Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 606a1d1f5984f20e79b0d2a2d38c121907426810 / . / docs / LangRef.html
blob: 9a563d670817e99bf0adfb53a7bf491b21c1dfc3 [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>
27 <li><a href="#linkage_internal">'<tt>internal</tt>' Linkage</a></li>
28 <li><a href="#linkage_available_externally">'<tt>available_externally</tt>' Linkage</a></li>
29 <li><a href="#linkage_linkonce">'<tt>linkonce</tt>' Linkage</a></li>
30 <li><a href="#linkage_common">'<tt>common</tt>' Linkage</a></li>
31 <li><a href="#linkage_weak">'<tt>weak</tt>' Linkage</a></li>
32 <li><a href="#linkage_appending">'<tt>appending</tt>' Linkage</a></li>
33 <li><a href="#linkage_externweak">'<tt>extern_weak</tt>' Linkage</a></li>
Chris Lattner5a2d8752009-10-10 18:26:06 +0000[diff] [blame]34 <li><a href="#linkage_linkonce_odr">'<tt>linkonce_odr</tt>' Linkage</a></li>
Bill Wendling987e7eb2009-07-20 02:41:50 +0000[diff] [blame]35 <li><a href="#linkage_weak">'<tt>weak_odr</tt>' Linkage</a></li>
36 <li><a href="#linkage_external">'<tt>externally visible</tt>' Linkage</a></li>
37 <li><a href="#linkage_dllimport">'<tt>dllimport</tt>' Linkage</a></li>
38 <li><a href="#linkage_dllexport">'<tt>dllexport</tt>' Linkage</a></li>
Bill Wendling3d10a5a2009-07-20 01:03:30 +0000[diff] [blame]39 </ol>
40 </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]41 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]42 <li><a href="#namedtypes">Named Types</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]43 <li><a href="#globalvars">Global Variables</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]44 <li><a href="#functionstructure">Functions</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]45 <li><a href="#aliasstructure">Aliases</a></li>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]46 <li><a href="#namedmetadatastructure">Named Metadata</a></li>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]47 <li><a href="#paramattrs">Parameter Attributes</a></li>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]48 <li><a href="#fnattrs">Function Attributes</a></li>
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]49 <li><a href="#gc">Garbage Collector Names</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]50 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]51 <li><a href="#datalayout">Data Layout</a></li>
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]52 <li><a href="#pointeraliasing">Pointer Aliasing Rules</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]53 </ol>
54 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]55 <li><a href="#typesystem">Type System</a>
56 <ol>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]57 <li><a href="#t_classifications">Type Classifications</a></li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]58 <li><a href="#t_primitive">Primitive Types</a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]59 <ol>
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]60 <li><a href="#t_integer">Integer Type</a></li>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]61 <li><a href="#t_floating">Floating Point Types</a></li>
62 <li><a href="#t_void">Void Type</a></li>
63 <li><a href="#t_label">Label Type</a></li>
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]64 <li><a href="#t_metadata">Metadata Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]65 </ol>
66 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]67 <li><a href="#t_derived">Derived Types</a>
68 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]69 <li><a href="#t_array">Array Type</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]70 <li><a href="#t_function">Function Type</a></li>
71 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]72 <li><a href="#t_struct">Structure Type</a></li>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]73 <li><a href="#t_pstruct">Packed Structure Type</a></li>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]74 <li><a href="#t_vector">Vector Type</a></li>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]75 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]76 </ol>
77 </li>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]78 <li><a href="#t_uprefs">Type Up-references</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]79 </ol>
80 </li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]81 <li><a href="#constants">Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]82 <ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]83 <li><a href="#simpleconstants">Simple Constants</a></li>
Chris Lattner70882792009-02-28 18:32:25 +0000[diff] [blame]84 <li><a href="#complexconstants">Complex Constants</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]85 <li><a href="#globalconstants">Global Variable and Function Addresses</a></li>
86 <li><a href="#undefvalues">Undefined Values</a></li>
Chris Lattnerf9d078e2009-10-27 21:19:13 +0000[diff] [blame]87 <li><a href="#blockaddress">Addresses of Basic Blocks</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]88 <li><a href="#constantexprs">Constant Expressions</a></li>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]89 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]90 </li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]91 <li><a href="#othervalues">Other Values</a>
92 <ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]93 <li><a href="#inlineasm">Inline Assembler Expressions</a></li>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]94 <li><a href="#metadata">Metadata Nodes and Metadata Strings</a></li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]95 </ol>
96 </li>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]97 <li><a href="#intrinsic_globals">Intrinsic Global Variables</a>
98 <ol>
99 <li><a href="#intg_used">The '<tt>llvm.used</tt>' Global Variable</a></li>
Chris Lattner401e10c2009-07-20 06:14:25 +0000[diff] [blame]100 <li><a href="#intg_compiler_used">The '<tt>llvm.compiler.used</tt>'
101 Global Variable</a></li>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]102 <li><a href="#intg_global_ctors">The '<tt>llvm.global_ctors</tt>'
103 Global Variable</a></li>
104 <li><a href="#intg_global_dtors">The '<tt>llvm.global_dtors</tt>'
105 Global Variable</a></li>
106 </ol>
107 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]108 <li><a href="#instref">Instruction Reference</a>
109 <ol>
110 <li><a href="#terminators">Terminator Instructions</a>
111 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]112 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
113 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]114 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]115 <li><a href="#i_indirectbr">'<tt>indirectbr</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]116 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]117 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]118 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]119 </ol>
120 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]121 <li><a href="#binaryops">Binary Operations</a>
122 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]123 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]124 <li><a href="#i_fadd">'<tt>fadd</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]125 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]126 <li><a href="#i_fsub">'<tt>fsub</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]127 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]128 <li><a href="#i_fmul">'<tt>fmul</tt>' Instruction</a></li>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]129 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
130 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
131 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]132 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
133 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
134 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]135 </ol>
136 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]137 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
138 <ol>
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]139 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
140 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
141 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]142 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]143 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]144 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]145 </ol>
146 </li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]147 <li><a href="#vectorops">Vector Operations</a>
148 <ol>
149 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
150 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
151 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]152 </ol>
153 </li>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]154 <li><a href="#aggregateops">Aggregate Operations</a>
155 <ol>
156 <li><a href="#i_extractvalue">'<tt>extractvalue</tt>' Instruction</a></li>
157 <li><a href="#i_insertvalue">'<tt>insertvalue</tt>' Instruction</a></li>
158 </ol>
159 </li>
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]160 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]161 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]162 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]163 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
164 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
165 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]166 </ol>
167 </li>
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]168 <li><a href="#convertops">Conversion Operations</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]169 <ol>
170 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
171 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
172 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
173 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
174 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]175 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
176 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
177 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
