords are present, the result value of the <tt>sub</tt>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]3297 is a <a href="#trapvalues">trap value</a> if unsigned and/or signed overflow,
3298 respectively, occurs.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3299
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3300<h5>Example:</h5>
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]3301<pre>
3302 &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3303 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3304</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3305
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3306</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3307
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3308<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3309<div class="doc_subsubsection">
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3310 <a name="i_fsub">'<tt>fsub</tt>' Instruction</a>
3311</div>
3312
3313<div class="doc_text">
3314
3315<h5>Syntax:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3316<pre>
3317 &lt;result&gt; = fsub &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3318</pre>
3319
3320<h5>Overview:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3321<p>The '<tt>fsub</tt>' instruction returns the difference of its two
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3322 operands.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3323
3324<p>Note that the '<tt>fsub</tt>' instruction is used to represent the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3325 '<tt>fneg</tt>' instruction present in most other intermediate
3326 representations.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3327
3328<h5>Arguments:</h5>
Bill Wendlingd9fe2982009-07-20 02:32:41 +0000[diff] [blame]3329<p>The two arguments to the '<tt>fsub</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3330 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3331 floating point values. Both arguments must have identical types.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3332
3333<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3334<p>The value produced is the floating point difference of the two operands.</p>
3335
3336<h5>Example:</h5>
3337<pre>
3338 &lt;result&gt; = fsub float 4.0, %var <i>; yields {float}:result = 4.0 - %var</i>
3339 &lt;result&gt; = fsub float -0.0, %val <i>; yields {float}:result = -%var</i>
3340</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3341
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3342</div>
3343
3344<!-- _______________________________________________________________________ -->
3345<div class="doc_subsubsection">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3346 <a name="i_mul">'<tt>mul</tt>' Instruction</a>
3347</div>
3348
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3349<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3350
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3351<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3352<pre>
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3353 &lt;result&gt; = mul &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Dan Gohmanfdfca792009-09-02 17:31:42 +0000[diff] [blame]3354 &lt;result&gt; = mul nuw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3355 &lt;result&gt; = mul nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3356 &lt;result&gt; = mul nuw nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3357</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3358
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3359<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3360<p>The '<tt>mul</tt>' instruction returns the product of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3361
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3362<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3363<p>The two arguments to the '<tt>mul</tt>' instruction must
3364 be <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3365 integer values. Both arguments must have identical types.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3366
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3367<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3368<p>The value produced is the integer product of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3369
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3370<p>If the result of the multiplication has unsigned overflow, the result
3371 returned is the mathematical result modulo 2<sup>n</sup>, where n is the bit
3372 width of the result.</p>
3373
3374<p>Because LLVM integers use a two's complement representation, and the result
3375 is the same width as the operands, this instruction returns the correct
3376 result for both signed and unsigned integers. If a full product
3377 (e.g. <tt>i32</tt>x<tt>i32</tt>-><tt>i64</tt>) is needed, the operands should
3378 be sign-extended or zero-extended as appropriate to the width of the full
3379 product.</p>
3380
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3381<p><tt>nuw</tt> and <tt>nsw</tt> stand for &quot;No Unsigned Wrap&quot;
3382 and &quot;No Signed Wrap&quot;, respectively. If the <tt>nuw</tt> and/or
3383 <tt>nsw</tt> keywords are present, the result value of the <tt>mul</tt>
Dan Gohmanfff6c532010-04-22 23:14:21 +0000[diff] [blame]3384 is a <a href="#trapvalues">trap value</a> if unsigned and/or signed overflow,
3385 respectively, occurs.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3386
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3387<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3388<pre>
3389 &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3390</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3391
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3392</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3393
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3394<!-- _______________________________________________________________________ -->
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3395<div class="doc_subsubsection">
3396 <a name="i_fmul">'<tt>fmul</tt>' Instruction</a>
3397</div>
3398
3399<div class="doc_text">
3400
3401<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3402<pre>
3403 &lt;result&gt; = fmul &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3404</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3405
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3406<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3407<p>The '<tt>fmul</tt>' instruction returns the product of its two operands.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3408
3409<h5>Arguments:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3410<p>The two arguments to the '<tt>fmul</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3411 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3412 floating point values. Both arguments must have identical types.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3413
3414<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3415<p>The value produced is the floating point product of the two operands.</p>
3416
3417<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3418<pre>
3419 &lt;result&gt; = fmul float 4.0, %var <i>; yields {float}:result = 4.0 * %var</i>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3420</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3421
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3422</div>
3423
3424<!-- _______________________________________________________________________ -->
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3425<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
3426</a></div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3427
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3428<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3429
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3430<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3431<pre>
3432 &lt;result&gt; = udiv &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3433</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3434
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3435<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3436<p>The '<tt>udiv</tt>' instruction returns the quotient of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3437
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3438<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3439<p>The two arguments to the '<tt>udiv</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3440 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3441 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3442
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3443<h5>Semantics:</h5>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3444<p>The value produced is the unsigned integer quotient of the two operands.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3445
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3446<p>Note that unsigned integer division and signed integer division are distinct
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3447 operations; for signed integer division, use '<tt>sdiv</tt>'.</p>
3448
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3449<p>Division by zero leads to undefined behavior.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3450
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3451<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3452<pre>
3453 &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3454</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3455
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3456</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3457
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3458<!-- _______________________________________________________________________ -->
3459<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
3460</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3461
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3462<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3463
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3464<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3465<pre>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3466 &lt;result&gt; = sdiv &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Dan Gohmanfdfca792009-09-02 17:31:42 +0000[diff] [blame]3467 &lt;result&gt; = sdiv exact &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3468</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3469
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3470<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3471<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3472
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3473<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3474<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3475 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3476 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3477
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3478<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3479<p>The value produced is the signed integer quotient of the two operands rounded
3480 towards zero.</p>
3481
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3482<p>Note that signed integer division and unsigned integer division are distinct
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3483 operations; for unsigned integer division, use '<tt>udiv</tt>'.</p>
3484
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3485<p>Division by zero leads to undefined behavior. Overflow also leads to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3486 undefined behavior; this is a rare case, but can occur, for example, by doing
3487 a 32-bit division of -2147483648 by -1.</p>
3488
Dan Gohman9c5beed2009-07-22 00:04:19 +0000[diff] [blame]3489<p>If the <tt>exact</tt> keyword is present, the result value of the
Dan Gohman27ef9972010-04-23 15:23:32 +0000[diff] [blame]3490 <tt>sdiv</tt> is a <a href="#trapvalues">trap value</a> if the result would
Dan Gohman38da9272010-07-11 00:08:34 +0000[diff] [blame]3491 be rounded.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3492
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3493<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3494<pre>
3495 &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3496</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3497
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3498</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3499
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3500<!-- _______________________________________________________________________ -->
3501<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3502Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3503
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3504<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3505
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3506<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3507<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]3508 &lt;result&gt; = fdiv &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3509</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3510
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3511<h5>Overview:</h5>
3512<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3513
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3514<h5>Arguments:</h5>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3515<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3516 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3517 floating point values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3518
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3519<h5>Semantics:</h5>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3520<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3521
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3522<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3523<pre>
3524 &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]3525</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3526
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3527</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3528
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3529<!-- _______________________________________________________________________ -->
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3530<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
3531</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3532
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3533<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3534
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3535<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3536<pre>
3537 &lt;result&gt; = urem &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3538</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3539
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3540<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3541<p>The '<tt>urem</tt>' instruction returns the remainder from the unsigned
3542 division of its two arguments.</p>
3543
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3544<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3545<p>The two arguments to the '<tt>urem</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3546 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3547 values. Both arguments must have identical types.</p>
3548
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3549<h5>Semantics:</h5>
3550<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3551 This instruction always performs an unsigned division to get the
3552 remainder.</p>
3553
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3554<p>Note that unsigned integer remainder and signed integer remainder are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3555 distinct operations; for signed integer remainder, use '<tt>srem</tt>'.</p>
3556
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3557<p>Taking the remainder of a division by zero leads to undefined behavior.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3558
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3559<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3560<pre>
3561 &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3562</pre>
3563
3564</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3565
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3566<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3567<div class="doc_subsubsection">
3568 <a name="i_srem">'<tt>srem</tt>' Instruction</a>
3569</div>
3570
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3571<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3572
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3573<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3574<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]3575 &lt;result&gt; = srem &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3576</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3577
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3578<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3579<p>The '<tt>srem</tt>' instruction returns the remainder from the signed
3580 division of its two operands. This instruction can also take
3581 <a href="#t_vector">vector</a> versions of the values in which case the
3582 elements must be integers.</p>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]3583
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3584<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3585<p>The two arguments to the '<tt>srem</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3586 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3587 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3588
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3589<h5>Semantics:</h5>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3590<p>This instruction returns the <i>remainder</i> of a division (where the result
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3591 has the same sign as the dividend, <tt>op1</tt>), not the <i>modulo</i>
3592 operator (where the result has the same sign as the divisor, <tt>op2</tt>) of
3593 a value. For more information about the difference,
3594 see <a href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
3595 Math Forum</a>. For a table of how this is implemented in various languages,
3596 please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
3597 Wikipedia: modulo operation</a>.</p>
3598
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3599<p>Note that signed integer remainder and unsigned integer remainder are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3600 distinct operations; for unsigned integer remainder, use '<tt>urem</tt>'.</p>
3601
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3602<p>Taking the remainder of a division by zero leads to undefined behavior.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3603 Overflow also leads to undefined behavior; this is a rare case, but can
3604 occur, for example, by taking the remainder of a 32-bit division of
3605 -2147483648 by -1. (The remainder doesn't actually overflow, but this rule
3606 lets srem be implemented using instructions that return both the result of
3607 the division and the remainder.)</p>
3608
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3609<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3610<pre>
3611 &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3612</pre>
3613
3614</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3615
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3616<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3617<div class="doc_subsubsection">
3618 <a name="i_frem">'<tt>frem</tt>' Instruction</a> </div>
3619
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3620<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3621
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3622<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3623<pre>
3624 &lt;result&gt; = frem &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3625</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3626
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3627<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3628<p>The '<tt>frem</tt>' instruction returns the remainder from the division of
3629 its two operands.</p>
3630
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3631<h5>Arguments:</h5>
3632<p>The two arguments to the '<tt>frem</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3633 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3634 floating point values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3635
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3636<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3637<p>This instruction returns the <i>remainder</i> of a division. The remainder
3638 has the same sign as the dividend.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3639
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3640<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3641<pre>
3642 &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]3643</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3644
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3645</div>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]3646
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3647<!-- ======================================================================= -->
3648<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
3649Operations</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3650
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3651<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3652
3653<p>Bitwise binary operators are used to do various forms of bit-twiddling in a
3654 program. They are generally very efficient instructions and can commonly be
3655 strength reduced from other instructions. They require two operands of the
3656 same type, execute an operation on them, and produce a single value. The
3657 resulting value is the same type as its operands.</p>
3658
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3659</div>
3660
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3661<!-- _______________________________________________________________________ -->
3662<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
3663Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3664
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3665<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3666
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3667<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3668<pre>
3669 &lt;result&gt; = shl &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3670</pre>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3671
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3672<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3673<p>The '<tt>shl</tt>' instruction returns the first operand shifted to the left
3674 a specified number of bits.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3675
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3676<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3677<p>Both arguments to the '<tt>shl</tt>' instruction must be the
3678 same <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3679 integer type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3680
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3681<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3682<p>The value produced is <tt>op1</tt> * 2<sup><tt>op2</tt></sup> mod
3683 2<sup>n</sup>, where <tt>n</tt> is the width of the result. If <tt>op2</tt>
3684 is (statically or dynamically) negative or equal to or larger than the number
3685 of bits in <tt>op1</tt>, the result is undefined. If the arguments are
3686 vectors, each vector element of <tt>op1</tt> is shifted by the corresponding
3687 shift amount in <tt>op2</tt>.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3688
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3689<h5>Example:</h5>
3690<pre>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3691 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
3692 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
3693 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3694 &lt;result&gt; = shl i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]3695 &lt;result&gt; = shl &lt;2 x i32&gt; &lt; i32 1, i32 1&gt;, &lt; i32 1, i32 2&gt; <i>; yields: result=&lt;2 x i32&gt; &lt; i32 2, i32 4&gt;</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3696</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3697
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3698</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3699
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3700<!-- _______________________________________________________________________ -->
3701<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
3702Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3703
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3704<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3705
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3706<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3707<pre>
3708 &lt;result&gt; = lshr &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3709</pre>
3710
3711<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3712<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
3713 operand shifted to the right a specified number of bits with zero fill.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3714
3715<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3716<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3717 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3718 type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3719
3720<h5>Semantics:</h5>
3721<p>This instruction always performs a logical shift right operation. The most
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3722 significant bits of the result will be filled with zero bits after the shift.
3723 If <tt>op2</tt> is (statically or dynamically) equal to or larger than the
3724 number of bits in <tt>op1</tt>, the result is undefined. If the arguments are
3725 vectors, each vector element of <tt>op1</tt> is shifted by the corresponding
3726 shift amount in <tt>op2</tt>.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3727
3728<h5>Example:</h5>
3729<pre>
3730 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
3731 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
3732 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
3733 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3734 &lt;result&gt; = lshr i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]3735 &lt;result&gt; = lshr &lt;2 x i32&gt; &lt; i32 -2, i32 4&gt;, &lt; i32 1, i32 2&gt; <i>; yields: result=&lt;2 x i32&gt; &lt; i32 0x7FFFFFFF, i32 1&gt;</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3736</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3737
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3738</div>
3739
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3740<!-- _______________________________________________________________________ -->
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3741<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
3742Instruction</a> </div>
3743<div class="doc_text">
3744
3745<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3746<pre>
3747 &lt;result&gt; = ashr &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3748</pre>
3749
3750<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3751<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
3752 operand shifted to the right a specified number of bits with sign
3753 extension.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3754
3755<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3756<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3757 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3758 type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3759
3760<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3761<p>This instruction always performs an arithmetic shift right operation, The
3762 most significant bits of the result will be filled with the sign bit
3763 of <tt>op1</tt>. If <tt>op2</tt> is (statically or dynamically) equal to or
3764 larger than the number of bits in <tt>op1</tt>, the result is undefined. If
3765 the arguments are vectors, each vector element of <tt>op1</tt> is shifted by
3766 the corresponding shift amount in <tt>op2</tt>.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3767
3768<h5>Example:</h5>
3769<pre>
3770 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
3771 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
3772 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
3773 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3774 &lt;result&gt; = ashr i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]3775 &lt;result&gt; = ashr &lt;2 x i32&gt; &lt; i32 -2, i32 4&gt;, &lt; i32 1, i32 3&gt; <i>; yields: result=&lt;2 x i32&gt; &lt; i32 -1, i32 0&gt;</i>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3776</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3777
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3778</div>
3779
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3780<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3781<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
3782Instruction</a> </div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3783
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3784<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3785
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3786<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3787<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]3788 &lt;result&gt; = and &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3789</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3790
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3791<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3792<p>The '<tt>and</tt>' instruction returns the bitwise logical and of its two
3793 operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3794
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3795<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3796<p>The two arguments to the '<tt>and</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3797 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3798 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3799
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3800<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3801<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3802
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3803<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3804 <tbody>
3805 <tr>
3806 <td>In0</td>
3807 <td>In1</td>
3808 <td>Out</td>
3809 </tr>
3810 <tr>
3811 <td>0</td>
3812 <td>0</td>
3813 <td>0</td>
3814 </tr>
3815 <tr>
3816 <td>0</td>
3817 <td>1</td>
3818 <td>0</td>
3819 </tr>
3820 <tr>
3821 <td>1</td>
3822 <td>0</td>
3823 <td>0</td>
3824 </tr>
3825 <tr>
3826 <td>1</td>
3827 <td>1</td>
3828 <td>1</td>
3829 </tr>
3830 </tbody>
3831</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3832
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3833<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3834<pre>
3835 &lt;result&gt; = and i32 4, %var <i>; yields {i32}:result = 4 &amp; %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3836 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
3837 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3838</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3839</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3840<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3841<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3842
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3843<div class="doc_text">
3844
3845<h5>Syntax:</h5>
3846<pre>
3847 &lt;result&gt; = or &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3848</pre>
3849
3850<h5>Overview:</h5>
3851<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive or of its
3852 two operands.</p>
3853
3854<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3855<p>The two arguments to the '<tt>or</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3856 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3857 values. Both arguments must have identical types.</p>
3858
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3859<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3860<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3861
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3862<table border="1" cellspacing="0" cellpadding="4">
3863 <tbody>
3864 <tr>
3865 <td>In0</td>
3866 <td>In1</td>
3867 <td>Out</td>
3868 </tr>
3869 <tr>
3870 <td>0</td>
3871 <td>0</td>
3872 <td>0</td>
3873 </tr>
3874 <tr>
3875 <td>0</td>
3876 <td>1</td>
3877 <td>1</td>
3878 </tr>
3879 <tr>
3880 <td>1</td>
3881 <td>0</td>
3882 <td>1</td>
3883 </tr>
3884 <tr>
3885 <td>1</td>
3886 <td>1</td>
3887 <td>1</td>
3888 </tr>
3889 </tbody>
3890</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3891
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3892<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3893<pre>
3894 &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3895 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
3896 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3897</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3898
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3899</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3900
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3901<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3902<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
3903Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3904
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3905<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3906
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3907<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3908<pre>
3909 &lt;result&gt; = xor &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3910</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3911
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3912<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3913<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive or of
3914 its two operands. The <tt>xor</tt> is used to implement the "one's
3915 complement" operation, which is the "~" operator in C.</p>
3916
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3917<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3918<p>The two arguments to the '<tt>xor</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3919 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3920 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3921
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3922<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3923<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3924
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3925<table border="1" cellspacing="0" cellpadding="4">
3926 <tbody>
3927 <tr>
3928 <td>In0</td>
3929 <td>In1</td>
3930 <td>Out</td>
3931 </tr>
3932 <tr>
3933 <td>0</td>
3934 <td>0</td>
3935 <td>0</td>
3936 </tr>
3937 <tr>
3938 <td>0</td>
3939 <td>1</td>
3940 <td>1</td>
3941 </tr>
3942 <tr>
3943 <td>1</td>
3944 <td>0</td>
3945 <td>1</td>
3946 </tr>
3947 <tr>
3948 <td>1</td>
3949 <td>1</td>
3950 <td>0</td>
3951 </tr>
3952 </tbody>
3953</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3954
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3955<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3956<pre>
3957 &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3958 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
3959 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
3960 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3961</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3962
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3963</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3964
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3965<!-- ======================================================================= -->
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3966<div class="doc_subsection">
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3967 <a name="vectorops">Vector Operations</a>
3968</div>
3969
3970<div class="doc_text">
3971
3972<p>LLVM supports several instructions to represent vector operations in a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3973 target-independent manner. These instructions cover the element-access and
3974 vector-specific operations needed to process vectors effectively. While LLVM
3975 does directly support these vector operations, many sophisticated algorithms
3976 will want to use target-specific intrinsics to take full advantage of a
3977 specific target.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3978
3979</div>
3980
3981<!-- _______________________________________________________________________ -->
3982<div class="doc_subsubsection">
3983 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
3984</div>
3985
3986<div class="doc_text">
3987
3988<h5>Syntax:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3989<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3990 &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]3991</pre>
3992
3993<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3994<p>The '<tt>extractelement</tt>' instruction extracts a single scalar element
3995 from a vector at a specified index.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3996
3997
3998<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3999<p>The first operand of an '<tt>extractelement</tt>' instruction is a value
4000 of <a href="#t_vector">vector</a> type. The second operand is an index
4001 indicating the position from which to extract the element. The index may be
4002 a variable.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4003
4004<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4005<p>The result is a scalar of the same type as the element type of
4006 <tt>val</tt>. Its value is the value at position <tt>idx</tt> of
4007 <tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
4008 results are undefined.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4009
4010<h5>Example:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4011<pre>
Gabor Greifa5b6f452009-10-28 13:14:50 +0000[diff] [blame]4012 &lt;result&gt; = extractelement &lt;4 x i32&gt; %vec, i32 0 <i>; yields i32</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4013</pre>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4014
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4015</div>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4016
4017<!-- _______________________________________________________________________ -->
4018<div class="doc_subsubsection">
4019 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
4020</div>
4021
4022<div class="doc_text">
4023
4024<h5>Syntax:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4025<pre>
Dan Gohmanf3480b92008-05-12 23:38:42 +0000[diff] [blame]4026 &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]4027</pre>
4028
4029<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4030<p>The '<tt>insertelement</tt>' instruction inserts a scalar element into a
4031 vector at a specified index.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4032
4033<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4034<p>The first operand of an '<tt>insertelement</tt>' instruction is a value
4035 of <a href="#t_vector">vector</a> type. The second operand is a scalar value
4036 whose type must equal the element type of the first operand. The third
4037 operand is an index indicating the position at which to insert the value.