178 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]179 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
180 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]181 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]182 </ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]183 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]184 <li><a href="#otherops">Other Operations</a>
185 <ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]186 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
187 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]188 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]189 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]190 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]191 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]192 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]193 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]194 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]195 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]196 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]197 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]198 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
199 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]200 <li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
201 <li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
202 <li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]203 </ol>
204 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]205 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
206 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]207 <li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
208 <li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
209 <li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]210 </ol>
211 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]212 <li><a href="#int_codegen">Code Generator Intrinsics</a>
213 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]214 <li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
215 <li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
216 <li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
217 <li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
218 <li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
219 <li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
220 <li><a href="#int_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]221 </ol>
222 </li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]223 <li><a href="#int_libc">Standard C Library Intrinsics</a>
224 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]225 <li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
226 <li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
227 <li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
228 <li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
229 <li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]230 <li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
231 <li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
232 <li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]233 </ol>
234 </li>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]235 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]236 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]237 <li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]238 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
239 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
240 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]241 </ol>
242 </li>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]243 <li><a href="#int_overflow">Arithmetic with Overflow Intrinsics</a>
244 <ol>
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]245 <li><a href="#int_sadd_overflow">'<tt>llvm.sadd.with.overflow.*</tt> Intrinsics</a></li>
246 <li><a href="#int_uadd_overflow">'<tt>llvm.uadd.with.overflow.*</tt> Intrinsics</a></li>
247 <li><a href="#int_ssub_overflow">'<tt>llvm.ssub.with.overflow.*</tt> Intrinsics</a></li>
248 <li><a href="#int_usub_overflow">'<tt>llvm.usub.with.overflow.*</tt> Intrinsics</a></li>
249 <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]250 <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]251 </ol>
252 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]253 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]254 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]255 <li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]256 <ol>
257 <li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]258 </ol>
259 </li>
Bill Wendling3c44f5b2008-11-18 22:10:53 +0000[diff] [blame]260 <li><a href="#int_atomics">Atomic intrinsics</a>
261 <ol>
262 <li><a href="#int_memory_barrier"><tt>llvm.memory_barrier</tt></a></li>
263 <li><a href="#int_atomic_cmp_swap"><tt>llvm.atomic.cmp.swap</tt></a></li>
264 <li><a href="#int_atomic_swap"><tt>llvm.atomic.swap</tt></a></li>
265 <li><a href="#int_atomic_load_add"><tt>llvm.atomic.load.add</tt></a></li>
266 <li><a href="#int_atomic_load_sub"><tt>llvm.atomic.load.sub</tt></a></li>
267 <li><a href="#int_atomic_load_and"><tt>llvm.atomic.load.and</tt></a></li>
268 <li><a href="#int_atomic_load_nand"><tt>llvm.atomic.load.nand</tt></a></li>
269 <li><a href="#int_atomic_load_or"><tt>llvm.atomic.load.or</tt></a></li>
270 <li><a href="#int_atomic_load_xor"><tt>llvm.atomic.load.xor</tt></a></li>
271 <li><a href="#int_atomic_load_max"><tt>llvm.atomic.load.max</tt></a></li>
272 <li><a href="#int_atomic_load_min"><tt>llvm.atomic.load.min</tt></a></li>
273 <li><a href="#int_atomic_load_umax"><tt>llvm.atomic.load.umax</tt></a></li>
274 <li><a href="#int_atomic_load_umin"><tt>llvm.atomic.load.umin</tt></a></li>
275 </ol>
276 </li>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]277 <li><a href="#int_memorymarkers">Memory Use Markers</a>
278 <ol>
279 <li><a href="#int_lifetime_start"><tt>llvm.lifetime.start</tt></a></li>
280 <li><a href="#int_lifetime_end"><tt>llvm.lifetime.end</tt></a></li>
281 <li><a href="#int_invariant_start"><tt>llvm.invariant.start</tt></a></li>
282 <li><a href="#int_invariant_end"><tt>llvm.invariant.end</tt></a></li>
283 </ol>
284 </li>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]285 <li><a href="#int_general">General intrinsics</a>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]286 <ol>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]287 <li><a href="#int_var_annotation">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]288 '<tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]289 <li><a href="#int_annotation">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]290 '<tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]291 <li><a href="#int_trap">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]292 '<tt>llvm.trap</tt>' Intrinsic</a></li>
293 <li><a href="#int_stackprotector">
294 '<tt>llvm.stackprotector</tt>' Intrinsic</a></li>
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]295 <li><a href="#int_objectsize">
296 '<tt>llvm.objectsize</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]297 </ol>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]298 </li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]299 </ol>
300 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]301</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]302
303<div class="doc_author">
304 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
305 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]306</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]307
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]308<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]309<div class="doc_section"> <a name="abstract">Abstract </a></div>
310<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]311
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]312<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]313
314<p>This document is a reference manual for the LLVM assembly language. LLVM is
315 a Static Single Assignment (SSA) based representation that provides type
316 safety, low-level operations, flexibility, and the capability of representing
317 'all' high-level languages cleanly. It is the common code representation
318 used throughout all phases of the LLVM compilation strategy.</p>
319
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]320</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]321
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]322<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]323<div class="doc_section"> <a name="introduction">Introduction</a> </div>
324<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]325
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]326<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]327
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]328<p>The LLVM code representation is designed to be used in three different forms:
329 as an in-memory compiler IR, as an on-disk bitcode representation (suitable
330 for fast loading by a Just-In-Time compiler), and as a human readable
331 assembly language representation. This allows LLVM to provide a powerful
332 intermediate representation for efficient compiler transformations and
333 analysis, while providing a natural means to debug and visualize the
334 transformations. The three different forms of LLVM are all equivalent. This
335 document describes the human readable representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]336
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]337<p>The LLVM representation aims to be light-weight and low-level while being
338 expressive, typed, and extensible at the same time. It aims to be a
339 "universal IR" of sorts, by being at a low enough level that high-level ideas
340 may be cleanly mapped to it (similar to how microprocessors are "universal
341 IR's", allowing many source languages to be mapped to them). By providing
342 type information, LLVM can be used as the target of optimizations: for
343 example, through pointer analysis, it can be proven that a C automatic
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]344 variable is never accessed outside of the current function, allowing it to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]345 be promoted to a simple SSA value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]346
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]347</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]348
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]349<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]350<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]351
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]352<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]353
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]354<p>It is important to note that this document describes 'well formed' LLVM
355 assembly language. There is a difference between what the parser accepts and
356 what is considered 'well formed'. For example, the following instruction is
357 syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]358
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]359<div class="doc_code">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]360<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]361%x = <a href="#i_add">add</a> i32 1, %x
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]362</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]363</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]364
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]365<p>because the definition of <tt>%x</tt> does not dominate all of its uses. The
366 LLVM infrastructure provides a verification pass that may be used to verify
367 that an LLVM module is well formed. This pass is automatically run by the
368 parser after parsing input assembly and by the optimizer before it outputs
369 bitcode. The violations pointed out by the verifier pass indicate bugs in
370 transformation passes or input to the parser.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]371
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]372</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]373
Chris Lattnercc689392007-10-03 17:34:29 +0000[diff] [blame]374<!-- Describe the typesetting conventions here. -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]375
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]376<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]377<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]378<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]379
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]380<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]381
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]382<p>LLVM identifiers come in two basic types: global and local. Global
383 identifiers (functions, global variables) begin with the <tt>'@'</tt>
384 character. Local identifiers (register names, types) begin with
385 the <tt>'%'</tt> character. Additionally, there are three different formats
386 for identifiers, for different purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]387
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]388<ol>
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]389 <li>Named values are represented as a string of characters with their prefix.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]390 For example, <tt>%foo</tt>, <tt>@DivisionByZero</tt>,
391 <tt>%a.really.long.identifier</tt>. The actual regular expression used is
392 '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'. Identifiers which require
393 other characters in their names can be surrounded with quotes. Special
394 characters may be escaped using <tt>"\xx"</tt> where <tt>xx</tt> is the
395 ASCII code for the character in hexadecimal. In this way, any character
396 can be used in a name value, even quotes themselves.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]397
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]398 <li>Unnamed values are represented as an unsigned numeric value with their
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]399 prefix. For example, <tt>%12</tt>, <tt>@2</tt>, <tt>%44</tt>.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]400
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]401 <li>Constants, which are described in a <a href="#constants">section about
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]402 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]403</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]404
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]405<p>LLVM requires that values start with a prefix for two reasons: Compilers
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]406 don't need to worry about name clashes with reserved words, and the set of
407 reserved words may be expanded in the future without penalty. Additionally,
408 unnamed identifiers allow a compiler to quickly come up with a temporary
409 variable without having to avoid symbol table conflicts.</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]410
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]411<p>Reserved words in LLVM are very similar to reserved words in other
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]412 languages. There are keywords for different opcodes
413 ('<tt><a href="#i_add">add</a></tt>',
414 '<tt><a href="#i_bitcast">bitcast</a></tt>',
415 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names
416 ('<tt><a href="#t_void">void</a></tt>',
417 '<tt><a href="#t_primitive">i32</a></tt>', etc...), and others. These
418 reserved words cannot conflict with variable names, because none of them
419 start with a prefix character (<tt>'%'</tt> or <tt>'@'</tt>).</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]420
421<p>Here is an example of LLVM code to multiply the integer variable
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]422 '<tt>%X</tt>' by 8:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]423
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]424<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]425
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]426<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]427<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]428%result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]429</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]430</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]431
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]432<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]433
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]434<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]435<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]436%result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]437</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]438</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]439
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]440<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]441
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]442<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]443<pre>
Gabor Greifec58f752009-10-28 13:05:07 +0000[diff] [blame]444%0 = <a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
445%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]446%result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]447</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]448</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]449
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]450<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several important
451 lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]452
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]453<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]454 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]455 line.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]456
457 <li>Unnamed temporaries are created when the result of a computation is not
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]458 assigned to a named value.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]459
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]460 <li>Unnamed temporaries are numbered sequentially</li>
461</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]462
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]463<p>It also shows a convention that we follow in this document. When
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]464 demonstrating instructions, we will follow an instruction with a comment that
465 defines the type and name of value produced. Comments are shown in italic
466 text.</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]467
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]468</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]469
470<!-- *********************************************************************** -->
471<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
472<!-- *********************************************************************** -->
473
474<!-- ======================================================================= -->
475<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
476</div>
477
478<div class="doc_text">
479
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]480<p>LLVM programs are composed of "Module"s, each of which is a translation unit
481 of the input programs. Each module consists of functions, global variables,
482 and symbol table entries. Modules may be combined together with the LLVM
483 linker, which merges function (and global variable) definitions, resolves
484 forward declarations, and merges symbol table entries. Here is an example of
485 the "hello world" module:</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]486
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]487<div class="doc_code">
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]488<pre>
489<i>; Declare the string constant as a global constant.</i>
490<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>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]491
492<i>; External declaration of the puts function</i>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]493<a href="#functionstructure">declare</a> i32 @puts(i8 *) <i>; i32(i8 *)* </i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]494