4038 The index may be a variable.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4039
4040<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4041<p>The result is a vector of the same type as <tt>val</tt>. Its element values
4042 are those of <tt>val</tt> except at position <tt>idx</tt>, where it gets the
4043 value <tt>elt</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
4044 results are undefined.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4045
4046<h5>Example:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4047<pre>
Gabor Greifa5b6f452009-10-28 13:14:50 +0000[diff] [blame]4048 &lt;result&gt; = 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]4049</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4050
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4051</div>
4052
4053<!-- _______________________________________________________________________ -->
4054<div class="doc_subsubsection">
4055 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
4056</div>
4057
4058<div class="doc_text">
4059
4060<h5>Syntax:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4061<pre>
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]4062 &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;m x i32&gt; &lt;mask&gt; <i>; yields &lt;m x &lt;ty&gt;&gt;</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4063</pre>
4064
4065<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4066<p>The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
4067 from two input vectors, returning a vector with the same element type as the
4068 input and length that is the same as the shuffle mask.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4069
4070<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4071<p>The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
4072 with types that match each other. The third argument is a shuffle mask whose
4073 element type is always 'i32'. The result of the instruction is a vector
4074 whose length is the same as the shuffle mask and whose element type is the
4075 same as the element type of the first two operands.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4076
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4077<p>The shuffle mask operand is required to be a constant vector with either
4078 constant integer or undef values.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4079
4080<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4081<p>The elements of the two input vectors are numbered from left to right across
4082 both of the vectors. The shuffle mask operand specifies, for each element of
4083 the result vector, which element of the two input vectors the result element
4084 gets. The element selector may be undef (meaning "don't care") and the
4085 second operand may be undef if performing a shuffle from only one vector.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4086
4087<h5>Example:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4088<pre>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4089 &lt;result&gt; = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]4090 &lt;4 x i32&gt; &lt;i32 0, i32 4, i32 1, i32 5&gt; <i>; yields &lt;4 x i32&gt;</i>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4091 &lt;result&gt; = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; undef,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4092 &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.
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4093 &lt;result&gt; = shufflevector &lt;8 x i32&gt; %v1, &lt;8 x i32&gt; undef,
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]4094 &lt;4 x i32&gt; &lt;i32 0, i32 1, i32 2, i32 3&gt; <i>; yields &lt;4 x i32&gt;</i>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4095 &lt;result&gt; = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]4096 &lt;8 x i32&gt; &lt;i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7 &gt; <i>; yields &lt;8 x i32&gt;</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4097</pre>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4098
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4099</div>
Tanya Lattner09474292006-04-14 19:24:33 +0000[diff] [blame]4100
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]4101<!-- ======================================================================= -->
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4102<div class="doc_subsection">
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4103 <a name="aggregateops">Aggregate Operations</a>
4104</div>
4105
4106<div class="doc_text">
4107
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4108<p>LLVM supports several instructions for working with
4109 <a href="#t_aggregate">aggregate</a> values.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4110
4111</div>
4112
4113<!-- _______________________________________________________________________ -->
4114<div class="doc_subsubsection">
4115 <a name="i_extractvalue">'<tt>extractvalue</tt>' Instruction</a>
4116</div>
4117
4118<div class="doc_text">
4119
4120<h5>Syntax:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4121<pre>
4122 &lt;result&gt; = extractvalue &lt;aggregate type&gt; &lt;val&gt;, &lt;idx&gt;{, &lt;idx&gt;}*
4123</pre>
4124
4125<h5>Overview:</h5>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4126<p>The '<tt>extractvalue</tt>' instruction extracts the value of a member field
4127 from an <a href="#t_aggregate">aggregate</a> value.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4128
4129<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4130<p>The first operand of an '<tt>extractvalue</tt>' instruction is a value
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]4131 of <a href="#t_struct">struct</a> or
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4132 <a href="#t_array">array</a> type. The operands are constant indices to
4133 specify which value to extract in a similar manner as indices in a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4134 '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.</p>
Frits van Bommel13242892010-12-05 20:54:38 +0000[diff] [blame]4135 <p>The major differences to <tt>getelementptr</tt> indexing are:</p>
4136 <ul>
4137 <li>Since the value being indexed is not a pointer, the first index is
4138 omitted and assumed to be zero.</li>
4139 <li>At least one index must be specified.</li>
4140 <li>Not only struct indices but also array indices must be in
4141 bounds.</li>
4142 </ul>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4143
4144<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4145<p>The result is the value at the position in the aggregate specified by the
4146 index operands.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4147
4148<h5>Example:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4149<pre>
Gabor Greifa5b6f452009-10-28 13:14:50 +0000[diff] [blame]4150 &lt;result&gt; = extractvalue {i32, float} %agg, 0 <i>; yields i32</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4151</pre>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4152
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4153</div>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4154
4155<!-- _______________________________________________________________________ -->
4156<div class="doc_subsubsection">
4157 <a name="i_insertvalue">'<tt>insertvalue</tt>' Instruction</a>
4158</div>
4159
4160<div class="doc_text">
4161
4162<h5>Syntax:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4163<pre>
Jeffrey Yasskin7a088cf2010-01-11 19:19:26 +0000[diff] [blame]4164 &lt;result&gt; = insertvalue &lt;aggregate type&gt; &lt;val&gt;, &lt;ty&gt; &lt;elt&gt;, &lt;idx&gt; <i>; yields &lt;aggregate type&gt;</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4165</pre>
4166
4167<h5>Overview:</h5>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4168<p>The '<tt>insertvalue</tt>' instruction inserts a value into a member field
4169 in an <a href="#t_aggregate">aggregate</a> value.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4170
4171<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4172<p>The first operand of an '<tt>insertvalue</tt>' instruction is a value
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]4173 of <a href="#t_struct">struct</a> or
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4174 <a href="#t_array">array</a> type. The second operand is a first-class
4175 value to insert. The following operands are constant indices indicating
4176 the position at which to insert the value in a similar manner as indices in a
Frits van Bommel13242892010-12-05 20:54:38 +0000[diff] [blame]4177 '<tt><a href="#i_extractvalue">extractvalue</a></tt>' instruction. The
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4178 value to insert must have the same type as the value identified by the
4179 indices.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4180
4181<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4182<p>The result is an aggregate of the same type as <tt>val</tt>. Its value is
4183 that of <tt>val</tt> except that the value at the position specified by the
4184 indices is that of <tt>elt</tt>.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4185
4186<h5>Example:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4187<pre>
Jeffrey Yasskin7a088cf2010-01-11 19:19:26 +0000[diff] [blame]4188 %agg1 = insertvalue {i32, float} undef, i32 1, 0 <i>; yields {i32 1, float undef}</i>
4189 %agg2 = insertvalue {i32, float} %agg1, float %val, 1 <i>; yields {i32 1, float %val}</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4190</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4191
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]4192</div>
4193
4194
4195<!-- ======================================================================= -->
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4196<div class="doc_subsection">
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]4197 <a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4198</div>
4199
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4200<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4201
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4202<p>A key design point of an SSA-based representation is how it represents
4203 memory. In LLVM, no memory locations are in SSA form, which makes things
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]4204 very simple. This section describes how to read, write, and allocate
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4205 memory in LLVM.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4206
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4207</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4208
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4209<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4210<div class="doc_subsubsection">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4211 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
4212</div>
4213
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4214<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4215
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4216<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4217<pre>
Dan Gohmanf75a7d32010-05-28 01:14:11 +0000[diff] [blame]4218 &lt;result&gt; = alloca &lt;type&gt;[, &lt;ty&gt; &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4219</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4220
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4221<h5>Overview:</h5>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]4222<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4223 currently executing function, to be automatically released when this function
4224 returns to its caller. The object is always allocated in the generic address
4225 space (address space zero).</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4226
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4227<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4228<p>The '<tt>alloca</tt>' instruction
4229 allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt> bytes of memory on the
4230 runtime stack, returning a pointer of the appropriate type to the program.
4231 If "NumElements" is specified, it is the number of elements allocated,
4232 otherwise "NumElements" is defaulted to be one. If a constant alignment is
4233 specified, the value result of the allocation is guaranteed to be aligned to
4234 at least that boundary. If not specified, or if zero, the target can choose
4235 to align the allocation on any convenient boundary compatible with the
4236 type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4237
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4238<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4239
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4240<h5>Semantics:</h5>
Bill Wendling871eb0a2009-05-08 20:49:29 +0000[diff] [blame]4241<p>Memory is allocated; a pointer is returned. The operation is undefined if
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4242 there is insufficient stack space for the allocation. '<tt>alloca</tt>'d
4243 memory is automatically released when the function returns. The
4244 '<tt>alloca</tt>' instruction is commonly used to represent automatic
4245 variables that must have an address available. When the function returns
4246 (either with the <tt><a href="#i_ret">ret</a></tt>
4247 or <tt><a href="#i_unwind">unwind</a></tt> instructions), the memory is
4248 reclaimed. Allocating zero bytes is legal, but the result is undefined.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4249
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4250<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4251<pre>
Dan Gohman81e21672009-01-04 23:49:44 +0000[diff] [blame]4252 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
4253 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
4254 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
4255 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4256</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4257
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4258</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]4259
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4260<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4261<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
4262Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4263
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4264<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4265
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4266<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4267<pre>
Bill Wendling7c78dbb2010-02-25 21:23:24 +0000[diff] [blame]4268 &lt;result&gt; = load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;][, !nontemporal !&lt;index&gt;]
4269 &lt;result&gt; = volatile load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;][, !nontemporal !&lt;index&gt;]
4270 !&lt;index&gt; = !{ i32 1 }
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4271</pre>
4272
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4273<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4274<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4275
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4276<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4277<p>The argument to the '<tt>load</tt>' instruction specifies the memory address
4278 from which to load. The pointer must point to
4279 a <a href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
4280 marked as <tt>volatile</tt>, then the optimizer is not allowed to modify the
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]4281 number or order of execution of this <tt>load</tt> with other <a
4282 href="#volatile">volatile operations</a>.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4283
Bill Wendling7c78dbb2010-02-25 21:23:24 +0000[diff] [blame]4284<p>The optional constant <tt>align</tt> argument specifies the alignment of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4285 operation (that is, the alignment of the memory address). A value of 0 or an
Bill Wendling7c78dbb2010-02-25 21:23:24 +0000[diff] [blame]4286 omitted <tt>align</tt> argument means that the operation has the preferential
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4287 alignment for the target. It is the responsibility of the code emitter to
4288 ensure that the alignment information is correct. Overestimating the
Bill Wendling7c78dbb2010-02-25 21:23:24 +0000[diff] [blame]4289 alignment results in undefined behavior. Underestimating the alignment may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4290 produce less efficient code. An alignment of 1 is always safe.</p>
4291
Bill Wendling7c78dbb2010-02-25 21:23:24 +0000[diff] [blame]4292<p>The optional <tt>!nontemporal</tt> metadata must reference a single
4293 metatadata name &lt;index&gt; corresponding to a metadata node with
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]4294 one <tt>i32</tt> entry of value 1. The existence of
Bill Wendling7c78dbb2010-02-25 21:23:24 +0000[diff] [blame]4295 the <tt>!nontemporal</tt> metatadata on the instruction tells the optimizer
4296 and code generator that this load is not expected to be reused in the cache.
4297 The code generator may select special instructions to save cache bandwidth,
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]4298 such as the <tt>MOVNT</tt> instruction on x86.</p>
David Greene8939b0d2010-02-16 20:50:18 +0000[diff] [blame]4299
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4300<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4301<p>The location of memory pointed to is loaded. If the value being loaded is of
4302 scalar type then the number of bytes read does not exceed the minimum number
4303 of bytes needed to hold all bits of the type. For example, loading an
4304 <tt>i24</tt> reads at most three bytes. When loading a value of a type like
4305 <tt>i20</tt> with a size that is not an integral number of bytes, the result
4306 is undefined if the value was not originally written using a store of the
4307 same type.</p>
4308
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4309<h5>Examples:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4310<pre>
4311 %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
4312 <a href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4313 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4314</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4315
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4316</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4317
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4318<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4319<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
4320Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4321
Reid Spencer035ab572006-11-09 21:18:01 +0000[diff] [blame]4322<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4323
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4324<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4325<pre>
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]4326 store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;][, !nontemporal !&lt;index&gt;] <i>; yields {void}</i>
4327 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;][, !nontemporal !&lt;index&gt;] <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4328</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4329
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4330<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4331<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4332
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4333<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4334<p>There are two arguments to the '<tt>store</tt>' instruction: a value to store
4335 and an address at which to store it. The type of the
4336 '<tt>&lt;pointer&gt;</tt>' operand must be a pointer to
4337 the <a href="#t_firstclass">first class</a> type of the
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]4338 '<tt>&lt;value&gt;</tt>' operand. If the <tt>store</tt> is marked as
4339 <tt>volatile</tt>, then the optimizer is not allowed to modify the number or
4340 order of execution of this <tt>store</tt> with other <a
4341 href="#volatile">volatile operations</a>.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4342
4343<p>The optional constant "align" argument specifies the alignment of the
4344 operation (that is, the alignment of the memory address). A value of 0 or an
4345 omitted "align" argument means that the operation has the preferential
4346 alignment for the target. It is the responsibility of the code emitter to
4347 ensure that the alignment information is correct. Overestimating the
4348 alignment results in an undefined behavior. Underestimating the alignment may
4349 produce less efficient code. An alignment of 1 is always safe.</p>
4350
David Greene8939b0d2010-02-16 20:50:18 +0000[diff] [blame]4351<p>The optional !nontemporal metadata must reference a single metatadata
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]4352 name &lt;index&gt; corresponding to a metadata node with one i32 entry of
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]4353 value 1. The existence of the !nontemporal metatadata on the
David Greene8939b0d2010-02-16 20:50:18 +0000[diff] [blame]4354 instruction tells the optimizer and code generator that this load is
4355 not expected to be reused in the cache. The code generator may
4356 select special instructions to save cache bandwidth, such as the
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]4357 MOVNT instruction on x86.</p>
David Greene8939b0d2010-02-16 20:50:18 +0000[diff] [blame]4358
4359
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4360<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4361<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>' at the
4362 location specified by the '<tt>&lt;pointer&gt;</tt>' operand. If
4363 '<tt>&lt;value&gt;</tt>' is of scalar type then the number of bytes written
4364 does not exceed the minimum number of bytes needed to hold all bits of the
4365 type. For example, storing an <tt>i24</tt> writes at most three bytes. When
4366 writing a value of a type like <tt>i20</tt> with a size that is not an
4367 integral number of bytes, it is unspecified what happens to the extra bits
4368 that do not belong to the type, but they will typically be overwritten.</p>
4369
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4370<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4371<pre>
4372 %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling8c6c72d2007-10-22 05:10:05 +0000[diff] [blame]4373 store i32 3, i32* %ptr <i>; yields {void}</i>
4374 %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]4375</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4376
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]4377</div>
4378
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4379<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4380<div class="doc_subsubsection">
4381 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
4382</div>
4383
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4384<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4385
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4386<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4387<pre>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4388 &lt;result&gt; = getelementptr &lt;pty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
Dan Gohmandd8004d2009-07-27 21:53:46 +0000[diff] [blame]4389 &lt;result&gt; = getelementptr inbounds &lt;pty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4390</pre>
4391
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4392<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4393<p>The '<tt>getelementptr</tt>' instruction is used to get the address of a
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4394 subelement of an <a href="#t_aggregate">aggregate</a> data structure.
4395 It performs address calculation only and does not access memory.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4396
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4397<h5>Arguments:</h5>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4398<p>The first argument is always a pointer, and forms the basis of the
Chris Lattnerc8eef442009-07-29 06:44:13 +0000[diff] [blame]4399 calculation. The remaining arguments are indices that indicate which of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4400 elements of the aggregate object are indexed. The interpretation of each
4401 index is dependent on the type being indexed into. The first index always
4402 indexes the pointer value given as the first argument, the second index
4403 indexes a value of the type pointed to (not necessarily the value directly
4404 pointed to, since the first index can be non-zero), etc. The first type
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4405 indexed into must be a pointer value, subsequent types can be arrays,
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]4406 vectors, and structs. Note that subsequent types being indexed into
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4407 can never be pointers, since that would require loading the pointer before
4408 continuing calculation.</p>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4409
4410<p>The type of each index argument depends on the type it is indexing into.