495<i>; Definition of main function</i>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]496define i32 @main() { <i>; i32()* </i>
497 <i>; Convert [13 x i8]* to i8 *...</i>
498 %cast210 = <a href="#i_getelementptr">getelementptr</a> [13 x i8]* @.LC0, i64 0, i64 0 <i>; i8 *</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]499
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]500 <i>; Call puts function to write out the string to stdout.</i>
501 <a href="#i_call">call</a> i32 @puts(i8 * %cast210) <i>; i32</i>
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]502 <a href="#i_ret">ret</a> i32 0<br>}
503
504<i>; Named metadata</i>
505!1 = metadata !{i32 41}
506!foo = !{!1, null}
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]507</pre>
508</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]509
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]510<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]511 "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>" function,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]512 a <a href="#functionstructure">function definition</a> for
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]513 "<tt>main</tt>" and <a href="#namedmetadatastructure">named metadata</a>
514 "<tt>foo"</tt>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]515
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]516<p>In general, a module is made up of a list of global values, where both
517 functions and global variables are global values. Global values are
518 represented by a pointer to a memory location (in this case, a pointer to an
519 array of char, and a pointer to a function), and have one of the
520 following <a href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]521
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]522</div>
523
524<!-- ======================================================================= -->
525<div class="doc_subsection">
526 <a name="linkage">Linkage Types</a>
527</div>
528
529<div class="doc_text">
530
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]531<p>All Global Variables and Functions have one of the following types of
532 linkage:</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]533
534<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]535 <dt><tt><b><a name="linkage_private">private</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]536 <dd>Global values with private linkage are only directly accessible by objects
537 in the current module. In particular, linking code into a module with an
538 private global value may cause the private to be renamed as necessary to
539 avoid collisions. Because the symbol is private to the module, all
540 references can be updated. This doesn't show up in any symbol table in the
541 object file.</dd>
Rafael Espindolabb46f522009-01-15 20:18:42 +0000[diff] [blame]542
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]543 <dt><tt><b><a name="linkage_linker_private">linker_private</a></b></tt></dt>
Bill Wendling3d10a5a2009-07-20 01:03:30 +0000[diff] [blame]544 <dd>Similar to private, but the symbol is passed through the assembler and
Chris Lattnere1eaf912009-08-24 04:32:16 +0000[diff] [blame]545 removed by the linker after evaluation. Note that (unlike private
546 symbols) linker_private symbols are subject to coalescing by the linker:
547 weak symbols get merged and redefinitions are rejected. However, unlike
548 normal strong symbols, they are removed by the linker from the final
549 linked image (executable or dynamic library).</dd>
Bill Wendling3d10a5a2009-07-20 01:03:30 +0000[diff] [blame]550
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]551 <dt><tt><b><a name="linkage_internal">internal</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]552 <dd>Similar to private, but the value shows as a local symbol
553 (<tt>STB_LOCAL</tt> in the case of ELF) in the object file. This
554 corresponds to the notion of the '<tt>static</tt>' keyword in C.</dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]555
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]556 <dt><tt><b><a name="linkage_available_externally">available_externally</a></b></tt></dt>
Chris Lattner266c7bb2009-04-13 05:44:34 +0000[diff] [blame]557 <dd>Globals with "<tt>available_externally</tt>" linkage are never emitted
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]558 into the object file corresponding to the LLVM module. They exist to
559 allow inlining and other optimizations to take place given knowledge of
560 the definition of the global, which is known to be somewhere outside the
561 module. Globals with <tt>available_externally</tt> linkage are allowed to
562 be discarded at will, and are otherwise the same as <tt>linkonce_odr</tt>.
563 This linkage type is only allowed on definitions, not declarations.</dd>
Chris Lattner266c7bb2009-04-13 05:44:34 +0000[diff] [blame]564
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]565 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt></dt>
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]566 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
Chris Lattner873187c2010-01-09 19:15:14 +0000[diff] [blame]567 the same name when linkage occurs. This can be used to implement
568 some forms of inline functions, templates, or other code which must be
569 generated in each translation unit that uses it, but where the body may
570 be overridden with a more definitive definition later. Unreferenced
571 <tt>linkonce</tt> globals are allowed to be discarded. Note that
572 <tt>linkonce</tt> linkage does not actually allow the optimizer to
573 inline the body of this function into callers because it doesn't know if
574 this definition of the function is the definitive definition within the
575 program or whether it will be overridden by a stronger definition.
576 To enable inlining and other optimizations, use "<tt>linkonce_odr</tt>"
577 linkage.</dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]578
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]579 <dt><tt><b><a name="linkage_weak">weak</a></b></tt></dt>
Chris Lattner26d054d2009-08-05 05:21:07 +0000[diff] [blame]580 <dd>"<tt>weak</tt>" linkage has the same merging semantics as
581 <tt>linkonce</tt> linkage, except that unreferenced globals with
582 <tt>weak</tt> linkage may not be discarded. This is used for globals that
583 are declared "weak" in C source code.</dd>
584
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]585 <dt><tt><b><a name="linkage_common">common</a></b></tt></dt>
Chris Lattner26d054d2009-08-05 05:21:07 +0000[diff] [blame]586 <dd>"<tt>common</tt>" linkage is most similar to "<tt>weak</tt>" linkage, but
587 they are used for tentative definitions in C, such as "<tt>int X;</tt>" at
588 global scope.
589 Symbols with "<tt>common</tt>" linkage are merged in the same way as
590 <tt>weak symbols</tt>, and they may not be deleted if unreferenced.
Chris Lattnercd81f5d2009-08-05 05:41:44 +0000[diff] [blame]591 <tt>common</tt> symbols may not have an explicit section,
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]592 must have a zero initializer, and may not be marked '<a
Chris Lattnercd81f5d2009-08-05 05:41:44 +0000[diff] [blame]593 href="#globalvars"><tt>constant</tt></a>'. Functions and aliases may not
594 have common linkage.</dd>
Chris Lattner26d054d2009-08-05 05:21:07 +0000[diff] [blame]595
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]596
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]597 <dt><tt><b><a name="linkage_appending">appending</a></b></tt></dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]598 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]599 pointer to array type. When two global variables with appending linkage
600 are linked together, the two global arrays are appended together. This is
601 the LLVM, typesafe, equivalent of having the system linker append together
602 "sections" with identical names when .o files are linked.</dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]603
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]604 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]605 <dd>The semantics of this linkage follow the ELF object file model: the symbol
606 is weak until linked, if not linked, the symbol becomes null instead of
607 being an undefined reference.</dd>
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]608
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]609 <dt><tt><b><a name="linkage_linkonce_odr">linkonce_odr</a></b></tt></dt>
610 <dt><tt><b><a name="linkage_weak_odr">weak_odr</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]611 <dd>Some languages allow differing globals to be merged, such as two functions
612 with different semantics. Other languages, such as <tt>C++</tt>, ensure
613 that only equivalent globals are ever merged (the "one definition rule" -
614 "ODR"). Such languages can use the <tt>linkonce_odr</tt>
615 and <tt>weak_odr</tt> linkage types to indicate that the global will only
616 be merged with equivalent globals. These linkage types are otherwise the
617 same as their non-<tt>odr</tt> versions.</dd>
Duncan Sands667d4b82009-03-07 15:45:40 +0000[diff] [blame]618
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]619 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]620 <dd>If none of the above identifiers are used, the global is externally
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]621 visible, meaning that it participates in linkage and can be used to
622 resolve external symbol references.</dd>
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]623</dl>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]624
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]625<p>The next two types of linkage are targeted for Microsoft Windows platform
626 only. They are designed to support importing (exporting) symbols from (to)
627 DLLs (Dynamic Link Libraries).</p>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]628
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]629<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]630 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt></dt>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]631 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]632 or variable via a global pointer to a pointer that is set up by the DLL
633 exporting the symbol. On Microsoft Windows targets, the pointer name is
634 formed by combining <code>__imp_</code> and the function or variable
635 name.</dd>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]636
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]637 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt></dt>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]638 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]639 pointer to a pointer in a DLL, so that it can be referenced with the
640 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
641 name is formed by combining <code>__imp_</code> and the function or
642 variable name.</dd>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]643</dl>
644
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]645<p>For example, since the "<tt>.LC0</tt>" variable is defined to be internal, if
646 another module defined a "<tt>.LC0</tt>" variable and was linked with this
647 one, one of the two would be renamed, preventing a collision. Since
648 "<tt>main</tt>" and "<tt>puts</tt>" are external (i.e., lacking any linkage
649 declarations), they are accessible outside of the current module.</p>
650
651<p>It is illegal for a function <i>declaration</i> to have any linkage type
652 other than "externally visible", <tt>dllimport</tt>
653 or <tt>extern_weak</tt>.</p>
654
Duncan Sands667d4b82009-03-07 15:45:40 +0000[diff] [blame]655<p>Aliases can have only <tt>external</tt>, <tt>internal</tt>, <tt>weak</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]656 or <tt>weak_odr</tt> linkages.</p>
657
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]658</div>
659
660<!-- ======================================================================= -->
661<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]662 <a name="callingconv">Calling Conventions</a>
663</div>
664
665<div class="doc_text">
666
667<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]668 and <a href="#i_invoke">invokes</a> can all have an optional calling
669 convention specified for the call. The calling convention of any pair of
670 dynamic caller/callee must match, or the behavior of the program is
671 undefined. The following calling conventions are supported by LLVM, and more
672 may be added in the future:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]673
674<dl>
675 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]676 <dd>This calling convention (the default if no other calling convention is
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]677 specified) matches the target C calling conventions. This calling
678 convention supports varargs function calls and tolerates some mismatch in
679 the declared prototype and implemented declaration of the function (as
680 does normal C).</dd>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]681
682 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]683 <dd>This calling convention attempts to make calls as fast as possible
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]684 (e.g. by passing things in registers). This calling convention allows the
685 target to use whatever tricks it wants to produce fast code for the
686 target, without having to conform to an externally specified ABI
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]687 (Application Binary Interface).
688 <a href="CodeGenerator.html#tailcallopt">Tail calls can only be optimized
689 when this convention is used.</a> This calling convention does not
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]690 support varargs and requires the prototype of all callees to exactly match
691 the prototype of the function definition.</dd>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]692
693 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]694 <dd>This calling convention attempts to make code in the caller as efficient
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]695 as possible under the assumption that the call is not commonly executed.
696 As such, these calls often preserve all registers so that the call does
697 not break any live ranges in the caller side. This calling convention
698 does not support varargs and requires the prototype of all callees to
699 exactly match the prototype of the function definition.</dd>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]700
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]701 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]702 <dd>Any calling convention may be specified by number, allowing
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]703 target-specific calling conventions to be used. Target specific calling
704 conventions start at 64.</dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]705</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]706
707<p>More calling conventions can be added/defined on an as-needed basis, to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]708 support Pascal conventions or any other well-known target-independent
709 convention.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]710
711</div>
712
713<!-- ======================================================================= -->
714<div class="doc_subsection">
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]715 <a name="visibility">Visibility Styles</a>
716</div>
717
718<div class="doc_text">
719
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]720<p>All Global Variables and Functions have one of the following visibility
721 styles:</p>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]722
723<dl>
724 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]725 <dd>On targets that use the ELF object file format, default visibility means
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]726 that the declaration is visible to other modules and, in shared libraries,
727 means that the declared entity may be overridden. On Darwin, default
728 visibility means that the declaration is visible to other modules. Default
729 visibility corresponds to "external linkage" in the language.</dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]730
731 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]732 <dd>Two declarations of an object with hidden visibility refer to the same
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]733 object if they are in the same shared object. Usually, hidden visibility
734 indicates that the symbol will not be placed into the dynamic symbol
735 table, so no other module (executable or shared library) can reference it
736 directly.</dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]737
Anton Korobeynikov6f9896f2007-04-29 18:35:00 +0000[diff] [blame]738 <dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
Anton Korobeynikov6f9896f2007-04-29 18:35:00 +0000[diff] [blame]739 <dd>On ELF, protected visibility indicates that the symbol will be placed in
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]740 the dynamic symbol table, but that references within the defining module
741 will bind to the local symbol. That is, the symbol cannot be overridden by
742 another module.</dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]743</dl>
744
745</div>
746
747<!-- ======================================================================= -->
748<div class="doc_subsection">
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]749 <a name="namedtypes">Named Types</a>
750</div>
751
752<div class="doc_text">
753
754<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]755 it easier to read the IR and make the IR more condensed (particularly when
756 recursive types are involved). An example of a name specification is:</p>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]757
758<div class="doc_code">
759<pre>
760%mytype = type { %mytype*, i32 }
761</pre>
762</div>
763
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]764<p>You may give a name to any <a href="#typesystem">type</a> except
765 "<a href="t_void">void</a>". Type name aliases may be used anywhere a type
766 is expected with the syntax "%mytype".</p>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]767
768<p>Note that type names are aliases for the structural type that they indicate,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]769 and that you can therefore specify multiple names for the same type. This
770 often leads to confusing behavior when dumping out a .ll file. Since LLVM IR
771 uses structural typing, the name is not part of the type. When printing out
772 LLVM IR, the printer will pick <em>one name</em> to render all types of a
773 particular shape. This means that if you have code where two different
774 source types end up having the same LLVM type, that the dumper will sometimes
775 print the "wrong" or unexpected type. This is an important design point and
776 isn't going to change.</p>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]777
778</div>
779
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]780<!-- ======================================================================= -->
781<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]782 <a name="globalvars">Global Variables</a>
783</div>
784
785<div class="doc_text">
786
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]787<p>Global variables define regions of memory allocated at compilation time
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]788 instead of run-time. Global variables may optionally be initialized, may
789 have an explicit section to be placed in, and may have an optional explicit
790 alignment specified. A variable may be defined as "thread_local", which
791 means that it will not be shared by threads (each thread will have a
792 separated copy of the variable). A variable may be defined as a global
793 "constant," which indicates that the contents of the variable
794 will <b>never</b> be modified (enabling better optimization, allowing the
795 global data to be placed in the read-only section of an executable, etc).