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]4411 When indexing into a (optionally packed) structure, only <tt>i32</tt>
Chris Lattnerfdfeb692010-02-12 20:49:41 +0000[diff] [blame]4412 integer <b>constants</b> are allowed. When indexing into an array, pointer
4413 or vector, integers of any width are allowed, and they are not required to be
Chris Lattnerc8eef442009-07-29 06:44:13 +0000[diff] [blame]4414 constant.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4415
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4416<p>For example, let's consider a C code fragment and how it gets compiled to
4417 LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4418
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]4419<pre class="doc_code">
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4420struct RT {
4421 char A;
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]4422 int B[10][20];
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4423 char C;
4424};
4425struct ST {
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]4426 int X;
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4427 double Y;
4428 struct RT Z;
4429};
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4430
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]4431int *foo(struct ST *s) {
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4432 return &amp;s[1].Z.B[5][13];
4433}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4434</pre>
4435
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4436<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4437
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]4438<pre class="doc_code">
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]4439%RT = <a href="#namedtypes">type</a> { i8 , [10 x [20 x i32]], i8 }
4440%ST = <a href="#namedtypes">type</a> { i32, double, %RT }
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4441
Dan Gohman4df605b2009-07-25 02:23:48 +0000[diff] [blame]4442define i32* @foo(%ST* %s) {
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4443entry:
4444 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
4445 ret i32* %reg
4446}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4447</pre>
4448
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4449<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4450<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4451 type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
4452 }</tt>' type, a structure. The second index indexes into the third element
4453 of the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
4454 i8 }</tt>' type, another structure. The third index indexes into the second
4455 element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
4456 array. The two dimensions of the array are subscripted into, yielding an
4457 '<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a
4458 pointer to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4459
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4460<p>Note that it is perfectly legal to index partially through a structure,
4461 returning a pointer to an inner element. Because of this, the LLVM code for
4462 the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4463
4464<pre>
Dan Gohman4df605b2009-07-25 02:23:48 +0000[diff] [blame]4465 define i32* @foo(%ST* %s) {
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4466 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]4467 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
4468 %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]4469 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
4470 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
4471 ret i32* %t5
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4472 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4473</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]4474
Dan Gohmandd8004d2009-07-27 21:53:46 +0000[diff] [blame]4475<p>If the <tt>inbounds</tt> keyword is present, the result value of the
Dan Gohman27ef9972010-04-23 15:23:32 +0000[diff] [blame]4476 <tt>getelementptr</tt> is a <a href="#trapvalues">trap value</a> if the
4477 base pointer is not an <i>in bounds</i> address of an allocated object,
4478 or if any of the addresses that would be formed by successive addition of
4479 the offsets implied by the indices to the base address with infinitely
4480 precise arithmetic are not an <i>in bounds</i> address of that allocated
4481 object. The <i>in bounds</i> addresses for an allocated object are all
4482 the addresses that point into the object, plus the address one byte past
4483 the end.</p>
Dan Gohmandd8004d2009-07-27 21:53:46 +0000[diff] [blame]4484
4485<p>If the <tt>inbounds</tt> keyword is not present, the offsets are added to
4486 the base address with silently-wrapping two's complement arithmetic, and
4487 the result value of the <tt>getelementptr</tt> may be outside the object
4488 pointed to by the base pointer. The result value may not necessarily be
4489 used to access memory though, even if it happens to point into allocated
4490 storage. See the <a href="#pointeraliasing">Pointer Aliasing Rules</a>
4491 section for more information.</p>
4492
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4493<p>The getelementptr instruction is often confusing. For some more insight into
4494 how it works, see <a href="GetElementPtr.html">the getelementptr FAQ</a>.</p>
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]4495
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4496<h5>Example:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4497<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4498 <i>; yields [12 x i8]*:aptr</i>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4499 %aptr = getelementptr {i32, [12 x i8]}* %saptr, i64 0, i32 1
4500 <i>; yields i8*:vptr</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4501 %vptr = getelementptr {i32, &lt;2 x i8&gt;}* %svptr, i64 0, i32 1, i32 1
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4502 <i>; yields i8*:eptr</i>
4503 %eptr = getelementptr [12 x i8]* %aptr, i64 0, i32 1
Sanjiv Gupta9f805c22009-04-25 07:27:44 +0000[diff] [blame]4504 <i>; yields i32*:iptr</i>
Sanjiv Gupta16ffa802009-04-24 16:38:13 +0000[diff] [blame]4505 %iptr = getelementptr [10 x i32]* @arr, i16 0, i16 0
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4506</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4507
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4508</div>
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]4509
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4510<!-- ======================================================================= -->
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4511<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4512</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4513
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4514<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4515
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4516<p>The instructions in this category are the conversion instructions (casting)
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4517 which all take a single operand and a type. They perform various bit
4518 conversions on the operand.</p>
4519
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4520</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4521
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4522<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4523<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4524 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
4525</div>
4526<div class="doc_text">
4527
4528<h5>Syntax:</h5>
4529<pre>
4530 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4531</pre>
4532
4533<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4534<p>The '<tt>trunc</tt>' instruction truncates its operand to the
4535 type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4536
4537<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4538<p>The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
4539 be an <a href="#t_integer">integer</a> type, and a type that specifies the
4540 size and type of the result, which must be
4541 an <a href="#t_integer">integer</a> type. The bit size of <tt>value</tt> must
4542 be larger than the bit size of <tt>ty2</tt>. Equal sized types are not
4543 allowed.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4544
4545<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4546<p>The '<tt>trunc</tt>' instruction truncates the high order bits
4547 in <tt>value</tt> and converts the remaining bits to <tt>ty2</tt>. Since the
4548 source size must be larger than the destination size, <tt>trunc</tt> cannot
4549 be a <i>no-op cast</i>. It will always truncate bits.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4550
4551<h5>Example:</h5>
4552<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4553 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4554 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4555 %Z = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4556</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4557
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4558</div>
4559
4560<!-- _______________________________________________________________________ -->
4561<div class="doc_subsubsection">
4562 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
4563</div>
4564<div class="doc_text">
4565
4566<h5>Syntax:</h5>
4567<pre>
4568 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4569</pre>
4570
4571<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4572<p>The '<tt>zext</tt>' instruction zero extends its operand to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4573 <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4574
4575
4576<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4577<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4578 <a href="#t_integer">integer</a> type, and a type to cast it to, which must
4579 also be of <a href="#t_integer">integer</a> type. The bit size of the
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4580 <tt>value</tt> must be smaller than the bit size of the destination type,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4581 <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4582
4583<h5>Semantics:</h5>
4584<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4585 bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4586
Reid Spencerb5929522007-01-12 15:46:11 +0000[diff] [blame]4587<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]4588
4589<h5>Example:</h5>
4590<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4591 %X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4592 %Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4593</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4594
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4595</div>
4596
4597<!-- _______________________________________________________________________ -->
4598<div class="doc_subsubsection">
4599 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
4600</div>
4601<div class="doc_text">
4602
4603<h5>Syntax:</h5>
4604<pre>
4605 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4606</pre>
4607
4608<h5>Overview:</h5>
4609<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
4610
4611<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4612<p>The '<tt>sext</tt>' instruction takes a value to cast, which must be of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4613 <a href="#t_integer">integer</a> type, and a type to cast it to, which must
4614 also be of <a href="#t_integer">integer</a> type. The bit size of the
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4615 <tt>value</tt> must be smaller than the bit size of the destination type,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4616 <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4617
4618<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4619<p>The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
4620 bit (highest order bit) of the <tt>value</tt> until it reaches the bit size
4621 of the type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4622
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4623<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]4624
4625<h5>Example:</h5>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4626<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4627 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4628 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4629</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4630
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4631</div>
4632
4633<!-- _______________________________________________________________________ -->
4634<div class="doc_subsubsection">
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4635 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
4636</div>
4637
4638<div class="doc_text">
4639
4640<h5>Syntax:</h5>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4641<pre>
4642 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4643</pre>
4644
4645<h5>Overview:</h5>
4646<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4647 <tt>ty2</tt>.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4648
4649<h5>Arguments:</h5>
4650<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4651 point</a> value to cast and a <a href="#t_floating">floating point</a> type
4652 to cast it to. The size of <tt>value</tt> must be larger than the size of
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4653 <tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4654 <i>no-op cast</i>.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4655
4656<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4657<p>The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4658 <a href="#t_floating">floating point</a> type to a smaller
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4659 <a href="#t_floating">floating point</a> type. If the value cannot fit
4660 within the destination type, <tt>ty2</tt>, then the results are
4661 undefined.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4662
4663<h5>Example:</h5>
4664<pre>
4665 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
4666 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
4667</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4668
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4669</div>
4670
4671<!-- _______________________________________________________________________ -->
4672<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4673 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
4674</div>
4675<div class="doc_text">
4676
4677<h5>Syntax:</h5>
4678<pre>
4679 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4680</pre>
4681
4682<h5>Overview:</h5>
4683<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4684 floating point value.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4685
4686<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4687<p>The '<tt>fpext</tt>' instruction takes a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4688 <a href="#t_floating">floating point</a> <tt>value</tt> to cast, and
4689 a <a href="#t_floating">floating point</a> type to cast it to. The source
4690 type must be smaller than the destination type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4691
4692<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4693<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4694 <a href="#t_floating">floating point</a> type to a larger
4695 <a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
4696 used to make a <i>no-op cast</i> because it always changes bits. Use
4697 <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]4698
4699<h5>Example:</h5>
4700<pre>
4701 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
4702 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
4703</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4704
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4705</div>
4706
4707<!-- _______________________________________________________________________ -->
4708<div class="doc_subsubsection">
Reid Spencer24d6da52007-01-21 00:29:26 +0000[diff] [blame]4709 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4710</div>
4711<div class="doc_text">
4712
4713<h5>Syntax:</h5>
4714<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]4715 &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]4716</pre>
4717
4718<h5>Overview:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]4719<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4720 unsigned integer equivalent of type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4721
4722<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4723<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
4724 scalar or vector <a href="#t_floating">floating point</a> value, and a type
4725 to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
4726 type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
4727 vector integer type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4728
4729<h5>Semantics:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4730<p>The '<tt>fptoui</tt>' instruction converts its
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4731 <a href="#t_floating">floating point</a> operand into the nearest (rounding
4732 towards zero) unsigned integer value. If the value cannot fit
4733 in <tt>ty2</tt>, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4734
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4735<h5>Example:</h5>
4736<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]4737 %X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]4738 %Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4739 %Z = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4740</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4741
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4742</div>
4743
4744<!-- _______________________________________________________________________ -->
4745<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4746 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4747</div>
4748<div class="doc_text">
4749
4750<h5>Syntax:</h5>
4751<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4752 &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]4753</pre>
4754
4755<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4756<p>The '<tt>fptosi</tt>' instruction converts
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4757 <a href="#t_floating">floating point</a> <tt>value</tt> to
4758 type <tt>ty2</tt>.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4759
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4760<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4761<p>The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
4762 scalar or vector <a href="#t_floating">floating point</a> value, and a type
4763 to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
4764 type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
4765 vector integer type with the same number of elements as <tt>ty</tt></p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4766
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4767<h5>Semantics:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4768<p>The '<tt>fptosi</tt>' instruction converts its
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4769 <a href="#t_floating">floating point</a> operand into the nearest (rounding
4770 towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
4771 the results are undefined.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4772
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]4773<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4774<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4775 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]4776 %Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4777 %Z = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4778</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4779
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4780</div>
4781
4782<!-- _______________________________________________________________________ -->
4783<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4784 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4785</div>
4786<div class="doc_text">
4787
4788<h5>Syntax:</h5>
4789<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4790 &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]4791</pre>
4792
4793<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4794<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4795 integer and converts that value to the <tt>ty2</tt> type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4796
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4797<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]4798<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4799 scalar or vector <a href="#t_integer">integer</a> value, and a type to cast
4800 it to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
4801 type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
4802 floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4803
4804<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4805<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4806 integer quantity and converts it to the corresponding floating point
4807 value. If the value cannot fit in the floating point value, the results are
4808 undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4809
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4810<h5>Example:</h5>
4811<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4812 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4813 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4814</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4815
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4816</div>
4817
4818<!-- _______________________________________________________________________ -->
4819<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4820 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4821</div>
4822<div class="doc_text">
4823
4824<h5>Syntax:</h5>
4825<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4826 &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]4827</pre>
4828
4829<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4830<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed integer
4831 and converts that value to the <tt>ty2</tt> type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4832
4833<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]4834<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4835 scalar or vector <a href="#t_integer">integer</a> value, and a type to cast
4836 it to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
4837 type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
4838 floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4839
4840<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4841<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed integer
4842 quantity and converts it to the corresponding floating point value. If the
4843 value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4844
4845<h5>Example:</h5>
4846<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4847 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4848 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4849</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4850
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4851</div>
4852
4853<!-- _______________________________________________________________________ -->
4854<div class="doc_subsubsection">
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4855 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
4856</div>
4857<div class="doc_text">
4858
4859<h5>Syntax:</h5>
4860<pre>
4861 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4862</pre>
4863
4864<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4865<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
4866 the integer type <tt>ty2</tt>.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4867
4868<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4869<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
4870 must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
4871 <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a> type.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4872
4873<h5>Semantics:</h5>
4874<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4875 <tt>ty2</tt> by interpreting the pointer value as an integer and either
4876 truncating or zero extending that value to the size of the integer type. If
4877 <tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
4878 <tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
4879 are the same size, then nothing is done (<i>no-op cast</i>) other than a type
4880 change.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4881
4882<h5>Example:</h5>
4883<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4884 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
4885 %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]4886</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4887
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4888</div>
4889
4890<!-- _______________________________________________________________________ -->
4891<div class="doc_subsubsection">
4892 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
4893</div>
4894<div class="doc_text">
4895
4896<h5>Syntax:</h5>
4897<pre>
4898 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4899</pre>
4900
4901<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4902<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to a
4903 pointer type, <tt>ty2</tt>.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4904
4905<h5>Arguments:</h5>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]4906<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4907 value to cast, and a type to cast it to, which must be a
4908 <a href="#t_pointer">pointer</a> type.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4909
4910<h5>Semantics:</h5>
4911<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4912 <tt>ty2</tt> by applying either a zero extension or a truncation depending on
4913 the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
4914 size of a pointer then a truncation is done. If <tt>value</tt> is smaller
4915 than the size of a pointer then a zero extension is done. If they are the
4916 same size, nothing is done (<i>no-op cast</i>).</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4917
4918<h5>Example:</h5>
4919<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4920 %X = inttoptr i32 255 to i32* <i>; yields zero extension on 64-bit architecture</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4921 %Y = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
4922 %Z = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4923</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4924
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4925</div>
4926
4927<!-- _______________________________________________________________________ -->
4928<div class="doc_subsubsection">
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4929 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4930</div>
4931<div class="doc_text">
4932
4933<h5>Syntax:</h5>
4934<pre>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4935 &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]4936</pre>
4937
4938<h5>Overview:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4939<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4940 <tt>ty2</tt> without changing any bits.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4941
4942<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4943<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be a
4944 non-aggregate first class value, and a type to cast it to, which must also be
4945 a non-aggregate <a href="#t_firstclass">first class</a> type. The bit sizes
4946 of <tt>value</tt> and the destination type, <tt>ty2</tt>, must be
4947 identical. If the source type is a pointer, the destination type must also be
4948 a pointer. This instruction supports bitwise conversion of vectors to
4949 integers and to vectors of other types (as long as they have the same
4950 size).</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4951
4952<h5>Semantics:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4953<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4954 <tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
4955 this conversion. The conversion is done as if the <tt>value</tt> had been
4956 stored to memory and read back as type <tt>ty2</tt>. Pointer types may only
4957 be converted to other pointer types with this instruction. To convert
4958 pointers to other types, use the <a href="#i_inttoptr">inttoptr</a> or
4959 <a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4960
4961<h5>Example:</h5>
4962<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4963 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4964 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4965 %Z = bitcast &lt;2 x int&gt; %V to i64; <i>; yields i64: %V</i>
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]4966</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4967
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4968</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4969
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4970<!-- ======================================================================= -->
4971<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4972
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4973<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4974
4975<p>The instructions in this category are the "miscellaneous" instructions, which
4976 defy better classification.</p>
4977
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4978</div>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4979
4980<!-- _______________________________________________________________________ -->
4981<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
4982</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4983
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4984<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4985
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4986<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4987<pre>
4988 &lt;result&gt; = icmp &lt;cond&gt; &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {i1} or {&lt;N x i1&gt;}:result</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4989</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4990
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4991<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4992<p>The '<tt>icmp</tt>' instruction returns a boolean value or a vector of
4993 boolean values based on comparison of its two integer, integer vector, or
4994 pointer operands.</p>
4995
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4996<h5>Arguments:</h5>
4997<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4998 the condition code indicating the kind of comparison to perform. It is not a
4999 value, just a keyword. The possible condition code are:</p>
5000
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5001<ol>
5002 <li><tt>eq</tt>: equal</li>
5003 <li><tt>ne</tt>: not equal </li>
5004 <li><tt>ugt</tt>: unsigned greater than</li>
5005 <li><tt>uge</tt>: unsigned greater or equal</li>
5006 <li><tt>ult</tt>: unsigned less than</li>
5007 <li><tt>ule</tt>: unsigned less or equal</li>
5008 <li><tt>sgt</tt>: signed greater than</li>
5009 <li><tt>sge</tt>: signed greater or equal</li>
5010 <li><tt>slt</tt>: signed less than</li>
5011 <li><tt>sle</tt>: signed less or equal</li>
5012</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5013
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]5014<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5015 <a href="#t_pointer">pointer</a> or integer <a href="#t_vector">vector</a>
5016 typed. They must also be identical types.</p>
5017
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5018<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5019<p>The '<tt>icmp</tt>' compares <tt>op1</tt> and <tt>op2</tt> according to the
5020 condition code given as <tt>cond</tt>. The comparison performed always yields
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]5021 either an <a href="#t_integer"><tt>i1</tt></a> or vector of <tt>i1</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5022 result, as follows:</p>
5023
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5024<ol>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5025 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5026 <tt>false</tt> otherwise. No sign interpretation is necessary or
5027 performed.</li>
5028
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5029 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5030 <tt>false</tt> otherwise. No sign interpretation is necessary or
5031 performed.</li>
5032
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5033 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5034 <tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
5035
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5036 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5037 <tt>true</tt> if <tt>op1</tt> is greater than or equal
5038 to <tt>op2</tt>.</li>
5039
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5040 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5041 <tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
5042
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5043 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5044 <tt>true</tt> if <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
5045
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5046 <li><tt>sgt</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5047 <tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
5048
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5049 <li><tt>sge</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5050 <tt>true</tt> if <tt>op1</tt> is greater than or equal
5051 to <tt>op2</tt>.</li>
5052
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5053 <li><tt>slt</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5054 <tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
5055
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5056 <li><tt>sle</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5057 <tt>true</tt> if <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5058</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5059
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5060<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5061 values are compared as if they were integers.</p>
5062
5063<p>If the operands are integer vectors, then they are compared element by
5064 element. The result is an <tt>i1</tt> vector with the same number of elements
5065 as the values being compared. Otherwise, the result is an <tt>i1</tt>.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5066
5067<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5068<pre>
5069 &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5070 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
5071 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
5072 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
5073 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
5074 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5075</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]5076
5077<p>Note that the code generator does not yet support vector types with
5078 the <tt>icmp</tt> instruction.</p>
5079
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5080</div>
5081
5082<!-- _______________________________________________________________________ -->
5083<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
5084</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5085
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5086<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5087
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5088<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5089<pre>
5090 &lt;result&gt; = fcmp &lt;cond&gt; &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {i1} or {&lt;N x i1&gt;}:result</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5091</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5092
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5093<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5094<p>The '<tt>fcmp</tt>' instruction returns a boolean value or vector of boolean
5095 values based on comparison of its operands.</p>
5096
5097<p>If the operands are floating point scalars, then the result type is a boolean
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]5098(<a href="#t_integer"><tt>i1</tt></a>).</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5099
5100<p>If the operands are floating point vectors, then the result type is a vector
5101 of boolean with the same number of elements as the operands being
5102 compared.</p>
5103
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5104<h5>Arguments:</h5>
5105<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5106 the condition code indicating the kind of comparison to perform. It is not a
5107 value, just a keyword. The possible condition code are:</p>
5108
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5109<ol>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]5110 <li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5111 <li><tt>oeq</tt>: ordered and equal</li>
5112 <li><tt>ogt</tt>: ordered and greater than </li>
5113 <li><tt>oge</tt>: ordered and greater than or equal</li>
5114 <li><tt>olt</tt>: ordered and less than </li>
5115 <li><tt>ole</tt>: ordered and less than or equal</li>
5116 <li><tt>one</tt>: ordered and not equal</li>
5117 <li><tt>ord</tt>: ordered (no nans)</li>
5118 <li><tt>ueq</tt>: unordered or equal</li>
5119 <li><tt>ugt</tt>: unordered or greater than </li>
5120 <li><tt>uge</tt>: unordered or greater than or equal</li>
5121 <li><tt>ult</tt>: unordered or less than </li>
5122 <li><tt>ule</tt>: unordered or less than or equal</li>
5123 <li><tt>une</tt>: unordered or not equal</li>
5124 <li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]5125 <li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5126</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5127
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5128<p><i>Ordered</i> means that neither operand is a QNAN while
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5129 <i>unordered</i> means that either operand may be a QNAN.</p>
5130
5131<p>Each of <tt>val1</tt> and <tt>val2</tt> arguments must be either
5132 a <a href="#t_floating">floating point</a> type or
5133 a <a href="#t_vector">vector</a> of floating point type. They must have
5134 identical types.</p>
5135
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5136<h5>Semantics:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]5137<p>The '<tt>fcmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5138 according to the condition code given as <tt>cond</tt>. If the operands are
5139 vectors, then the vectors are compared element by element. Each comparison
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]5140 performed always yields an <a href="#t_integer">i1</a> result, as
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5141 follows:</p>
5142
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5143<ol>
5144 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5145
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5146 <li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5147 <tt>op1</tt> is equal to <tt>op2</tt>.</li>
5148
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]5149 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]5150 <tt>op1</tt> is greater than <tt>op2</tt>.</li>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5151
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5152 <li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5153 <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
5154
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5155 <li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5156 <tt>op1</tt> is less than <tt>op2</tt>.</li>
5157
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5158 <li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5159 <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
5160
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5161 <li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5162 <tt>op1</tt> is not equal to <tt>op2</tt>.</li>
5163
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]5164 <li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5165
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5166 <li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5167 <tt>op1</tt> is equal to <tt>op2</tt>.</li>
5168
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5169 <li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5170 <tt>op1</tt> is greater than <tt>op2</tt>.</li>
5171
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5172 <li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5173 <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
5174
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5175 <li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5176 <tt>op1</tt> is less than <tt>op2</tt>.</li>
5177
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5178 <li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5179 <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
5180
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5181 <li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5182 <tt>op1</tt> is not equal to <tt>op2</tt>.</li>
5183
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]5184 <li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5185
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5186 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
5187</ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5188
5189<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5190<pre>
5191 &lt;result&gt; = fcmp oeq float 4.0, 5.0 <i>; yields: result=false</i>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]5192 &lt;result&gt; = fcmp one float 4.0, 5.0 <i>; yields: result=true</i>
5193 &lt;result&gt; = fcmp olt float 4.0, 5.0 <i>; yields: result=true</i>
5194 &lt;result&gt; = fcmp ueq double 1.0, 2.0 <i>; yields: result=false</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5195</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]5196
5197<p>Note that the code generator does not yet support vector types with
5198 the <tt>fcmp</tt> instruction.</p>
5199
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]5200</div>
5201
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5202<!-- _______________________________________________________________________ -->
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]5203<div class="doc_subsubsection">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5204 <a name="i_phi">'<tt>phi</tt>' Instruction</a>
5205</div>
5206
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5207<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5208
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5209<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5210<pre>
5211 &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...