796 Note that variables that need runtime initialization cannot be marked
797 "constant" as there is a store to the variable.</p>
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]798
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]799<p>LLVM explicitly allows <em>declarations</em> of global variables to be marked
800 constant, even if the final definition of the global is not. This capability
801 can be used to enable slightly better optimization of the program, but
802 requires the language definition to guarantee that optimizations based on the
803 'constantness' are valid for the translation units that do not include the
804 definition.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]805
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]806<p>As SSA values, global variables define pointer values that are in scope
807 (i.e. they dominate) all basic blocks in the program. Global variables
808 always define a pointer to their "content" type because they describe a
809 region of memory, and all memory objects in LLVM are accessed through
810 pointers.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]811
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]812<p>A global variable may be declared to reside in a target-specific numbered
813 address space. For targets that support them, address spaces may affect how
814 optimizations are performed and/or what target instructions are used to
815 access the variable. The default address space is zero. The address space
816 qualifier must precede any other attributes.</p>
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]817
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]818<p>LLVM allows an explicit section to be specified for globals. If the target
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]819 supports it, it will emit globals to the section specified.</p>
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]820
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]821<p>An explicit alignment may be specified for a global. If not present, or if
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]822 the alignment is set to zero, the alignment of the global is set by the
823 target to whatever it feels convenient. If an explicit alignment is
824 specified, the global is forced to have at least that much alignment. All
825 alignments must be a power of 2.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]826
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]827<p>For example, the following defines a global in a numbered address space with
828 an initializer, section, and alignment:</p>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]829
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]830<div class="doc_code">
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]831<pre>
Dan Gohman398873c2009-01-11 00:40:00 +0000[diff] [blame]832@G = addrspace(5) constant float 1.0, section "foo", align 4
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]833</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]834</div>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]835
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]836</div>
837
838
839<!-- ======================================================================= -->
840<div class="doc_subsection">
841 <a name="functionstructure">Functions</a>
842</div>
843
844<div class="doc_text">
845
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]846<p>LLVM function definitions consist of the "<tt>define</tt>" keyord, an
847 optional <a href="#linkage">linkage type</a>, an optional
848 <a href="#visibility">visibility style</a>, an optional
849 <a href="#callingconv">calling convention</a>, a return type, an optional
850 <a href="#paramattrs">parameter attribute</a> for the return type, a function
851 name, a (possibly empty) argument list (each with optional
852 <a href="#paramattrs">parameter attributes</a>), optional
853 <a href="#fnattrs">function attributes</a>, an optional section, an optional
854 alignment, an optional <a href="#gc">garbage collector name</a>, an opening
855 curly brace, a list of basic blocks, and a closing curly brace.</p>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]856
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]857<p>LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
858 optional <a href="#linkage">linkage type</a>, an optional
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]859 <a href="#visibility">visibility style</a>, an optional
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]860 <a href="#callingconv">calling convention</a>, a return type, an optional
861 <a href="#paramattrs">parameter attribute</a> for the return type, a function
862 name, a possibly empty list of arguments, an optional alignment, and an
863 optional <a href="#gc">garbage collector name</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]864
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]865<p>A function definition contains a list of basic blocks, forming the CFG
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]866 (Control Flow Graph) for the function. Each basic block may optionally start
867 with a label (giving the basic block a symbol table entry), contains a list
868 of instructions, and ends with a <a href="#terminators">terminator</a>
869 instruction (such as a branch or function return).</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]870
Chris Lattner4a3c9012007-06-08 16:52:14 +0000[diff] [blame]871<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]872 executed on entrance to the function, and it is not allowed to have
873 predecessor basic blocks (i.e. there can not be any branches to the entry
874 block of a function). Because the block can have no predecessors, it also
875 cannot have any <a href="#i_phi">PHI nodes</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]876
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]877<p>LLVM allows an explicit section to be specified for functions. If the target
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]878 supports it, it will emit functions to the section specified.</p>
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]879
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]880<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]881 the alignment is set to zero, the alignment of the function is set by the
882 target to whatever it feels convenient. If an explicit alignment is
883 specified, the function is forced to have at least that much alignment. All
884 alignments must be a power of 2.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]885
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]886<h5>Syntax:</h5>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]887<div class="doc_code">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]888<pre>
Chris Lattner50ad45c2008-10-13 16:55:18 +0000[diff] [blame]889define [<a href="#linkage">linkage</a>] [<a href="#visibility">visibility</a>]
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]890 [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>]
891 &lt;ResultType&gt; @&lt;FunctionName&gt; ([argument list])
892 [<a href="#fnattrs">fn Attrs</a>] [section "name"] [align N]
893 [<a href="#gc">gc</a>] { ... }
894</pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]895</div>
896
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]897</div>
898
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]899<!-- ======================================================================= -->
900<div class="doc_subsection">
901 <a name="aliasstructure">Aliases</a>
902</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]903
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]904<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]905
906<p>Aliases act as "second name" for the aliasee value (which can be either
907 function, global variable, another alias or bitcast of global value). Aliases
908 may have an optional <a href="#linkage">linkage type</a>, and an
909 optional <a href="#visibility">visibility style</a>.</p>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]910
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]911<h5>Syntax:</h5>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]912<div class="doc_code">
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]913<pre>
Duncan Sands0b23ac12008-09-12 20:48:21 +0000[diff] [blame]914@&lt;Name&gt; = alias [Linkage] [Visibility] &lt;AliaseeTy&gt; @&lt;Aliasee&gt;
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]915</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]916</div>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]917
918</div>
919
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]920<!-- ======================================================================= -->
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]921<div class="doc_subsection">
922 <a name="namedmetadatastructure">Named Metadata</a>
923</div>
924
925<div class="doc_text">
926
927<p>Named metadata is a collection of metadata. <a href="#metadata"> Metadata </a>
928 node and null are the only valid named metadata operands.