5212</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5213
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5214<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5215<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in the
5216 SSA graph representing the function.</p>
5217
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5218<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5219<p>The type of the incoming values is specified with the first type field. After
5220 this, the '<tt>phi</tt>' instruction takes a list of pairs as arguments, with
5221 one pair for each predecessor basic block of the current block. Only values
5222 of <a href="#t_firstclass">first class</a> type may be used as the value
5223 arguments to the PHI node. Only labels may be used as the label
5224 arguments.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5225
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5226<p>There must be no non-phi instructions between the start of a basic block and
5227 the PHI instructions: i.e. PHI instructions must be first in a basic
5228 block.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5229
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5230<p>For the purposes of the SSA form, the use of each incoming value is deemed to
5231 occur on the edge from the corresponding predecessor block to the current
5232 block (but after any definition of an '<tt>invoke</tt>' instruction's return
5233 value on the same edge).</p>
Jay Foadd2449092009-06-03 10:20:10 +0000[diff] [blame]5234
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5235<h5>Semantics:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5236<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5237 specified by the pair corresponding to the predecessor basic block that
5238 executed just prior to the current block.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5239
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5240<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]5241<pre>
5242Loop: ; Infinite loop that counts from 0 on up...
5243 %indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]
5244 %nextindvar = add i32 %indvar, 1
5245 br label %Loop
5246</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5247
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]5248</div>
5249
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5250<!-- _______________________________________________________________________ -->
5251<div class="doc_subsubsection">
5252 <a name="i_select">'<tt>select</tt>' Instruction</a>
5253</div>
5254
5255<div class="doc_text">
5256
5257<h5>Syntax:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5258<pre>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]5259 &lt;result&gt; = select <i>selty</i> &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
5260
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5261 <i>selty</i> is either i1 or {&lt;N x i1&gt;}
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5262</pre>
5263
5264<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5265<p>The '<tt>select</tt>' instruction is used to choose one value based on a
5266 condition, without branching.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5267
5268
5269<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5270<p>The '<tt>select</tt>' instruction requires an 'i1' value or a vector of 'i1'
5271 values indicating the condition, and two values of the
5272 same <a href="#t_firstclass">first class</a> type. If the val1/val2 are
5273 vectors and the condition is a scalar, then entire vectors are selected, not
5274 individual elements.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5275
5276<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5277<p>If the condition is an i1 and it evaluates to 1, the instruction returns the
5278 first value argument; otherwise, it returns the second value argument.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5279
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5280<p>If the condition is a vector of i1, then the value arguments must be vectors
5281 of the same size, and the selection is done element by element.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5282
5283<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5284<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]5285 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5286</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]5287
5288<p>Note that the code generator does not yet support conditions
5289 with vector type.</p>
5290
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5291</div>
5292
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]5293<!-- _______________________________________________________________________ -->
5294<div class="doc_subsubsection">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5295 <a name="i_call">'<tt>call</tt>' Instruction</a>
5296</div>
5297
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5298<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5299
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5300<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5301<pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]5302 &lt;result&gt; = [tail] call [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>] &lt;ty&gt; [&lt;fnty&gt;*] &lt;fnptrval&gt;(&lt;function args&gt;) [<a href="#fnattrs">fn attrs</a>]
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5303</pre>
5304
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5305<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5306<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5307
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5308<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5309<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5310
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]5311<ol>
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]5312 <li>The optional "tail" marker indicates that the callee function does not
5313 access any allocas or varargs in the caller. Note that calls may be
5314 marked "tail" even if they do not occur before
5315 a <a href="#i_ret"><tt>ret</tt></a> instruction. If the "tail" marker is
5316 present, the function call is eligible for tail call optimization,
5317 but <a href="CodeGenerator.html#tailcallopt">might not in fact be
Evan Chengdc444e92010-03-08 21:05:02 +0000[diff] [blame]5318 optimized into a jump</a>. The code generator may optimize calls marked
5319 "tail" with either 1) automatic <a href="CodeGenerator.html#sibcallopt">
5320 sibling call optimization</a> when the caller and callee have
5321 matching signatures, or 2) forced tail call optimization when the
5322 following extra requirements are met:
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]5323 <ul>
5324 <li>Caller and callee both have the calling
5325 convention <tt>fastcc</tt>.</li>
5326 <li>The call is in tail position (ret immediately follows call and ret
5327 uses value of call or is void).</li>
5328 <li>Option <tt>-tailcallopt</tt> is enabled,
Dan Gohmanfbbee8d2010-03-02 01:08:11 +0000[diff] [blame]5329 or <code>llvm::GuaranteedTailCallOpt</code> is <code>true</code>.</li>
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]5330 <li><a href="CodeGenerator.html#tailcallopt">Platform specific
5331 constraints are met.</a></li>
5332 </ul>
5333 </li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]5334
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5335 <li>The optional "cconv" marker indicates which <a href="#callingconv">calling
5336 convention</a> the call should use. If none is specified, the call
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]5337 defaults to using C calling conventions. The calling convention of the
5338 call must match the calling convention of the target function, or else the
5339 behavior is undefined.</li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]5340
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5341 <li>The optional <a href="#paramattrs">Parameter Attributes</a> list for
5342 return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>', and
5343 '<tt>inreg</tt>' attributes are valid here.</li>
5344
5345 <li>'<tt>ty</tt>': the type of the call instruction itself which is also the
5346 type of the return value. Functions that return no value are marked
5347 <tt><a href="#t_void">void</a></tt>.</li>
5348
5349 <li>'<tt>fnty</tt>': shall be the signature of the pointer to function value
5350 being invoked. The argument types must match the types implied by this
5351 signature. This type can be omitted if the function is not varargs and if
5352 the function type does not return a pointer to a function.</li>
5353
5354 <li>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
5355 be invoked. In most cases, this is a direct function invocation, but
5356 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
5357 to function value.</li>
5358
5359 <li>'<tt>function args</tt>': argument list whose types match the function
Chris Lattner0724fbd2010-03-02 06:36:51 +0000[diff] [blame]5360 signature argument types and parameter attributes. All arguments must be
5361 of <a href="#t_firstclass">first class</a> type. If the function
5362 signature indicates the function accepts a variable number of arguments,
5363 the extra arguments can be specified.</li>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5364
5365 <li>The optional <a href="#fnattrs">function attributes</a> list. Only
5366 '<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
5367 '<tt>readnone</tt>' attributes are valid here.</li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]5368</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5369
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5370<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5371<p>The '<tt>call</tt>' instruction is used to cause control flow to transfer to
5372 a specified function, with its incoming arguments bound to the specified
5373 values. Upon a '<tt><a href="#i_ret">ret</a></tt>' instruction in the called
5374 function, control flow continues with the instruction after the function
5375 call, and the return value of the function is bound to the result
5376 argument.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5377
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5378<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5379<pre>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]5380 %retval = call i32 @test(i32 %argc)
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]5381 call i32 (i8*, ...)* @printf(i8* %msg, i32 12, i8 42) <i>; yields i32</i>
Chris Lattner772fccf2008-03-21 17:24:17 +0000[diff] [blame]5382 %X = tail call i32 @foo() <i>; yields i32</i>
5383 %Y = tail call <a href="#callingconv">fastcc</a> i32 @foo() <i>; yields i32</i>
5384 call void %foo(i8 97 signext)
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]5385
5386 %struct.A = type { i32, i8 }
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]5387 %r = call %struct.A @foo() <i>; yields { 32, i8 }</i>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]5388 %gr = extractvalue %struct.A %r, 0 <i>; yields i32</i>
5389 %gr1 = extractvalue %struct.A %r, 1 <i>; yields i8</i>
Chris Lattner85a350f2008-10-08 06:26:11 +0000[diff] [blame]5390 %Z = call void @foo() noreturn <i>; indicates that %foo never returns normally</i>
Matthijs Kooijmancb73d192008-10-07 10:03:45 +0000[diff] [blame]5391 %ZZ = call zeroext i32 @bar() <i>; Return value is %zero extended</i>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5392</pre>
5393
Dale Johannesen07de8d12009-09-24 18:38:21 +0000[diff] [blame]5394<p>llvm treats calls to some functions with names and arguments that match the
Dale Johannesen9f8380b2009-09-25 17:04:42 +0000[diff] [blame]5395standard C99 library as being the C99 library functions, and may perform
5396optimizations or generate code for them under that assumption. This is
5397something we'd like to change in the future to provide better support for
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]5398freestanding environments and non-C-based languages.</p>
Dale Johannesen07de8d12009-09-24 18:38:21 +0000[diff] [blame]5399
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5400</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5401
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5402<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5403<div class="doc_subsubsection">
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]5404 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5405</div>
5406
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5407<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5408
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5409<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5410<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5411 &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]5412</pre>
5413
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5414<h5>Overview:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5415<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5416 the "variable argument" area of a function call. It is used to implement the
5417 <tt>va_arg</tt> macro in C.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5418
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5419<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5420<p>This instruction takes a <tt>va_list*</tt> value and the type of the
5421 argument. It returns a value of the specified argument type and increments
5422 the <tt>va_list</tt> to point to the next argument. The actual type
5423 of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5424
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5425<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5426<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified type
5427 from the specified <tt>va_list</tt> and causes the <tt>va_list</tt> to point
5428 to the next argument. For more information, see the variable argument
5429 handling <a href="#int_varargs">Intrinsic Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5430
5431<p>It is legal for this instruction to be called in a function which does not
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5432 take a variable number of arguments, for example, the <tt>vfprintf</tt>
5433 function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5434
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5435<p><tt>va_arg</tt> is an LLVM instruction instead of
5436 an <a href="#intrinsics">intrinsic function</a> because it takes a type as an
5437 argument.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5438
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5439<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5440<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
5441
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5442<p>Note that the code generator does not yet fully support va_arg on many
5443 targets. Also, it does not currently support va_arg with aggregate types on
5444 any target.</p>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]5445
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5446</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5447
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5448<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]5449<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
5450<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5451
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5452<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5453
5454<p>LLVM supports the notion of an "intrinsic function". These functions have
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5455 well known names and semantics and are required to follow certain
5456 restrictions. Overall, these intrinsics represent an extension mechanism for
5457 the LLVM language that does not require changing all of the transformations
5458 in LLVM when adding to the language (or the bitcode reader/writer, the
5459 parser, etc...).</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5460
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]5461<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5462 prefix is reserved in LLVM for intrinsic names; thus, function names may not
5463 begin with this prefix. Intrinsic functions must always be external
5464 functions: you cannot define the body of intrinsic functions. Intrinsic
5465 functions may only be used in call or invoke instructions: it is illegal to
5466 take the address of an intrinsic function. Additionally, because intrinsic
5467 functions are part of the LLVM language, it is required if any are added that
5468 they be documented here.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5469
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5470<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents a
5471 family of functions that perform the same operation but on different data
5472 types. Because LLVM can represent over 8 million different integer types,
5473 overloading is used commonly to allow an intrinsic function to operate on any
5474 integer type. One or more of the argument types or the result type can be
5475 overloaded to accept any integer type. Argument types may also be defined as
5476 exactly matching a previous argument's type or the result type. This allows
5477 an intrinsic function which accepts multiple arguments, but needs all of them
5478 to be of the same type, to only be overloaded with respect to a single
5479 argument or the result.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5480
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5481<p>Overloaded intrinsics will have the names of its overloaded argument types
5482 encoded into its function name, each preceded by a period. Only those types
5483 which are overloaded result in a name suffix. Arguments whose type is matched
5484 against another type do not. For example, the <tt>llvm.ctpop</tt> function
5485 can take an integer of any width and returns an integer of exactly the same
5486 integer width. This leads to a family of functions such as
5487 <tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29
5488 %val)</tt>. Only one type, the return type, is overloaded, and only one type
5489 suffix is required. Because the argument's type is matched against the return
5490 type, it does not require its own name suffix.</p>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5491
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5492<p>To learn how to add an intrinsic function, please see the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5493 <a href="ExtendingLLVM.html">Extending LLVM Guide</a>.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5494
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5495</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5496
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5497<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5498<div class="doc_subsection">
5499 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
5500</div>
5501
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5502<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5503
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5504<p>Variable argument support is defined in LLVM with
5505 the <a href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
5506 intrinsic functions. These functions are related to the similarly named
5507 macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5508
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5509<p>All of these functions operate on arguments that use a target-specific value
5510 type "<tt>va_list</tt>". The LLVM assembly language reference manual does
5511 not define what this type is, so all transformations should be prepared to
5512 handle these functions regardless of the type used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5513
Chris Lattner374ab302006-05-15 17:26:46 +0000[diff] [blame]5514<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5515 instruction and the variable argument handling intrinsic functions are
5516 used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5517
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]5518<pre class="doc_code">
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5519define i32 @test(i32 %X, ...) {
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5520 ; Initialize variable argument processing
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5521 %ap = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5522 %ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5523 call void @llvm.va_start(i8* %ap2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5524
5525 ; Read a single integer argument
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5526 %tmp = va_arg i8** %ap, i32
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5527
5528 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5529 %aq = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5530 %aq2 = bitcast i8** %aq to i8*
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5531 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5532 call void @llvm.va_end(i8* %aq2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5533
5534 ; Stop processing of arguments.
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5535 call void @llvm.va_end(i8* %ap2)
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5536 ret i32 %tmp
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5537}
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5538
5539declare void @llvm.va_start(i8*)
5540declare void @llvm.va_copy(i8*, i8*)
5541declare void @llvm.va_end(i8*)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5542</pre>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5543
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]5544</div>
5545
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5546<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5547<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5548 <a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5549</div>
5550
5551
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5552<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5553
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5554<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5555<pre>
5556 declare void %llvm.va_start(i8* &lt;arglist&gt;)
5557</pre>
5558
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5559<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5560<p>The '<tt>llvm.va_start</tt>' intrinsic initializes <tt>*&lt;arglist&gt;</tt>
5561 for subsequent use by <tt><a href="#i_va_arg">va_arg</a></tt>.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5562
5563<h5>Arguments:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5564<p>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5565
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5566<h5>Semantics:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5567<p>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5568 macro available in C. In a target-dependent way, it initializes
5569 the <tt>va_list</tt> element to which the argument points, so that the next
5570 call to <tt>va_arg</tt> will produce the first variable argument passed to
5571 the function. Unlike the C <tt>va_start</tt> macro, this intrinsic does not
5572 need to know the last argument of the function as the compiler can figure
5573 that out.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5574
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5575</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5576
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5577<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5578<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5579 <a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5580</div>
5581
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5582<div class="doc_text">
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5583
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5584<h5>Syntax:</h5>
5585<pre>
5586 declare void @llvm.va_end(i8* &lt;arglist&gt;)
5587</pre>
5588
5589<h5>Overview:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5590<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*&lt;arglist&gt;</tt>,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5591 which has been initialized previously
5592 with <tt><a href="#int_va_start">llvm.va_start</a></tt>
5593 or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5594
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5595<h5>Arguments:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5596<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]5597
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5598<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5599<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5600 macro available in C. In a target-dependent way, it destroys
5601 the <tt>va_list</tt> element to which the argument points. Calls
5602 to <a href="#int_va_start"><tt>llvm.va_start</tt></a>
5603 and <a href="#int_va_copy"> <tt>llvm.va_copy</tt></a> must be matched exactly
5604 with calls to <tt>llvm.va_end</tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5605
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5606</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5607
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5608<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5609<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5610 <a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5611</div>
5612
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5613<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5614
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5615<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5616<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5617 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5618</pre>
5619
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5620<h5>Overview:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5621<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5622 from the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5623
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5624<h5>Arguments:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5625<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5626 The second argument is a pointer to a <tt>va_list</tt> element to copy
5627 from.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5628
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5629<h5>Semantics:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5630<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5631 macro available in C. In a target-dependent way, it copies the
5632 source <tt>va_list</tt> element into the destination <tt>va_list</tt>
5633 element. This intrinsic is necessary because
5634 the <tt><a href="#int_va_start"> llvm.va_start</a></tt> intrinsic may be
5635 arbitrarily complex and require, for example, memory allocation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5636
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5637</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5638
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5639<!-- ======================================================================= -->
5640<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5641 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
5642</div>
5643
5644<div class="doc_text">
5645
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5646<p>LLVM support for <a href="GarbageCollection.html">Accurate Garbage
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]5647Collection</a> (GC) requires the implementation and generation of these
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5648intrinsics. These intrinsics allow identification of <a href="#int_gcroot">GC
5649roots on the stack</a>, as well as garbage collector implementations that
5650require <a href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a>
5651barriers. Front-ends for type-safe garbage collected languages should generate
5652these intrinsics to make use of the LLVM garbage collectors. For more details,
5653see <a href="GarbageCollection.html">Accurate Garbage Collection with
5654LLVM</a>.</p>
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]5655
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5656<p>The garbage collection intrinsics only operate on objects in the generic
5657 address space (address space zero).</p>
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]5658
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5659</div>
5660
5661<!-- _______________________________________________________________________ -->
5662<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5663 <a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5664</div>
5665
5666<div class="doc_text">
5667
5668<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5669<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5670 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5671</pre>
5672
5673<h5>Overview:</h5>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]5674<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5675 the code generator, and allows some metadata to be associated with it.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5676
5677<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5678<p>The first argument specifies the address of a stack object that contains the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5679 root pointer. The second pointer (which must be either a constant or a
5680 global value address) contains the meta-data to be associated with the
5681 root.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5682
5683<h5>Semantics:</h5>
Chris Lattner05d67092008-04-24 05:59:56 +0000[diff] [blame]5684<p>At runtime, a call to this intrinsic stores a null pointer into the "ptrloc"
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5685 location. At compile-time, the code generator generates information to allow
5686 the runtime to find the pointer at GC safe points. The '<tt>llvm.gcroot</tt>'
5687 intrinsic may only be used in a function which <a href="#gc">specifies a GC
5688 algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5689
5690</div>
5691
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5692<!-- _______________________________________________________________________ -->
5693<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5694 <a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5695</div>
5696
5697<div class="doc_text">
5698
5699<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5700<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5701 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5702</pre>
5703
5704<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5705<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5706 locations, allowing garbage collector implementations that require read
5707 barriers.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5708
5709<h5>Arguments:</h5>
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]5710<p>The second argument is the address to read from, which should be an address
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5711 allocated from the garbage collector. The first object is a pointer to the
5712 start of the referenced object, if needed by the language runtime (otherwise
5713 null).</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5714
5715<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5716<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5717 instruction, but may be replaced with substantially more complex code by the
5718 garbage collector runtime, as needed. The '<tt>llvm.gcread</tt>' intrinsic
5719 may only be used in a function which <a href="#gc">specifies a GC
5720 algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5721
5722</div>
5723
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5724<!-- _______________________________________________________________________ -->
5725<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5726 <a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5727</div>
5728
5729<div class="doc_text">
5730
5731<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5732<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5733 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5734</pre>
5735
5736<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5737<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5738 locations, allowing garbage collector implementations that require write
5739 barriers (such as generational or reference counting collectors).</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5740
5741<h5>Arguments:</h5>
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]5742<p>The first argument is the reference to store, the second is the start of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5743 object to store it to, and the third is the address of the field of Obj to
5744 store to. If the runtime does not require a pointer to the object, Obj may
5745 be null.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5746
5747<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5748<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5749 instruction, but may be replaced with substantially more complex code by the
5750 garbage collector runtime, as needed. The '<tt>llvm.gcwrite</tt>' intrinsic
5751 may only be used in a function which <a href="#gc">specifies a GC
5752 algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5753
5754</div>
5755
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5756<!-- ======================================================================= -->
5757<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5758 <a name="int_codegen">Code Generator Intrinsics</a>
5759</div>
5760
5761<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5762
5763<p>These intrinsics are provided by LLVM to expose special features that may
5764 only be implemented with code generator support.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5765
5766</div>
5767
5768<!-- _______________________________________________________________________ -->
5769<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5770 <a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5771</div>
5772
5773<div class="doc_text">
5774
5775<h5>Syntax:</h5>
5776<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5777 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5778</pre>
5779
5780<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5781<p>The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
5782 target-specific value indicating the return address of the current function
5783 or one of its callers.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5784
5785<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5786<p>The argument to this intrinsic indicates which function to return the address
5787 for. Zero indicates the calling function, one indicates its caller, etc.
5788 The argument is <b>required</b> to be a constant integer value.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5789
5790<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5791<p>The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer
5792 indicating the return address of the specified call frame, or zero if it
5793 cannot be identified. The value returned by this intrinsic is likely to be
5794 incorrect or 0 for arguments other than zero, so it should only be used for
5795 debugging purposes.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5796
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5797<p>Note that calling this intrinsic does not prevent function inlining or other
5798 aggressive transformations, so the value returned may not be that of the
5799 obvious source-language caller.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5800
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5801</div>
5802
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5803<!-- _______________________________________________________________________ -->
5804<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5805 <a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5806</div>
5807
5808<div class="doc_text">
5809
5810<h5>Syntax:</h5>
5811<pre>
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]5812 declare i8* @llvm.frameaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5813</pre>
5814
5815<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5816<p>The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
5817 target-specific frame pointer value for the specified stack frame.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5818
5819<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5820<p>The argument to this intrinsic indicates which function to return the frame
5821 pointer for. Zero indicates the calling function, one indicates its caller,
5822 etc. The argument is <b>required</b> to be a constant integer value.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5823
5824<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5825<p>The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer
5826 indicating the frame address of the specified call frame, or zero if it
5827 cannot be identified. The value returned by this intrinsic is likely to be
5828 incorrect or 0 for arguments other than zero, so it should only be used for
5829 debugging purposes.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5830
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5831<p>Note that calling this intrinsic does not prevent function inlining or other
5832 aggressive transformations, so the value returned may not be that of the
5833 obvious source-language caller.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5834
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5835</div>
5836
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5837<!-- _______________________________________________________________________ -->
5838<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5839 <a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5840</div>
5841
5842<div class="doc_text">
5843
5844<h5>Syntax:</h5>
5845<pre>
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]5846 declare i8* @llvm.stacksave()
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5847</pre>
5848
5849<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5850<p>The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state
5851 of the function stack, for use
5852 with <a href="#int_stackrestore"> <tt>llvm.stackrestore</tt></a>. This is
5853 useful for implementing language features like scoped automatic variable
5854 sized arrays in C99.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5855
5856<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5857<p>This intrinsic returns a opaque pointer value that can be passed
5858 to <a href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When
5859 an <tt>llvm.stackrestore</tt> intrinsic is executed with a value saved
5860 from <tt>llvm.stacksave</tt>, it effectively restores the state of the stack
5861 to the state it was in when the <tt>llvm.stacksave</tt> intrinsic executed.
5862 In practice, this pops any <a href="#i_alloca">alloca</a> blocks from the
5863 stack that were allocated after the <tt>llvm.stacksave</tt> was executed.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5864
5865</div>
5866
5867<!-- _______________________________________________________________________ -->
5868<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5869 <a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5870</div>
5871
5872<div class="doc_text">
5873
5874<h5>Syntax:</h5>
5875<pre>
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]5876 declare void @llvm.stackrestore(i8* %ptr)
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5877</pre>
5878
5879<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5880<p>The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
5881 the function stack to the state it was in when the
5882 corresponding <a href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic
5883 executed. This is useful for implementing language features like scoped
5884 automatic variable sized arrays in C99.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5885
5886<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5887<p>See the description
5888 for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5889
5890</div>
5891
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5892<!-- _______________________________________________________________________ -->
5893<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5894 <a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5895</div>
5896
5897<div class="doc_text">
5898
5899<h5>Syntax:</h5>
5900<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5901 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]5902</pre>
5903
5904<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5905<p>The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to
5906 insert a prefetch instruction if supported; otherwise, it is a noop.
5907 Prefetches have no effect on the behavior of the program but can change its
5908 performance characteristics.</p>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5909
5910<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5911<p><tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the
5912 specifier determining if the fetch should be for a read (0) or write (1),
5913 and <tt>locality</tt> is a temporal locality specifier ranging from (0) - no
5914 locality, to (3) - extremely local keep in cache. The <tt>rw</tt>
5915 and <tt>locality</tt> arguments must be constant integers.</p>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5916
5917<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5918<p>This intrinsic does not modify the behavior of the program. In particular,
5919 prefetches cannot trap and do not produce a value. On targets that support
5920 this intrinsic, the prefetch can provide hints to the processor cache for
5921 better performance.</p>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5922
5923</div>
5924
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5925<!-- _______________________________________________________________________ -->
5926<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5927 <a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5928</div>
5929
5930<div class="doc_text">
5931
5932<h5>Syntax:</h5>
5933<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5934 declare void @llvm.pcmarker(i32 &lt;id&gt;)
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5935</pre>
5936
5937<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5938<p>The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program
5939 Counter (PC) in a region of code to simulators and other tools. The method
5940 is target specific, but it is expected that the marker will use exported
5941 symbols to transmit the PC of the marker. The marker makes no guarantees
5942 that it will remain with any specific instruction after optimizations. It is
5943 possible that the presence of a marker will inhibit optimizations. The
5944 intended use is to be inserted after optimizations to allow correlations of
5945 simulation runs.</p>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5946
5947<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5948<p><tt>id</tt> is a numerical id identifying the marker.</p>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5949
5950<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5951<p>This intrinsic does not modify the behavior of the program. Backends that do
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]5952 not support this intrinsic may ignore it.</p>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5953
5954</div>
5955
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5956<!-- _______________________________________________________________________ -->
5957<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5958 <a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5959</div>
5960
5961<div class="doc_text">
5962
5963<h5>Syntax:</h5>
5964<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]5965 declare i64 @llvm.readcyclecounter()
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5966</pre>
5967
5968<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5969<p>The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
5970 counter register (or similar low latency, high accuracy clocks) on those
5971 targets that support it. On X86, it should map to RDTSC. On Alpha, it
5972 should map to RPCC. As the backing counters overflow quickly (on the order
5973 of 9 seconds on alpha), this should only be used for small timings.</p>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5974
5975<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5976<p>When directly supported, reading the cycle counter should not modify any
5977 memory. Implementations are allowed to either return a application specific
5978 value or a system wide value. On backends without support, this is lowered
5979 to a constant 0.</p>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5980
5981</div>
5982
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5983<!-- ======================================================================= -->
5984<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5985 <a name="int_libc">Standard C Library Intrinsics</a>
5986</div>
5987
5988<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5989
5990<p>LLVM provides intrinsics for a few important standard C library functions.
5991 These intrinsics allow source-language front-ends to pass information about
5992 the alignment of the pointer arguments to the code generator, providing
5993 opportunity for more efficient code generation.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5994
5995</div>
5996
5997<!-- _______________________________________________________________________ -->
5998<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5999 <a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6000</div>
6001
6002<div class="doc_text">
6003
6004<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6005<p>This is an overloaded intrinsic. You can use <tt>llvm.memcpy</tt> on any
Mon P Wange88909b2010-04-07 06:35:53 +0000[diff] [blame]6006 integer bit width and for different address spaces. Not all targets support
6007 all bit widths however.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6008
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6009<pre>
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]6010 declare void @llvm.memcpy.p0i8.p0i8.i32(i8* &lt;dest&gt;, i8* &lt;src&gt;,
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6011 i32 &lt;len&gt;, i32 &lt;align&gt;, i1 &lt;isvolatile&gt;)
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]6012 declare void @llvm.memcpy.p0i8.p0i8.i64(i8* &lt;dest&gt;, i8* &lt;src&gt;,
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6013 i64 &lt;len&gt;, i32 &lt;align&gt;, i1 &lt;isvolatile&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6014</pre>
6015
6016<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6017<p>The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the
6018 source location to the destination location.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6019
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6020<p>Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6021 intrinsics do not return a value, takes extra alignment/isvolatile arguments
6022 and the pointers can be in specified address spaces.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6023
6024<h5>Arguments:</h5>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6025
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6026<p>The first argument is a pointer to the destination, the second is a pointer
6027 to the source. The third argument is an integer argument specifying the
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6028 number of bytes to copy, the fourth argument is the alignment of the
6029 source and destination locations, and the fifth is a boolean indicating a
6030 volatile access.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6031
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]6032<p>If the call to this intrinsic has an alignment value that is not 0 or 1,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6033 then the caller guarantees that both the source and destination pointers are
6034 aligned to that boundary.</p>
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]6035
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]6036<p>If the <tt>isvolatile</tt> parameter is <tt>true</tt>, the
6037 <tt>llvm.memcpy</tt> call is a <a href="#volatile">volatile operation</a>.
6038 The detailed access behavior is not very cleanly specified and it is unwise
6039 to depend on it.</p>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6040
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6041<h5>Semantics:</h5>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6042
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6043<p>The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the
6044 source location to the destination location, which are not allowed to
6045 overlap. It copies "len" bytes of memory over. If the argument is known to
6046 be aligned to some boundary, this can be specified as the fourth argument,
6047 otherwise it should be set to 0 or 1.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6048
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]6049</div>
6050
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6051<!-- _______________________________________________________________________ -->
6052<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]6053 <a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6054</div>
6055
6056<div class="doc_text">
6057
6058<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]6059<p>This is an overloaded intrinsic. You can use llvm.memmove on any integer bit
Mon P Wange88909b2010-04-07 06:35:53 +0000[diff] [blame]6060 width and for different address space. Not all targets support all bit
6061 widths however.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6062
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6063<pre>
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]6064 declare void @llvm.memmove.p0i8.p0i8.i32(i8* &lt;dest&gt;, i8* &lt;src&gt;,
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6065 i32 &lt;len&gt;, i32 &lt;align&gt;, i1 &lt;isvolatile&gt;)
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]6066 declare void @llvm.memmove.p0i8.p0i8.i64(i8* &lt;dest&gt;, i8* &lt;src&gt;,
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6067 i64 &lt;len&gt;, i32 &lt;align&gt;, i1 &lt;isvolatile&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6068</pre>
6069
6070<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6071<p>The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the
6072 source location to the destination location. It is similar to the
6073 '<tt>llvm.memcpy</tt>' intrinsic but allows the two memory locations to
6074 overlap.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6075
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6076<p>Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6077 intrinsics do not return a value, takes extra alignment/isvolatile arguments
6078 and the pointers can be in specified address spaces.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6079
6080<h5>Arguments:</h5>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6081
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6082<p>The first argument is a pointer to the destination, the second is a pointer
6083 to the source. The third argument is an integer argument specifying the
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6084 number of bytes to copy, the fourth argument is the alignment of the
6085 source and destination locations, and the fifth is a boolean indicating a
6086 volatile access.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6087
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]6088<p>If the call to this intrinsic has an alignment value that is not 0 or 1,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6089 then the caller guarantees that the source and destination pointers are
6090 aligned to that boundary.</p>
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]6091
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]6092<p>If the <tt>isvolatile</tt> parameter is <tt>true</tt>, the
6093 <tt>llvm.memmove</tt> call is a <a href="#volatile">volatile operation</a>.
6094 The detailed access behavior is not very cleanly specified and it is unwise
6095 to depend on it.</p>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6096
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6097<h5>Semantics:</h5>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6098
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6099<p>The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the
6100 source location to the destination location, which may overlap. It copies
6101 "len" bytes of memory over. If the argument is known to be aligned to some
6102 boundary, this can be specified as the fourth argument, otherwise it should
6103 be set to 0 or 1.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6104
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]6105</div>
6106
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6107<!-- _______________________________________________________________________ -->
6108<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]6109 <a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6110</div>
6111
6112<div class="doc_text">
6113
6114<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]6115<p>This is an overloaded intrinsic. You can use llvm.memset on any integer bit
John Criswellcdcbbfc2010-07-30 16:30:28 +0000[diff] [blame]6116 width and for different address spaces. However, not all targets support all
6117 bit widths.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6118
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6119<pre>
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]6120 declare void @llvm.memset.p0i8.i32(i8* &lt;dest&gt;, i8 &lt;val&gt;,
Chris Lattnerff35c3f2010-04-08 00:54:34 +0000[diff] [blame]6121 i32 &lt;len&gt;, i32 &lt;align&gt;, i1 &lt;isvolatile&gt;)
Dan Gohmanfe47aae2010-05-28 17:13:49 +0000[diff] [blame]6122 declare void @llvm.memset.p0i8.i64(i8* &lt;dest&gt;, i8 &lt;val&gt;,
Chris Lattnerff35c3f2010-04-08 00:54:34 +0000[diff] [blame]6123 i64 &lt;len&gt;, i32 &lt;align&gt;, i1 &lt;isvolatile&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6124</pre>
6125
6126<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6127<p>The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a
6128 particular byte value.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6129
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6130<p>Note that, unlike the standard libc function, the <tt>llvm.memset</tt>
John Criswellcdcbbfc2010-07-30 16:30:28 +0000[diff] [blame]6131 intrinsic does not return a value and takes extra alignment/volatile
6132 arguments. Also, the destination can be in an arbitrary address space.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6133
6134<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6135<p>The first argument is a pointer to the destination to fill, the second is the
John Criswellcdcbbfc2010-07-30 16:30:28 +0000[diff] [blame]6136 byte value with which to fill it, the third argument is an integer argument
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6137 specifying the number of bytes to fill, and the fourth argument is the known
John Criswellcdcbbfc2010-07-30 16:30:28 +0000[diff] [blame]6138 alignment of the destination location.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6139
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]6140<p>If the call to this intrinsic has an alignment value that is not 0 or 1,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6141 then the caller guarantees that the destination pointer is aligned to that
6142 boundary.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6143
Jeffrey Yasskin93e066d2010-04-26 21:21:24 +0000[diff] [blame]6144<p>If the <tt>isvolatile</tt> parameter is <tt>true</tt>, the
6145 <tt>llvm.memset</tt> call is a <a href="#volatile">volatile operation</a>.
6146 The detailed access behavior is not very cleanly specified and it is unwise
6147 to depend on it.</p>
Chris Lattner9f636de2010-04-08 00:53:57 +0000[diff] [blame]6148
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6149<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6150<p>The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting
6151 at the destination location. If the argument is known to be aligned to some
6152 boundary, this can be specified as the fourth argument, otherwise it should
6153 be set to 0 or 1.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6154
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]6155</div>
6156
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]6157<!-- _______________________________________________________________________ -->
6158<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]6159 <a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6160</div>
6161
6162<div class="doc_text">
6163
6164<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6165<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
6166 floating point or vector of floating point type. Not all targets support all
6167 types however.</p>
6168
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6169<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]6170 declare float @llvm.sqrt.f32(float %Val)
6171 declare double @llvm.sqrt.f64(double %Val)
6172 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
6173 declare fp128 @llvm.sqrt.f128(fp128 %Val)
6174 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6175</pre>
6176
6177<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6178<p>The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
6179 returning the same value as the libm '<tt>sqrt</tt>' functions would.