929 Metadata strings are not allowed as an named metadata operand.</p>
930
931<h5>Syntax:</h5>
932<div class="doc_code">
933<pre>
934!1 = metadata !{metadata !"one"}
935!name = !{null, !1}
936</pre>
937</div>
938
939</div>
940
941<!-- ======================================================================= -->
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]942<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]943
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]944<div class="doc_text">
945
946<p>The return type and each parameter of a function type may have a set of
947 <i>parameter attributes</i> associated with them. Parameter attributes are
948 used to communicate additional information about the result or parameters of
949 a function. Parameter attributes are considered to be part of the function,
950 not of the function type, so functions with different parameter attributes
951 can have the same function type.</p>
952
953<p>Parameter attributes are simple keywords that follow the type specified. If
954 multiple parameter attributes are needed, they are space separated. For
955 example:</p>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]956
957<div class="doc_code">
958<pre>
Nick Lewyckyb6a7d252009-02-15 23:06:14 +0000[diff] [blame]959declare i32 @printf(i8* noalias nocapture, ...)
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]960declare i32 @atoi(i8 zeroext)
961declare signext i8 @returns_signed_char()
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]962</pre>
963</div>
964
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]965<p>Note that any attributes for the function result (<tt>nounwind</tt>,
966 <tt>readonly</tt>) come immediately after the argument list.</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]967
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]968<p>Currently, only the following parameter attributes are defined:</p>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]969
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]970<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]971 <dt><tt><b>zeroext</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]972 <dd>This indicates to the code generator that the parameter or return value
973 should be zero-extended to a 32-bit value by the caller (for a parameter)
974 or the callee (for a return value).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]975
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]976 <dt><tt><b>signext</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]977 <dd>This indicates to the code generator that the parameter or return value
978 should be sign-extended to a 32-bit value by the caller (for a parameter)
979 or the callee (for a return value).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]980
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]981 <dt><tt><b>inreg</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]982 <dd>This indicates that this parameter or return value should be treated in a
983 special target-dependent fashion during while emitting code for a function
984 call or return (usually, by putting it in a register as opposed to memory,
985 though some targets use it to distinguish between two different kinds of
986 registers). Use of this attribute is target-specific.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]987
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]988 <dt><tt><b><a name="byval">byval</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]989 <dd>This indicates that the pointer parameter should really be passed by value
990 to the function. The attribute implies that a hidden copy of the pointee
991 is made between the caller and the callee, so the callee is unable to
992 modify the value in the callee. This attribute is only valid on LLVM
993 pointer arguments. It is generally used to pass structs and arrays by
994 value, but is also valid on pointers to scalars. The copy is considered
995 to belong to the caller not the callee (for example,
996 <tt><a href="#readonly">readonly</a></tt> functions should not write to
997 <tt>byval</tt> parameters). This is not a valid attribute for return
998 values. The byval attribute also supports specifying an alignment with
999 the align attribute. This has a target-specific effect on the code
1000 generator that usually indicates a desired alignment for the synthesized
1001 stack slot.</dd>
1002
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1003 <dt><tt><b>sret</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1004 <dd>This indicates that the pointer parameter specifies the address of a
1005 structure that is the return value of the function in the source program.
1006 This pointer must be guaranteed by the caller to be valid: loads and
1007 stores to the structure may be assumed by the callee to not to trap. This
1008 may only be applied to the first parameter. This is not a valid attribute
1009 for return values. </dd>
1010
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1011 <dt><tt><b>noalias</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1012 <dd>This indicates that the pointer does not alias any global or any other
1013 parameter. The caller is responsible for ensuring that this is the
1014 case. On a function return value, <tt>noalias</tt> additionally indicates
1015 that the pointer does not alias any other pointers visible to the
1016 caller. For further details, please see the discussion of the NoAlias
1017 response in
1018 <a href="http://llvm.org/docs/AliasAnalysis.html#MustMayNo">alias
1019 analysis</a>.</dd>
1020
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1021 <dt><tt><b>nocapture</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1022 <dd>This indicates that the callee does not make any copies of the pointer
1023 that outlive the callee itself. This is not a valid attribute for return
1024 values.</dd>
1025
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1026 <dt><tt><b>nest</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1027 <dd>This indicates that the pointer parameter can be excised using the
1028 <a href="#int_trampoline">trampoline intrinsics</a>. This is not a valid
1029 attribute for return values.</dd>
1030</dl>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1031
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1032</div>
1033
1034<!-- ======================================================================= -->
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1035<div class="doc_subsection">
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1036 <a name="gc">Garbage Collector Names</a>
1037</div>
1038
1039<div class="doc_text">
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1040
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1041<p>Each function may specify a garbage collector name, which is simply a
1042 string:</p>
1043
1044<div class="doc_code">
1045<pre>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1046define void @f() gc "name" { ... }
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1047</pre>
1048</div>
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1049
1050<p>The compiler declares the supported values of <i>name</i>. Specifying a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1051 collector which will cause the compiler to alter its output in order to
1052 support the named garbage collection algorithm.</p>
1053
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]1054</div>
1055
1056<!-- ======================================================================= -->
1057<div class="doc_subsection">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1058 <a name="fnattrs">Function Attributes</a>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1059</div>
1060
1061<div class="doc_text">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1062
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1063<p>Function attributes are set to communicate additional information about a
1064 function. Function attributes are considered to be part of the function, not
1065 of the function type, so functions with different parameter attributes can
1066 have the same function type.</p>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1067
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1068<p>Function attributes are simple keywords that follow the type specified. If
1069 multiple attributes are needed, they are space separated. For example:</p>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1070
1071<div class="doc_code">
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1072<pre>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1073define void @f() noinline { ... }
1074define void @f() alwaysinline { ... }
1075define void @f() alwaysinline optsize { ... }
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1076define void @f() optsize { ... }
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1077</pre>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1078</div>
1079
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1080<dl>
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1081 <dt><tt><b>alwaysinline</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1082 <dd>This attribute indicates that the inliner should attempt to inline this
1083 function into callers whenever possible, ignoring any active inlining size
1084 threshold for this caller.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1085
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1086 <dt><tt><b>noinline</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1087 <dd>This attribute indicates that the inliner should never inline this
1088 function in any situation. This attribute may not be used together with
1089 the <tt>alwaysinline</tt> attribute.</dd>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1090
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1091 <dt><tt><b>optsize</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1092 <dd>This attribute suggests that optimization passes and code generator passes
1093 make choices that keep the code size of this function low, and otherwise
1094 do optimizations specifically to reduce code size.</dd>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1095
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1096 <dt><tt><b>noreturn</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1097 <dd>This function attribute indicates that the function never returns
1098 normally. This produces undefined behavior at runtime if the function
1099 ever does dynamically return.</dd>
Bill Wendling31359ba2008-11-13 01:02:51 +0000[diff] [blame]1100
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1101 <dt><tt><b>nounwind</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1102 <dd>This function attribute indicates that the function never returns with an
1103 unwind or exceptional control flow. If the function does unwind, its
1104 runtime behavior is undefined.</dd>
Bill Wendlingfbaa7ed2008-11-26 19:07:40 +0000[diff] [blame]1105
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1106 <dt><tt><b>readnone</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1107 <dd>This attribute indicates that the function computes its result (or decides
1108 to unwind an exception) based strictly on its arguments, without
1109 dereferencing any pointer arguments or otherwise accessing any mutable
1110 state (e.g. memory, control registers, etc) visible to caller functions.