6180 Unlike <tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined
6181 behavior for negative numbers other than -0.0 (which allows for better
6182 optimization, because there is no need to worry about errno being
6183 set). <tt>llvm.sqrt(-0.0)</tt> is defined to return -0.0 like IEEE sqrt.</p>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6184
6185<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6186<p>The argument and return value are floating point numbers of the same
6187 type.</p>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6188
6189<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6190<p>This function returns the sqrt of the specified operand if it is a
6191 nonnegative floating point number.</p>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6192
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]6193</div>
6194
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6195<!-- _______________________________________________________________________ -->
6196<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]6197 <a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6198</div>
6199
6200<div class="doc_text">
6201
6202<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6203<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
6204 floating point or vector of floating point type. Not all targets support all
6205 types however.</p>
6206
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6207<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]6208 declare float @llvm.powi.f32(float %Val, i32 %power)
6209 declare double @llvm.powi.f64(double %Val, i32 %power)
6210 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
6211 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
6212 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6213</pre>
6214
6215<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6216<p>The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
6217 specified (positive or negative) power. The order of evaluation of
6218 multiplications is not defined. When a vector of floating point type is
6219 used, the second argument remains a scalar integer value.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6220
6221<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6222<p>The second argument is an integer power, and the first is a value to raise to
6223 that power.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6224
6225<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6226<p>This function returns the first value raised to the second power with an
6227 unspecified sequence of rounding operations.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6228
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]6229</div>
6230
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6231<!-- _______________________________________________________________________ -->
6232<div class="doc_subsubsection">
6233 <a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
6234</div>
6235
6236<div class="doc_text">
6237
6238<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6239<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
6240 floating point or vector of floating point type. Not all targets support all
6241 types however.</p>
6242
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6243<pre>
6244 declare float @llvm.sin.f32(float %Val)
6245 declare double @llvm.sin.f64(double %Val)
6246 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
6247 declare fp128 @llvm.sin.f128(fp128 %Val)
6248 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
6249</pre>
6250
6251<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6252<p>The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6253
6254<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6255<p>The argument and return value are floating point numbers of the same
6256 type.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6257
6258<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6259<p>This function returns the sine of the specified operand, returning the same
6260 values as the libm <tt>sin</tt> functions would, and handles error conditions
6261 in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6262
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6263</div>
6264
6265<!-- _______________________________________________________________________ -->
6266<div class="doc_subsubsection">
6267 <a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
6268</div>
6269
6270<div class="doc_text">
6271
6272<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6273<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
6274 floating point or vector of floating point type. Not all targets support all
6275 types however.</p>
6276
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6277<pre>
6278 declare float @llvm.cos.f32(float %Val)
6279 declare double @llvm.cos.f64(double %Val)
6280 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
6281 declare fp128 @llvm.cos.f128(fp128 %Val)
6282 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
6283</pre>
6284
6285<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6286<p>The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6287
6288<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6289<p>The argument and return value are floating point numbers of the same
6290 type.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6291
6292<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6293<p>This function returns the cosine of the specified operand, returning the same
6294 values as the libm <tt>cos</tt> functions would, and handles error conditions
6295 in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6296
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6297</div>
6298
6299<!-- _______________________________________________________________________ -->
6300<div class="doc_subsubsection">
6301 <a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
6302</div>
6303
6304<div class="doc_text">
6305
6306<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6307<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
6308 floating point or vector of floating point type. Not all targets support all
6309 types however.</p>
6310
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6311<pre>
6312 declare float @llvm.pow.f32(float %Val, float %Power)
6313 declare double @llvm.pow.f64(double %Val, double %Power)
6314 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
6315 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
6316 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
6317</pre>
6318
6319<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6320<p>The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
6321 specified (positive or negative) power.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6322
6323<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6324<p>The second argument is a floating point power, and the first is a value to
6325 raise to that power.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6326
6327<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6328<p>This function returns the first value raised to the second power, returning
6329 the same values as the libm <tt>pow</tt> functions would, and handles error
6330 conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6331
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6332</div>
6333
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6334<!-- ======================================================================= -->
6335<div class="doc_subsection">
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6336 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6337</div>
6338
6339<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6340
6341<p>LLVM provides intrinsics for a few important bit manipulation operations.
6342 These allow efficient code generation for some algorithms.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6343
6344</div>
6345
6346<!-- _______________________________________________________________________ -->
6347<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]6348 <a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6349</div>
6350
6351<div class="doc_text">
6352
6353<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]6354<p>This is an overloaded intrinsic function. You can use bswap on any integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6355 type that is an even number of bytes (i.e. BitWidth % 16 == 0).</p>
6356
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6357<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6358 declare i16 @llvm.bswap.i16(i16 &lt;id&gt;)
6359 declare i32 @llvm.bswap.i32(i32 &lt;id&gt;)
6360 declare i64 @llvm.bswap.i64(i64 &lt;id&gt;)
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6361</pre>
6362
6363<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6364<p>The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
6365 values with an even number of bytes (positive multiple of 16 bits). These
6366 are useful for performing operations on data that is not in the target's
6367 native byte order.</p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6368
6369<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6370<p>The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
6371 and low byte of the input i16 swapped. Similarly,
6372 the <tt>llvm.bswap.i32</tt> intrinsic returns an i32 value that has the four
6373 bytes of the input i32 swapped, so that if the input bytes are numbered 0, 1,
6374 2, 3 then the returned i32 will have its bytes in 3, 2, 1, 0 order.
6375 The <tt>llvm.bswap.i48</tt>, <tt>llvm.bswap.i64</tt> and other intrinsics
6376 extend this concept to additional even-byte lengths (6 bytes, 8 bytes and
6377 more, respectively).</p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6378
6379</div>
6380
6381<!-- _______________________________________________________________________ -->
6382<div class="doc_subsubsection">
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]6383 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6384</div>
6385
6386<div class="doc_text">
6387
6388<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]6389<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6390 width. Not all targets support all bit widths however.</p>
6391
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6392<pre>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6393 declare i8 @llvm.ctpop.i8(i8 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6394 declare i16 @llvm.ctpop.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]6395 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6396 declare i64 @llvm.ctpop.i64(i64 &lt;src&gt;)
6397 declare i256 @llvm.ctpop.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6398</pre>
6399
6400<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6401<p>The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set
6402 in a value.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6403
6404<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6405<p>The only argument is the value to be counted. The argument may be of any
6406 integer type. The return type must match the argument type.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6407
6408<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6409<p>The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6410
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6411</div>
6412
6413<!-- _______________________________________________________________________ -->
6414<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]6415 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6416</div>
6417
6418<div class="doc_text">
6419
6420<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6421<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
6422 integer bit width. Not all targets support all bit widths however.</p>
6423
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6424<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6425 declare i8 @llvm.ctlz.i8 (i8 &lt;src&gt;)
6426 declare i16 @llvm.ctlz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]6427 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6428 declare i64 @llvm.ctlz.i64(i64 &lt;src&gt;)
6429 declare i256 @llvm.ctlz.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6430</pre>
6431
6432<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6433<p>The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
6434 leading zeros in a variable.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6435
6436<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6437<p>The only argument is the value to be counted. The argument may be of any
6438 integer type. The return type must match the argument type.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6439
6440<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6441<p>The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant)
6442 zeros in a variable. If the src == 0 then the result is the size in bits of
6443 the type of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6444
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6445</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]6446
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6447<!-- _______________________________________________________________________ -->
6448<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]6449 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6450</div>
6451
6452<div class="doc_text">
6453
6454<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6455<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
6456 integer bit width. Not all targets support all bit widths however.</p>
6457
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6458<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6459 declare i8 @llvm.cttz.i8 (i8 &lt;src&gt;)
6460 declare i16 @llvm.cttz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]6461 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6462 declare i64 @llvm.cttz.i64(i64 &lt;src&gt;)
6463 declare i256 @llvm.cttz.i256(i256 &lt;src&gt;)
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6464</pre>
6465
6466<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6467<p>The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
6468 trailing zeros.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6469
6470<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6471<p>The only argument is the value to be counted. The argument may be of any
6472 integer type. The return type must match the argument type.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6473
6474<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6475<p>The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant)
6476 zeros in a variable. If the src == 0 then the result is the size in bits of
6477 the type of src. For example, <tt>llvm.cttz(2) = 1</tt>.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6478
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6479</div>
6480
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6481<!-- ======================================================================= -->
6482<div class="doc_subsection">
6483 <a name="int_overflow">Arithmetic with Overflow Intrinsics</a>
6484</div>
6485
6486<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6487
6488<p>LLVM provides intrinsics for some arithmetic with overflow operations.</p>
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6489
6490</div>
6491
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6492<!-- _______________________________________________________________________ -->
6493<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6494 <a name="int_sadd_overflow">'<tt>llvm.sadd.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6495</div>
6496
6497<div class="doc_text">
6498
6499<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6500<p>This is an overloaded intrinsic. You can use <tt>llvm.sadd.with.overflow</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6501 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6502
6503<pre>
6504 declare {i16, i1} @llvm.sadd.with.overflow.i16(i16 %a, i16 %b)
6505 declare {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
6506 declare {i64, i1} @llvm.sadd.with.overflow.i64(i64 %a, i64 %b)
6507</pre>
6508
6509<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6510<p>The '<tt>llvm.sadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6511 a signed addition of the two arguments, and indicate whether an overflow
6512 occurred during the signed summation.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6513
6514<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6515<p>The arguments (%a and %b) and the first element of the result structure may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6516 be of integer types of any bit width, but they must have the same bit
6517 width. The second element of the result structure must be of
6518 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6519 undergo signed addition.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6520
6521<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6522<p>The '<tt>llvm.sadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6523 a signed addition of the two variables. They return a structure &mdash; the
6524 first element of which is the signed summation, and the second element of
6525 which is a bit specifying if the signed summation resulted in an
6526 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6527
6528<h5>Examples:</h5>
6529<pre>
6530 %res = call {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
6531 %sum = extractvalue {i32, i1} %res, 0
6532 %obit = extractvalue {i32, i1} %res, 1
6533 br i1 %obit, label %overflow, label %normal
6534</pre>
6535
6536</div>
6537
6538<!-- _______________________________________________________________________ -->
6539<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6540 <a name="int_uadd_overflow">'<tt>llvm.uadd.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6541</div>
6542
6543<div class="doc_text">
6544
6545<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6546<p>This is an overloaded intrinsic. You can use <tt>llvm.uadd.with.overflow</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6547 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6548
6549<pre>
6550 declare {i16, i1} @llvm.uadd.with.overflow.i16(i16 %a, i16 %b)
6551 declare {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
6552 declare {i64, i1} @llvm.uadd.with.overflow.i64(i64 %a, i64 %b)
6553</pre>
6554
6555<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6556<p>The '<tt>llvm.uadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6557 an unsigned addition of the two arguments, and indicate whether a carry
6558 occurred during the unsigned summation.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6559
6560<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6561<p>The arguments (%a and %b) and the first element of the result structure may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6562 be of integer types of any bit width, but they must have the same bit
6563 width. The second element of the result structure must be of
6564 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6565 undergo unsigned addition.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6566
6567<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6568<p>The '<tt>llvm.uadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6569 an unsigned addition of the two arguments. They return a structure &mdash;
6570 the first element of which is the sum, and the second element of which is a
6571 bit specifying if the unsigned summation resulted in a carry.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6572
6573<h5>Examples:</h5>
6574<pre>
6575 %res = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
6576 %sum = extractvalue {i32, i1} %res, 0
6577 %obit = extractvalue {i32, i1} %res, 1
6578 br i1 %obit, label %carry, label %normal
6579</pre>
6580
6581</div>
6582
6583<!-- _______________________________________________________________________ -->
6584<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6585 <a name="int_ssub_overflow">'<tt>llvm.ssub.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6586</div>
6587
6588<div class="doc_text">
6589
6590<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6591<p>This is an overloaded intrinsic. You can use <tt>llvm.ssub.with.overflow</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6592 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6593
6594<pre>
6595 declare {i16, i1} @llvm.ssub.with.overflow.i16(i16 %a, i16 %b)
6596 declare {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
6597 declare {i64, i1} @llvm.ssub.with.overflow.i64(i64 %a, i64 %b)
6598</pre>
6599
6600<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6601<p>The '<tt>llvm.ssub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6602 a signed subtraction of the two arguments, and indicate whether an overflow
6603 occurred during the signed subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6604
6605<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6606<p>The arguments (%a and %b) and the first element of the result structure may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6607 be of integer types of any bit width, but they must have the same bit
6608 width. The second element of the result structure must be of
6609 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6610 undergo signed subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6611
6612<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6613<p>The '<tt>llvm.ssub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6614 a signed subtraction of the two arguments. They return a structure &mdash;
6615 the first element of which is the subtraction, and the second element of
6616 which is a bit specifying if the signed subtraction resulted in an
6617 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6618
6619<h5>Examples:</h5>
6620<pre>
6621 %res = call {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
6622 %sum = extractvalue {i32, i1} %res, 0
6623 %obit = extractvalue {i32, i1} %res, 1
6624 br i1 %obit, label %overflow, label %normal
6625</pre>
6626
6627</div>
6628
6629<!-- _______________________________________________________________________ -->
6630<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6631 <a name="int_usub_overflow">'<tt>llvm.usub.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6632</div>
6633
6634<div class="doc_text">
6635
6636<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6637<p>This is an overloaded intrinsic. You can use <tt>llvm.usub.with.overflow</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6638 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6639
6640<pre>
6641 declare {i16, i1} @llvm.usub.with.overflow.i16(i16 %a, i16 %b)
6642 declare {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
6643 declare {i64, i1} @llvm.usub.with.overflow.i64(i64 %a, i64 %b)
6644</pre>
6645
6646<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6647<p>The '<tt>llvm.usub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6648 an unsigned subtraction of the two arguments, and indicate whether an
6649 overflow occurred during the unsigned subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6650
6651<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6652<p>The arguments (%a and %b) and the first element of the result structure may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6653 be of integer types of any bit width, but they must have the same bit
6654 width. The second element of the result structure must be of
6655 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6656 undergo unsigned subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6657
6658<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6659<p>The '<tt>llvm.usub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6660 an unsigned subtraction of the two arguments. They return a structure &mdash;
6661 the first element of which is the subtraction, and the second element of
6662 which is a bit specifying if the unsigned subtraction resulted in an
6663 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6664
6665<h5>Examples:</h5>
6666<pre>
6667 %res = call {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
6668 %sum = extractvalue {i32, i1} %res, 0
6669 %obit = extractvalue {i32, i1} %res, 1
6670 br i1 %obit, label %overflow, label %normal
6671</pre>
6672
6673</div>
6674
6675<!-- _______________________________________________________________________ -->
6676<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6677 <a name="int_smul_overflow">'<tt>llvm.smul.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6678</div>
6679
6680<div class="doc_text">
6681
6682<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6683<p>This is an overloaded intrinsic. You can use <tt>llvm.smul.with.overflow</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6684 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6685
6686<pre>
6687 declare {i16, i1} @llvm.smul.with.overflow.i16(i16 %a, i16 %b)
6688 declare {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
6689 declare {i64, i1} @llvm.smul.with.overflow.i64(i64 %a, i64 %b)
6690</pre>
6691
6692<h5>Overview:</h5>
6693
6694<p>The '<tt>llvm.smul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6695 a signed multiplication of the two arguments, and indicate whether an
6696 overflow occurred during the signed multiplication.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6697
6698<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6699<p>The arguments (%a and %b) and the first element of the result structure may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6700 be of integer types of any bit width, but they must have the same bit
6701 width. The second element of the result structure must be of
6702 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6703 undergo signed multiplication.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6704
6705<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6706<p>The '<tt>llvm.smul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6707 a signed multiplication of the two arguments. They return a structure &mdash;
6708 the first element of which is the multiplication, and the second element of
6709 which is a bit specifying if the signed multiplication resulted in an
6710 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6711
6712<h5>Examples:</h5>
6713<pre>
6714 %res = call {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
6715 %sum = extractvalue {i32, i1} %res, 0
6716 %obit = extractvalue {i32, i1} %res, 1
6717 br i1 %obit, label %overflow, label %normal
6718</pre>
6719
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]6720</div>
6721
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6722<!-- _______________________________________________________________________ -->
6723<div class="doc_subsubsection">
6724 <a name="int_umul_overflow">'<tt>llvm.umul.with.overflow.*</tt>' Intrinsics</a>
6725</div>
6726
6727<div class="doc_text">
6728
6729<h5>Syntax:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6730<p>This is an overloaded intrinsic. You can use <tt>llvm.umul.with.overflow</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6731 on any integer bit width.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6732
6733<pre>
6734 declare {i16, i1} @llvm.umul.with.overflow.i16(i16 %a, i16 %b)
6735 declare {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
6736 declare {i64, i1} @llvm.umul.with.overflow.i64(i64 %a, i64 %b)
6737</pre>
6738
6739<h5>Overview:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6740<p>The '<tt>llvm.umul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6741 a unsigned multiplication of the two arguments, and indicate whether an
6742 overflow occurred during the unsigned multiplication.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6743
6744<h5>Arguments:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6745<p>The arguments (%a and %b) and the first element of the result structure may
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6746 be of integer types of any bit width, but they must have the same bit
6747 width. The second element of the result structure must be of
6748 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6749 undergo unsigned multiplication.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6750
6751<h5>Semantics:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6752<p>The '<tt>llvm.umul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6753 an unsigned multiplication of the two arguments. They return a structure
6754 &mdash; the first element of which is the multiplication, and the second
6755 element of which is a bit specifying if the unsigned multiplication resulted
6756 in an overflow.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6757
6758<h5>Examples:</h5>
6759<pre>
6760 %res = call {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
6761 %sum = extractvalue {i32, i1} %res, 0
6762 %obit = extractvalue {i32, i1} %res, 1
6763 br i1 %obit, label %overflow, label %normal
6764</pre>
6765
6766</div>
6767
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6768<!-- ======================================================================= -->
6769<div class="doc_subsection">
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]6770 <a name="int_fp16">Half Precision Floating Point Intrinsics</a>
6771</div>
6772
6773<div class="doc_text">
6774
Chris Lattner0cec9c82010-03-15 04:12:21 +0000[diff] [blame]6775<p>Half precision floating point is a storage-only format. This means that it is
6776 a dense encoding (in memory) but does not support computation in the
6777 format.</p>
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6778
Chris Lattner0cec9c82010-03-15 04:12:21 +0000[diff] [blame]6779<p>This means that code must first load the half-precision floating point
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6780 value as an i16, then convert it to float with <a
6781 href="#int_convert_from_fp16"><tt>llvm.convert.from.fp16</tt></a>.
6782 Computation can then be performed on the float value (including extending to
Chris Lattner0cec9c82010-03-15 04:12:21 +0000[diff] [blame]6783 double etc). To store the value back to memory, it is first converted to
6784 float if needed, then converted to i16 with
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6785 <a href="#int_convert_to_fp16"><tt>llvm.convert.to.fp16</tt></a>, then
6786 storing as an i16 value.</p>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]6787</div>
6788
6789<!-- _______________________________________________________________________ -->
6790<div class="doc_subsubsection">
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6791 <a name="int_convert_to_fp16">'<tt>llvm.convert.to.fp16</tt>' Intrinsic</a>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]6792</div>
6793
6794<div class="doc_text">
6795
6796<h5>Syntax:</h5>
6797<pre>
6798 declare i16 @llvm.convert.to.fp16(f32 %a)
6799</pre>
6800
6801<h5>Overview:</h5>
6802<p>The '<tt>llvm.convert.to.fp16</tt>' intrinsic function performs
6803 a conversion from single precision floating point format to half precision
6804 floating point format.</p>
6805
6806<h5>Arguments:</h5>
6807<p>The intrinsic function contains single argument - the value to be
6808 converted.</p>
6809
6810<h5>Semantics:</h5>
6811<p>The '<tt>llvm.convert.to.fp16</tt>' intrinsic function performs
6812 a conversion from single precision floating point format to half precision
Chris Lattner0cec9c82010-03-15 04:12:21 +0000[diff] [blame]6813 floating point format. The return value is an <tt>i16</tt> which
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6814 contains the converted number.</p>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]6815
6816<h5>Examples:</h5>
6817<pre>
6818 %res = call i16 @llvm.convert.to.fp16(f32 %a)
6819 store i16 %res, i16* @x, align 2
6820</pre>
6821
6822</div>
6823
6824<!-- _______________________________________________________________________ -->
6825<div class="doc_subsubsection">
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6826 <a name="int_convert_from_fp16">'<tt>llvm.convert.from.fp16</tt>' Intrinsic</a>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]6827</div>
6828
6829<div class="doc_text">
6830
6831<h5>Syntax:</h5>
6832<pre>
6833 declare f32 @llvm.convert.from.fp16(i16 %a)
6834</pre>
6835
6836<h5>Overview:</h5>
6837<p>The '<tt>llvm.convert.from.fp16</tt>' intrinsic function performs
6838 a conversion from half precision floating point format to single precision
6839 floating point format.</p>
6840
6841<h5>Arguments:</h5>
6842<p>The intrinsic function contains single argument - the value to be
6843 converted.</p>
6844
6845<h5>Semantics:</h5>
6846<p>The '<tt>llvm.convert.from.fp16</tt>' intrinsic function performs a
Chris Lattner0cec9c82010-03-15 04:12:21 +0000[diff] [blame]6847 conversion from half single precision floating point format to single
Chris Lattner82c3dc62010-03-14 23:03:31 +0000[diff] [blame]6848 precision floating point format. The input half-float value is represented by
6849 an <tt>i16</tt> value.</p>
Anton Korobeynikovf02e7302010-03-14 18:42:47 +0000[diff] [blame]6850
6851<h5>Examples:</h5>
6852<pre>
6853 %a = load i16* @x, align 2
6854 %res = call f32 @llvm.convert.from.fp16(i16 %a)
6855</pre>
6856
6857</div>
6858
6859<!-- ======================================================================= -->
6860<div class="doc_subsection">
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6861 <a name="int_debugger">Debugger Intrinsics</a>
6862</div>
6863
6864<div class="doc_text">
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6865
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6866<p>The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt>
6867 prefix), are described in
6868 the <a href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source
6869 Level Debugging</a> document.</p>
6870
6871</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6872
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]6873<!-- ======================================================================= -->
6874<div class="doc_subsection">
6875 <a name="int_eh">Exception Handling Intrinsics</a>
6876</div>
6877
6878<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6879
6880<p>The LLVM exception handling intrinsics (which all start with
6881 <tt>llvm.eh.</tt> prefix), are described in
6882 the <a href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
6883 Handling</a> document.</p>
6884
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]6885</div>
6886
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]6887<!-- ======================================================================= -->
6888<div class="doc_subsection">
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]6889 <a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6890</div>
6891
6892<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6893
6894<p>This intrinsic makes it possible to excise one parameter, marked with
Dan Gohmanff235352010-07-02 23:18:08 +0000[diff] [blame]6895 the <a href="#nest"><tt>nest</tt></a> attribute, from a function.
6896 The result is a callable
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6897 function pointer lacking the nest parameter - the caller does not need to
6898 provide a value for it. Instead, the value to use is stored in advance in a
6899 "trampoline", a block of memory usually allocated on the stack, which also
6900 contains code to splice the nest value into the argument list. This is used
6901 to implement the GCC nested function address extension.</p>
6902
6903<p>For example, if the function is
6904 <tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
6905 pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as
6906 follows:</p>
6907
Benjamin Kramer26fe25f2010-07-13 12:26:09 +0000[diff] [blame]6908<pre class="doc_code">
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]6909 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
6910 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]6911 %p = call i8* @llvm.init.trampoline(i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8*), i8* %nval)
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]6912 %fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6913</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6914
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]6915<p>The call <tt>%val = call i32 %fp(i32 %x, i32 %y)</tt> is then equivalent
6916 to <tt>%val = call i32 %f(i8* %nval, i32 %x, i32 %y)</tt>.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6917
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6918</div>
6919
6920<!-- _______________________________________________________________________ -->
6921<div class="doc_subsubsection">
6922 <a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
6923</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6924
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6925<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6926
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6927<h5>Syntax:</h5>
6928<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6929 declare 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]6930</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6931
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6932<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6933<p>This fills the memory pointed to by <tt>tramp</tt> with code and returns a
6934 function pointer suitable for executing it.</p>
6935
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6936<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6937<p>The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
6938 pointers. The <tt>tramp</tt> argument must point to a sufficiently large and
6939 sufficiently aligned block of memory; this memory is written to by the
6940 intrinsic. Note that the size and the alignment are target-specific - LLVM
6941 currently provides no portable way of determining them, so a front-end that
6942 generates this intrinsic needs to have some target-specific knowledge.
6943 The <tt>func</tt> argument must hold a function bitcast to
6944 an <tt>i8*</tt>.</p>
6945
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6946<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6947<p>The block of memory pointed to by <tt>tramp</tt> is filled with target
6948 dependent code, turning it into a function. A pointer to this function is
6949 returned, but needs to be bitcast to an <a href="#int_trampoline">appropriate
6950 function pointer type</a> before being called. The new function's signature
6951 is the same as that of <tt>func</tt> with any arguments marked with
6952 the <tt>nest</tt> attribute removed. At most one such <tt>nest</tt> argument
6953 is allowed, and it must be of pointer type. Calling the new function is
6954 equivalent to calling <tt>func</tt> with the same argument list, but
6955 with <tt>nval</tt> used for the missing <tt>nest</tt> argument. If, after
6956 calling <tt>llvm.init.trampoline</tt>, the memory pointed to
6957 by <tt>tramp</tt> is modified, then the effect of any later call to the
6958 returned function pointer is undefined.</p>
6959
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6960</div>
6961
6962<!-- ======================================================================= -->
6963<div class="doc_subsection">
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6964 <a name="int_atomics">Atomic Operations and Synchronization Intrinsics</a>
6965</div>
6966
6967<div class="doc_text">
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6968
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6969<p>These intrinsic functions expand the "universal IR" of LLVM to represent
6970 hardware constructs for atomic operations and memory synchronization. This
6971 provides an interface to the hardware, not an interface to the programmer. It
6972 is aimed at a low enough level to allow any programming models or APIs
6973 (Application Programming Interfaces) which need atomic behaviors to map
6974 cleanly onto it. It is also modeled primarily on hardware behavior. Just as
6975 hardware provides a "universal IR" for source languages, it also provides a
6976 starting point for developing a "universal" atomic operation and
6977 synchronization IR.</p>
6978
6979<p>These do <em>not</em> form an API such as high-level threading libraries,
6980 software transaction memory systems, atomic primitives, and intrinsic
6981 functions as found in BSD, GNU libc, atomic_ops, APR, and other system and
6982 application libraries. The hardware interface provided by LLVM should allow
6983 a clean implementation of all of these APIs and parallel programming models.
6984 No one model or paradigm should be selected above others unless the hardware
6985 itself ubiquitously does so.</p>
6986
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6987</div>
6988
6989<!-- _______________________________________________________________________ -->
6990<div class="doc_subsubsection">
6991 <a name="int_memory_barrier">'<tt>llvm.memory.barrier</tt>' Intrinsic</a>
6992</div>
6993<div class="doc_text">
6994<h5>Syntax:</h5>
6995<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]6996 declare void @llvm.memory.barrier(i1 &lt;ll&gt;, i1 &lt;ls&gt;, i1 &lt;sl&gt;, i1 &lt;ss&gt;, i1 &lt;device&gt;)
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6997</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6998
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6999<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7000<p>The <tt>llvm.memory.barrier</tt> intrinsic guarantees ordering between
7001 specific pairs of memory access types.</p>
7002
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7003<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7004<p>The <tt>llvm.memory.barrier</tt> intrinsic requires five boolean arguments.
7005 The first four arguments enables a specific barrier as listed below. The
Dan Gohmanb55a1ee2010-03-01 17:41:39 +0000[diff] [blame]7006 fifth argument specifies that the barrier applies to io or device or uncached
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7007 memory.</p>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7008
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7009<ul>
7010 <li><tt>ll</tt>: load-load barrier</li>
7011 <li><tt>ls</tt>: load-store barrier</li>
7012 <li><tt>sl</tt>: store-load barrier</li>
7013 <li><tt>ss</tt>: store-store barrier</li>
7014 <li><tt>device</tt>: barrier applies to device and uncached memory also.</li>
7015</ul>
7016
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7017<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7018<p>This intrinsic causes the system to enforce some ordering constraints upon
7019 the loads and stores of the program. This barrier does not
7020 indicate <em>when</em> any events will occur, it only enforces
7021 an <em>order</em> in which they occur. For any of the specified pairs of load
7022 and store operations (f.ex. load-load, or store-load), all of the first
7023 operations preceding the barrier will complete before any of the second
7024 operations succeeding the barrier begin. Specifically the semantics for each
7025 pairing is as follows:</p>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7026
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7027<ul>
7028 <li><tt>ll</tt>: All loads before the barrier must complete before any load
7029 after the barrier begins.</li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7030 <li><tt>ls</tt>: All loads before the barrier must complete before any
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7031 store after the barrier begins.</li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7032 <li><tt>ss</tt>: All stores before the barrier must complete before any
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7033 store after the barrier begins.</li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7034 <li><tt>sl</tt>: All stores before the barrier must complete before any
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7035 load after the barrier begins.</li>
7036</ul>
7037
7038<p>These semantics are applied with a logical "and" behavior when more than one
7039 is enabled in a single memory barrier intrinsic.</p>
7040
7041<p>Backends may implement stronger barriers than those requested when they do
7042 not support as fine grained a barrier as requested. Some architectures do
7043 not need all types of barriers and on such architectures, these become
7044 noops.</p>
7045
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7046<h5>Example:</h5>
7047<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7048%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7049%ptr = bitcast i8* %mallocP to i32*
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7050 store i32 4, %ptr
7051
7052%result1 = load i32* %ptr <i>; yields {i32}:result1 = 4</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7053 call void @llvm.memory.barrier(i1 false, i1 true, i1 false, i1 false)
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7054 <i>; guarantee the above finishes</i>
7055 store i32 8, %ptr <i>; before this begins</i>
7056</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7057
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7058</div>
7059
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7060<!-- _______________________________________________________________________ -->
7061<div class="doc_subsubsection">
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7062 <a name="int_atomic_cmp_swap">'<tt>llvm.atomic.cmp.swap.*</tt>' Intrinsic</a>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7063</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7064
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7065<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7066
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7067<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7068<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.cmp.swap</tt> on
7069 any integer bit width and for different address spaces. Not all targets
7070 support all bit widths however.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7071
7072<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7073 declare i8 @llvm.atomic.cmp.swap.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;cmp&gt;, i8 &lt;val&gt;)
7074 declare i16 @llvm.atomic.cmp.swap.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;cmp&gt;, i16 &lt;val&gt;)
7075 declare i32 @llvm.atomic.cmp.swap.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;cmp&gt;, i32 &lt;val&gt;)
7076 declare i64 @llvm.atomic.cmp.swap.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;cmp&gt;, i64 &lt;val&gt;)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7077</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7078
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7079<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7080<p>This loads a value in memory and compares it to a given value. If they are
7081 equal, it stores a new value into the memory.</p>
7082
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7083<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7084<p>The <tt>llvm.atomic.cmp.swap</tt> intrinsic takes three arguments. The result
7085 as well as both <tt>cmp</tt> and <tt>val</tt> must be integer values with the
7086 same bit width. The <tt>ptr</tt> argument must be a pointer to a value of
7087 this integer type. While any bit width integer may be used, targets may only
7088 lower representations they support in hardware.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7089
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7090<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7091<p>This entire intrinsic must be executed atomically. It first loads the value
7092 in memory pointed to by <tt>ptr</tt> and compares it with the
7093 value <tt>cmp</tt>. If they are equal, <tt>val</tt> is stored into the
7094 memory. The loaded value is yielded in all cases. This provides the
7095 equivalent of an atomic compare-and-swap operation within the SSA
7096 framework.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7097
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7098<h5>Examples:</h5>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7099<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7100%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7101%ptr = bitcast i8* %mallocP to i32*
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7102 store i32 4, %ptr
7103
7104%val1 = add i32 4, 4
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7105%result1 = call i32 @llvm.atomic.cmp.swap.i32.p0i32(i32* %ptr, i32 4, %val1)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7106 <i>; yields {i32}:result1 = 4</i>
7107%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
7108%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
7109
7110%val2 = add i32 1, 1
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7111%result2 = call i32 @llvm.atomic.cmp.swap.i32.p0i32(i32* %ptr, i32 5, %val2)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7112 <i>; yields {i32}:result2 = 8</i>
7113%stored2 = icmp eq i32 %result2, 5 <i>; yields {i1}:stored2 = false</i>
7114
7115%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 8</i>
7116</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7117
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7118</div>
7119
7120<!-- _______________________________________________________________________ -->
7121<div class="doc_subsubsection">
7122 <a name="int_atomic_swap">'<tt>llvm.atomic.swap.*</tt>' Intrinsic</a>
7123</div>
7124<div class="doc_text">
7125<h5>Syntax:</h5>
7126
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7127<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.swap</tt> on any
7128 integer bit width. Not all targets support all bit widths however.</p>
7129
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7130<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7131 declare i8 @llvm.atomic.swap.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;val&gt;)
7132 declare i16 @llvm.atomic.swap.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;val&gt;)
7133 declare i32 @llvm.atomic.swap.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;val&gt;)
7134 declare i64 @llvm.atomic.swap.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;val&gt;)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7135</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7136
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7137<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7138<p>This intrinsic loads the value stored in memory at <tt>ptr</tt> and yields
7139 the value from memory. It then stores the value in <tt>val</tt> in the memory
7140 at <tt>ptr</tt>.</p>
7141
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7142<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7143<p>The <tt>llvm.atomic.swap</tt> intrinsic takes two arguments. Both
7144 the <tt>val</tt> argument and the result must be integers of the same bit
7145 width. The first argument, <tt>ptr</tt>, must be a pointer to a value of this
7146 integer type. The targets may only lower integer representations they
7147 support.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7148
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7149<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7150<p>This intrinsic loads the value pointed to by <tt>ptr</tt>, yields it, and
7151 stores <tt>val</tt> back into <tt>ptr</tt> atomically. This provides the
7152 equivalent of an atomic swap operation within the SSA framework.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7153
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7154<h5>Examples:</h5>
7155<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7156%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7157%ptr = bitcast i8* %mallocP to i32*
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7158 store i32 4, %ptr
7159
7160%val1 = add i32 4, 4
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7161%result1 = call i32 @llvm.atomic.swap.i32.p0i32(i32* %ptr, i32 %val1)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7162 <i>; yields {i32}:result1 = 4</i>
7163%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
7164%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
7165
7166%val2 = add i32 1, 1
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7167%result2 = call i32 @llvm.atomic.swap.i32.p0i32(i32* %ptr, i32 %val2)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7168 <i>; yields {i32}:result2 = 8</i>
7169
7170%stored2 = icmp eq i32 %result2, 8 <i>; yields {i1}:stored2 = true</i>
7171%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 2</i>
7172</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7173
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7174</div>
7175
7176<!-- _______________________________________________________________________ -->
7177<div class="doc_subsubsection">
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7178 <a name="int_atomic_load_add">'<tt>llvm.atomic.load.add.*</tt>' Intrinsic</a>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7179
7180</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7181
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7182<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7183
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7184<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7185<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.add</tt> on
7186 any integer bit width. Not all targets support all bit widths however.</p>
7187
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7188<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7189 declare i8 @llvm.atomic.load.add.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7190 declare i16 @llvm.atomic.load.add.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7191 declare i32 @llvm.atomic.load.add.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7192 declare i64 @llvm.atomic.load.add.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7193</pre>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7194
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7195<h5>Overview:</h5>
7196<p>This intrinsic adds <tt>delta</tt> to the value stored in memory
7197 at <tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.</p>
7198
7199<h5>Arguments:</h5>
7200<p>The intrinsic takes two arguments, the first a pointer to an integer value
7201 and the second an integer value. The result is also an integer value. These
7202 integer types can have any bit width, but they must all have the same bit
7203 width. The targets may only lower integer representations they support.</p>
7204
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7205<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7206<p>This intrinsic does a series of operations atomically. It first loads the
7207 value stored at <tt>ptr</tt>. It then adds <tt>delta</tt>, stores the result
7208 to <tt>ptr</tt>. It yields the original value stored at <tt>ptr</tt>.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7209
7210<h5>Examples:</h5>
7211<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7212%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7213%ptr = bitcast i8* %mallocP to i32*
7214 store i32 4, %ptr
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7215%result1 = call i32 @llvm.atomic.load.add.i32.p0i32(i32* %ptr, i32 4)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7216 <i>; yields {i32}:result1 = 4</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7217%result2 = call i32 @llvm.atomic.load.add.i32.p0i32(i32* %ptr, i32 2)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7218 <i>; yields {i32}:result2 = 8</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7219%result3 = call i32 @llvm.atomic.load.add.i32.p0i32(i32* %ptr, i32 5)
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7220 <i>; yields {i32}:result3 = 10</i>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7221%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 15</i>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7222</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7223
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]7224</div>
7225
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7226<!-- _______________________________________________________________________ -->
7227<div class="doc_subsubsection">
7228 <a name="int_atomic_load_sub">'<tt>llvm.atomic.load.sub.*</tt>' Intrinsic</a>
7229
7230</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7231
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7232<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7233
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7234<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7235<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.sub</tt> on
7236 any integer bit width and for different address spaces. Not all targets
7237 support all bit widths however.</p>
7238
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7239<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7240 declare i8 @llvm.atomic.load.sub.i8.p0i32(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7241 declare i16 @llvm.atomic.load.sub.i16.p0i32(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7242 declare i32 @llvm.atomic.load.sub.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7243 declare i64 @llvm.atomic.load.sub.i64.p0i32(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7244</pre>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7245
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7246<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7247<p>This intrinsic subtracts <tt>delta</tt> to the value stored in memory at
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7248 <tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.</p>
7249
7250<h5>Arguments:</h5>
7251<p>The intrinsic takes two arguments, the first a pointer to an integer value
7252 and the second an integer value. The result is also an integer value. These
7253 integer types can have any bit width, but they must all have the same bit
7254 width. The targets may only lower integer representations they support.</p>
7255
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7256<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7257<p>This intrinsic does a series of operations atomically. It first loads the
7258 value stored at <tt>ptr</tt>. It then subtracts <tt>delta</tt>, stores the
7259 result to <tt>ptr</tt>. It yields the original value stored
7260 at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7261
7262<h5>Examples:</h5>
7263<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7264%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7265%ptr = bitcast i8* %mallocP to i32*
7266 store i32 8, %ptr
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7267%result1 = call i32 @llvm.atomic.load.sub.i32.p0i32(i32* %ptr, i32 4)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7268 <i>; yields {i32}:result1 = 8</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7269%result2 = call i32 @llvm.atomic.load.sub.i32.p0i32(i32* %ptr, i32 2)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7270 <i>; yields {i32}:result2 = 4</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7271%result3 = call i32 @llvm.atomic.load.sub.i32.p0i32(i32* %ptr, i32 5)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7272 <i>; yields {i32}:result3 = 2</i>
7273%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = -3</i>
7274</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7275
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7276</div>
7277
7278<!-- _______________________________________________________________________ -->
7279<div class="doc_subsubsection">
7280 <a name="int_atomic_load_and">'<tt>llvm.atomic.load.and.*</tt>' Intrinsic</a><br>
7281 <a name="int_atomic_load_nand">'<tt>llvm.atomic.load.nand.*</tt>' Intrinsic</a><br>
7282 <a name="int_atomic_load_or">'<tt>llvm.atomic.load.or.*</tt>' Intrinsic</a><br>
7283 <a name="int_atomic_load_xor">'<tt>llvm.atomic.load.xor.*</tt>' Intrinsic</a><br>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7284</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7285
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7286<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7287
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7288<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7289<p>These are overloaded intrinsics. You can
7290 use <tt>llvm.atomic.load_and</tt>, <tt>llvm.atomic.load_nand</tt>,
7291 <tt>llvm.atomic.load_or</tt>, and <tt>llvm.atomic.load_xor</tt> on any integer
7292 bit width and for different address spaces. Not all targets support all bit
7293 widths however.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7294
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7295<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7296 declare i8 @llvm.atomic.load.and.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7297 declare i16 @llvm.atomic.load.and.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7298 declare i32 @llvm.atomic.load.and.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7299 declare i64 @llvm.atomic.load.and.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7300</pre>
7301
7302<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7303 declare i8 @llvm.atomic.load.or.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7304 declare i16 @llvm.atomic.load.or.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7305 declare i32 @llvm.atomic.load.or.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7306 declare i64 @llvm.atomic.load.or.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7307</pre>
7308
7309<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7310 declare i8 @llvm.atomic.load.nand.i8.p0i32(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7311 declare i16 @llvm.atomic.load.nand.i16.p0i32(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7312 declare i32 @llvm.atomic.load.nand.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7313 declare i64 @llvm.atomic.load.nand.i64.p0i32(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7314</pre>
7315
7316<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7317 declare i8 @llvm.atomic.load.xor.i8.p0i32(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7318 declare i16 @llvm.atomic.load.xor.i16.p0i32(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7319 declare i32 @llvm.atomic.load.xor.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7320 declare i64 @llvm.atomic.load.xor.i64.p0i32(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7321</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7322
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7323<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7324<p>These intrinsics bitwise the operation (and, nand, or, xor) <tt>delta</tt> to
7325 the value stored in memory at <tt>ptr</tt>. It yields the original value
7326 at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7327
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7328<h5>Arguments:</h5>
7329<p>These intrinsics take two arguments, the first a pointer to an integer value
7330 and the second an integer value. The result is also an integer value. These
7331 integer types can have any bit width, but they must all have the same bit
7332 width. The targets may only lower integer representations they support.</p>
7333
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7334<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7335<p>These intrinsics does a series of operations atomically. They first load the
7336 value stored at <tt>ptr</tt>. They then do the bitwise
7337 operation <tt>delta</tt>, store the result to <tt>ptr</tt>. They yield the
7338 original value stored at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7339
7340<h5>Examples:</h5>
7341<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7342%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7343%ptr = bitcast i8* %mallocP to i32*
7344 store i32 0x0F0F, %ptr
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7345%result0 = call i32 @llvm.atomic.load.nand.i32.p0i32(i32* %ptr, i32 0xFF)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7346 <i>; yields {i32}:result0 = 0x0F0F</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7347%result1 = call i32 @llvm.atomic.load.and.i32.p0i32(i32* %ptr, i32 0xFF)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7348 <i>; yields {i32}:result1 = 0xFFFFFFF0</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7349%result2 = call i32 @llvm.atomic.load.or.i32.p0i32(i32* %ptr, i32 0F)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7350 <i>; yields {i32}:result2 = 0xF0</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7351%result3 = call i32 @llvm.atomic.load.xor.i32.p0i32(i32* %ptr, i32 0F)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7352 <i>; yields {i32}:result3 = FF</i>
7353%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = F0</i>
7354</pre>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7355
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7356</div>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7357
7358<!-- _______________________________________________________________________ -->
7359<div class="doc_subsubsection">
7360 <a name="int_atomic_load_max">'<tt>llvm.atomic.load.max.*</tt>' Intrinsic</a><br>
7361 <a name="int_atomic_load_min">'<tt>llvm.atomic.load.min.*</tt>' Intrinsic</a><br>
7362 <a name="int_atomic_load_umax">'<tt>llvm.atomic.load.umax.*</tt>' Intrinsic</a><br>
7363 <a name="int_atomic_load_umin">'<tt>llvm.atomic.load.umin.*</tt>' Intrinsic</a><br>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7364</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7365
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7366<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7367
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7368<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7369<p>These are overloaded intrinsics. You can use <tt>llvm.atomic.load_max</tt>,
7370 <tt>llvm.atomic.load_min</tt>, <tt>llvm.atomic.load_umax</tt>, and
7371 <tt>llvm.atomic.load_umin</tt> on any integer bit width and for different
7372 address spaces. Not all targets support all bit widths however.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7373
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7374<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7375 declare i8 @llvm.atomic.load.max.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7376 declare i16 @llvm.atomic.load.max.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7377 declare i32 @llvm.atomic.load.max.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7378 declare i64 @llvm.atomic.load.max.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7379</pre>
7380
7381<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7382 declare i8 @llvm.atomic.load.min.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7383 declare i16 @llvm.atomic.load.min.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7384 declare i32 @llvm.atomic.load.min.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7385 declare i64 @llvm.atomic.load.min.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7386</pre>
7387
7388<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7389 declare i8 @llvm.atomic.load.umax.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7390 declare i16 @llvm.atomic.load.umax.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7391 declare i32 @llvm.atomic.load.umax.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7392 declare i64 @llvm.atomic.load.umax.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7393</pre>
7394
7395<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7396 declare i8 @llvm.atomic.load.umin.i8.p0i8(i8* &lt;ptr&gt;, i8 &lt;delta&gt;)
7397 declare i16 @llvm.atomic.load.umin.i16.p0i16(i16* &lt;ptr&gt;, i16 &lt;delta&gt;)
7398 declare i32 @llvm.atomic.load.umin.i32.p0i32(i32* &lt;ptr&gt;, i32 &lt;delta&gt;)
7399 declare i64 @llvm.atomic.load.umin.i64.p0i64(i64* &lt;ptr&gt;, i64 &lt;delta&gt;)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7400</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7401
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7402<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7403<p>These intrinsics takes the signed or unsigned minimum or maximum of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7404 <tt>delta</tt> and the value stored in memory at <tt>ptr</tt>. It yields the
7405 original value at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7406
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7407<h5>Arguments:</h5>
7408<p>These intrinsics take two arguments, the first a pointer to an integer value
7409 and the second an integer value. The result is also an integer value. These
7410 integer types can have any bit width, but they must all have the same bit
7411 width. The targets may only lower integer representations they support.</p>
7412
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7413<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7414<p>These intrinsics does a series of operations atomically. They first load the
7415 value stored at <tt>ptr</tt>. They then do the signed or unsigned min or
7416 max <tt>delta</tt> and the value, store the result to <tt>ptr</tt>. They
7417 yield the original value stored at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7418
7419<h5>Examples:</h5>
7420<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7421%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7422%ptr = bitcast i8* %mallocP to i32*
7423 store i32 7, %ptr
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7424%result0 = call i32 @llvm.atomic.load.min.i32.p0i32(i32* %ptr, i32 -2)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7425 <i>; yields {i32}:result0 = 7</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7426%result1 = call i32 @llvm.atomic.load.max.i32.p0i32(i32* %ptr, i32 8)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7427 <i>; yields {i32}:result1 = -2</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7428%result2 = call i32 @llvm.atomic.load.umin.i32.p0i32(i32* %ptr, i32 10)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7429 <i>; yields {i32}:result2 = 8</i>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7430%result3 = call i32 @llvm.atomic.load.umax.i32.p0i32(i32* %ptr, i32 30)
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7431 <i>; yields {i32}:result3 = 8</i>
7432%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 30</i>
7433</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7434
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7435</div>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7436
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7437
7438<!-- ======================================================================= -->
7439<div class="doc_subsection">
7440 <a name="int_memorymarkers">Memory Use Markers</a>
7441</div>
7442
7443<div class="doc_text">
7444
7445<p>This class of intrinsics exists to information about the lifetime of memory
7446 objects and ranges where variables are immutable.</p>
7447
7448</div>
7449
7450<!-- _______________________________________________________________________ -->
7451<div class="doc_subsubsection">
7452 <a name="int_lifetime_start">'<tt>llvm.lifetime.start</tt>' Intrinsic</a>
7453</div>
7454
7455<div class="doc_text">
7456
7457<h5>Syntax:</h5>
7458<pre>
7459 declare void @llvm.lifetime.start(i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
7460</pre>
7461
7462<h5>Overview:</h5>
7463<p>The '<tt>llvm.lifetime.start</tt>' intrinsic specifies the start of a memory
7464 object's lifetime.</p>
7465
7466<h5>Arguments:</h5>
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7467<p>The first argument is a constant integer representing the size of the
7468 object, or -1 if it is variable sized. The second argument is a pointer to
7469 the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7470
7471<h5>Semantics:</h5>
7472<p>This intrinsic indicates that before this point in the code, the value of the
7473 memory pointed to by <tt>ptr</tt> is dead. This means that it is known to
Nick Lewycky8d336592009-10-27 16:56:58 +0000[diff] [blame]7474 never be used and has an undefined value. A load from the pointer that
7475 precedes this intrinsic can be replaced with
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7476 <tt>'<a href="#undefvalues">undef</a>'</tt>.</p>
7477
7478</div>
7479
7480<!-- _______________________________________________________________________ -->
7481<div class="doc_subsubsection">
7482 <a name="int_lifetime_end">'<tt>llvm.lifetime.end</tt>' Intrinsic</a>
7483</div>
7484
7485<div class="doc_text">
7486
7487<h5>Syntax:</h5>
7488<pre>
7489 declare void @llvm.lifetime.end(i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
7490</pre>
7491
7492<h5>Overview:</h5>
7493<p>The '<tt>llvm.lifetime.end</tt>' intrinsic specifies the end of a memory
7494 object's lifetime.</p>
7495
7496<h5>Arguments:</h5>
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7497<p>The first argument is a constant integer representing the size of the
7498 object, or -1 if it is variable sized. The second argument is a pointer to
7499 the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7500
7501<h5>Semantics:</h5>
7502<p>This intrinsic indicates that after this point in the code, the value of the
7503 memory pointed to by <tt>ptr</tt> is dead. This means that it is known to
7504 never be used and has an undefined value. Any stores into the memory object
7505 following this intrinsic may be removed as dead.
7506
7507</div>
7508
7509<!-- _______________________________________________________________________ -->
7510<div class="doc_subsubsection">
7511 <a name="int_invariant_start">'<tt>llvm.invariant.start</tt>' Intrinsic</a>
7512</div>
7513
7514<div class="doc_text">
7515
7516<h5>Syntax:</h5>
7517<pre>
Nick Lewycky29b6cb42010-11-30 04:13:41 +0000[diff] [blame]7518 declare {}* @llvm.invariant.start(i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7519</pre>
7520
7521<h5>Overview:</h5>
7522<p>The '<tt>llvm.invariant.start</tt>' intrinsic specifies that the contents of
7523 a memory object will not change.</p>
7524
7525<h5>Arguments:</h5>
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7526<p>The first argument is a constant integer representing the size of the
7527 object, or -1 if it is variable sized. The second argument is a pointer to
7528 the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7529
7530<h5>Semantics:</h5>
7531<p>This intrinsic indicates that until an <tt>llvm.invariant.end</tt> that uses
7532 the return value, the referenced memory location is constant and
7533 unchanging.</p>
7534
7535</div>
7536
7537<!-- _______________________________________________________________________ -->
7538<div class="doc_subsubsection">
7539 <a name="int_invariant_end">'<tt>llvm.invariant.end</tt>' Intrinsic</a>
7540</div>
7541
7542<div class="doc_text">
7543
7544<h5>Syntax:</h5>
7545<pre>
7546 declare void @llvm.invariant.end({}* &lt;start&gt;, i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
7547</pre>
7548
7549<h5>Overview:</h5>
7550<p>The '<tt>llvm.invariant.end</tt>' intrinsic specifies that the contents of
7551 a memory object are mutable.</p>
7552
7553<h5>Arguments:</h5>
7554<p>The first argument is the matching <tt>llvm.invariant.start</tt> intrinsic.
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7555 The second argument is a constant integer representing the size of the
7556 object, or -1 if it is variable sized and the third argument is a pointer
7557 to the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7558
7559<h5>Semantics:</h5>
7560<p>This intrinsic indicates that the memory is mutable again.</p>
7561
7562</div>
7563
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7564<!-- ======================================================================= -->
7565<div class="doc_subsection">
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7566 <a name="int_general">General Intrinsics</a>
7567</div>
7568
7569<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7570
7571<p>This class of intrinsics is designed to be generic and has no specific
7572 purpose.</p>
7573
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7574</div>
7575
7576<!-- _______________________________________________________________________ -->
7577<div class="doc_subsubsection">
7578 <a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
7579</div>
7580
7581<div class="doc_text">
7582
7583<h5>Syntax:</h5>
7584<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7585 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]7586</pre>
7587
7588<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7589<p>The '<tt>llvm.var.annotation</tt>' intrinsic.</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7590
7591<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7592<p>The first argument is a pointer to a value, the second is a pointer to a
7593 global string, the third is a pointer to a global string which is the source
7594 file name, and the last argument is the line number.</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7595
7596<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7597<p>This intrinsic allows annotation of local variables with arbitrary strings.
7598 This can be useful for special purpose optimizations that want to look for
7599 these annotations. These have no other defined use, they are ignored by code
7600 generation and optimization.</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7601
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7602</div>
7603
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7604<!-- _______________________________________________________________________ -->
7605<div class="doc_subsubsection">
Tanya Lattnere1a8da02007-09-21 23:57:59 +0000[diff] [blame]7606 <a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7607</div>
7608
7609<div class="doc_text">
7610
7611<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7612<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
7613 any integer bit width.</p>
7614
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7615<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7616 declare i8 @llvm.annotation.i8(i8 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt;)
7617 declare i16 @llvm.annotation.i16(i16 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt;)
7618 declare i32 @llvm.annotation.i32(i32 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt;)
7619 declare i64 @llvm.annotation.i64(i64 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt;)
7620 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]7621</pre>
7622
7623<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7624<p>The '<tt>llvm.annotation</tt>' intrinsic.</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7625
7626<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7627<p>The first argument is an integer value (result of some expression), the
7628 second is a pointer to a global string, the third is a pointer to a global
7629 string which is the source file name, and the last argument is the line
7630 number. It returns the value of the first argument.</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7631
7632<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7633<p>This intrinsic allows annotations to be put on arbitrary expressions with
7634 arbitrary strings. This can be useful for special purpose optimizations that
7635 want to look for these annotations. These have no other defined use, they
7636 are ignored by code generation and optimization.</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7637
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7638</div>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]7639
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7640<!-- _______________________________________________________________________ -->
7641<div class="doc_subsubsection">
7642 <a name="int_trap">'<tt>llvm.trap</tt>' Intrinsic</a>
7643</div>
7644
7645<div class="doc_text">
7646
7647<h5>Syntax:</h5>
7648<pre>
7649 declare void @llvm.trap()
7650</pre>
7651
7652<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7653<p>The '<tt>llvm.trap</tt>' intrinsic.</p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7654
7655<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7656<p>None.</p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7657
7658<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7659<p>This intrinsics is lowered to the target dependent trap instruction. If the
7660 target does not have a trap instruction, this intrinsic will be lowered to
7661 the call of the <tt>abort()</tt> function.</p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7662
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7663</div>
7664
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7665<!-- _______________________________________________________________________ -->
7666<div class="doc_subsubsection">
Misha Brukmandccb0252008-11-22 23:55:29 +0000[diff] [blame]7667 <a name="int_stackprotector">'<tt>llvm.stackprotector</tt>' Intrinsic</a>
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7668</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7669
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7670<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7671
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7672<h5>Syntax:</h5>
7673<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7674 declare void @llvm.stackprotector(i8* &lt;guard&gt;, i8** &lt;slot&gt;)
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7675</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7676
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7677<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7678<p>The <tt>llvm.stackprotector</tt> intrinsic takes the <tt>guard</tt> and
7679 stores it onto the stack at <tt>slot</tt>. The stack slot is adjusted to
7680 ensure that it is placed on the stack before local variables.</p>
7681
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7682<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7683<p>The <tt>llvm.stackprotector</tt> intrinsic requires two pointer
7684 arguments. The first argument is the value loaded from the stack
7685 guard <tt>@__stack_chk_guard</tt>. The second variable is an <tt>alloca</tt>
7686 that has enough space to hold the value of the guard.</p>
7687
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7688<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7689<p>This intrinsic causes the prologue/epilogue inserter to force the position of
7690 the <tt>AllocaInst</tt> stack slot to be before local variables on the
7691 stack. This is to ensure that if a local variable on the stack is
7692 overwritten, it will destroy the value of the guard. When the function exits,
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]7693 the guard on the stack is checked against the original guard. If they are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7694 different, then the program aborts by calling the <tt>__stack_chk_fail()</tt>
7695 function.</p>
7696
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7697</div>
7698
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7699<!-- _______________________________________________________________________ -->
7700<div class="doc_subsubsection">
7701 <a name="int_objectsize">'<tt>llvm.objectsize</tt>' Intrinsic</a>
7702</div>
7703
7704<div class="doc_text">
7705
7706<h5>Syntax:</h5>
7707<pre>
Dan Gohman3dfb3cf2010-05-28 17:07:41 +0000[diff] [blame]7708 declare i32 @llvm.objectsize.i32(i8* &lt;object&gt;, i1 &lt;type&gt;)
7709 declare i64 @llvm.objectsize.i64(i8* &lt;object&gt;, i1 &lt;type&gt;)
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7710</pre>
7711
7712<h5>Overview:</h5>
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]7713<p>The <tt>llvm.objectsize</tt> intrinsic is designed to provide information to
7714 the optimizers to determine at compile time whether a) an operation (like
7715 memcpy) will overflow a buffer that corresponds to an object, or b) that a
7716 runtime check for overflow isn't necessary. An object in this context means
7717 an allocation of a specific class, structure, array, or other object.</p>
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7718
7719<h5>Arguments:</h5>
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]7720<p>The <tt>llvm.objectsize</tt> intrinsic takes two arguments. The first
Eric Christopher8295a0a2009-12-23 00:29:49 +0000[diff] [blame]7721 argument is a pointer to or into the <tt>object</tt>. The second argument
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]7722 is a boolean 0 or 1. This argument determines whether you want the
7723 maximum (0) or minimum (1) bytes remaining. This needs to be a literal 0 or
Eric Christopher8295a0a2009-12-23 00:29:49 +0000[diff] [blame]7724 1, variables are not allowed.</p>
7725
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7726<h5>Semantics:</h5>
7727<p>The <tt>llvm.objectsize</tt> intrinsic is lowered to either a constant
Bill Wendling1b383ba2010-10-27 01:07:41 +0000[diff] [blame]7728 representing the size of the object concerned, or <tt>i32/i64 -1 or 0</tt>,
7729 depending on the <tt>type</tt> argument, if the size cannot be determined at
7730 compile time.</p>
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7731
7732</div>
7733
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]7734<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]7735<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7736<address>
7737 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]7738 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7739 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]7740 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7741
7742 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencer05fe4b02006-03-14 05:39:39 +0000[diff] [blame]7743 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7744 Last modified: $Date$
7745</address>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]7746
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]7747</body>
7748</html>