1111 It does not write through any pointer arguments
1112 (including <tt><a href="#byval">byval</a></tt> arguments) and never
1113 changes any state visible to callers. This means that it cannot unwind
1114 exceptions by calling the <tt>C++</tt> exception throwing methods, but
1115 could use the <tt>unwind</tt> instruction.</dd>
Devang Patel5d96fda2009-06-12 19:45:19 +0000[diff] [blame]1116
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1117 <dt><tt><b><a name="readonly">readonly</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1118 <dd>This attribute indicates that the function does not write through any
1119 pointer arguments (including <tt><a href="#byval">byval</a></tt>
1120 arguments) or otherwise modify any state (e.g. memory, control registers,
1121 etc) visible to caller functions. It may dereference pointer arguments
1122 and read state that may be set in the caller. A readonly function always
1123 returns the same value (or unwinds an exception identically) when called
1124 with the same set of arguments and global state. It cannot unwind an
1125 exception by calling the <tt>C++</tt> exception throwing methods, but may
1126 use the <tt>unwind</tt> instruction.</dd>
Anton Korobeynikovc5ec8a72009-07-17 18:07:26 +0000[diff] [blame]1127
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1128 <dt><tt><b><a name="ssp">ssp</a></b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1129 <dd>This attribute indicates that the function should emit a stack smashing
1130 protector. It is in the form of a "canary"&mdash;a random value placed on
1131 the stack before the local variables that's checked upon return from the
1132 function to see if it has been overwritten. A heuristic is used to
1133 determine if a function needs stack protectors or not.<br>
1134<br>
1135 If a function that has an <tt>ssp</tt> attribute is inlined into a
1136 function that doesn't have an <tt>ssp</tt> attribute, then the resulting
1137 function will have an <tt>ssp</tt> attribute.</dd>
1138
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1139 <dt><tt><b>sspreq</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1140 <dd>This attribute indicates that the function should <em>always</em> emit a
1141 stack smashing protector. This overrides
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1142 the <tt><a href="#ssp">ssp</a></tt> function attribute.<br>
1143<br>
1144 If a function that has an <tt>sspreq</tt> attribute is inlined into a
1145 function that doesn't have an <tt>sspreq</tt> attribute or which has
1146 an <tt>ssp</tt> attribute, then the resulting function will have
1147 an <tt>sspreq</tt> attribute.</dd>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1148
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1149 <dt><tt><b>noredzone</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1150 <dd>This attribute indicates that the code generator should not use a red
1151 zone, even if the target-specific ABI normally permits it.</dd>
1152
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1153 <dt><tt><b>noimplicitfloat</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1154 <dd>This attributes disables implicit floating point instructions.</dd>
1155
Bill Wendlingf82d40a2009-11-02 00:24:16 +0000[diff] [blame]1156 <dt><tt><b>naked</b></tt></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1157 <dd>This attribute disables prologue / epilogue emission for the function.
1158 This can have very system-specific consequences.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1159</dl>
1160
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1161</div>
1162
1163<!-- ======================================================================= -->
1164<div class="doc_subsection">
Chris Lattner1eeeb0c2006-04-08 04:40:53 +0000[diff] [blame]1165 <a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1166</div>
1167
1168<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1169
1170<p>Modules may contain "module-level inline asm" blocks, which corresponds to
1171 the GCC "file scope inline asm" blocks. These blocks are internally
1172 concatenated by LLVM and treated as a single unit, but may be separated in
1173 the <tt>.ll</tt> file if desired. The syntax is very simple:</p>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1174
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1175<div class="doc_code">
1176<pre>
1177module asm "inline asm code goes here"
1178module asm "more can go here"
1179</pre>
1180</div>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1181
1182<p>The strings can contain any character by escaping non-printable characters.
1183 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]1184 for the number.</p>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1185
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1186<p>The inline asm code is simply printed to the machine code .s file when
1187 assembly code is generated.</p>
1188
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1189</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1190
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1191<!-- ======================================================================= -->
1192<div class="doc_subsection">
1193 <a name="datalayout">Data Layout</a>
1194</div>
1195
1196<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1197
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1198<p>A module may specify a target specific data layout string that specifies how
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1199 data is to be laid out in memory. The syntax for the data layout is
1200 simply:</p>
1201
1202<div class="doc_code">
1203<pre>
1204target datalayout = "<i>layout specification</i>"
1205</pre>
1206</div>
1207
1208<p>The <i>layout specification</i> consists of a list of specifications
1209 separated by the minus sign character ('-'). Each specification starts with
1210 a letter and may include other information after the letter to define some
1211 aspect of the data layout. The specifications accepted are as follows:</p>
1212
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1213<dl>
1214 <dt><tt>E</tt></dt>
1215 <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]1216 bits with the most significance have the lowest address location.</dd>
1217
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1218 <dt><tt>e</tt></dt>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1219 <dd>Specifies that the target lays out data in little-endian form. That is,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1220 the bits with the least significance have the lowest address
1221 location.</dd>
1222
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1223 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1224 <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]1225 <i>preferred</i> alignments. All sizes are in bits. Specifying
1226 the <i>pref</i> alignment is optional. If omitted, the
1227 preceding <tt>:</tt> should be omitted too.</dd>
1228
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1229 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1230 <dd>This specifies the alignment for an integer type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1231 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
1232
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1233 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1234 <dd>This specifies the alignment for a vector type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1235 <i>size</i>.</dd>
1236
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1237 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1238 <dd>This specifies the alignment for a floating point type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1239 <i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
1240 (double).</dd>
1241
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1242 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1243 <dd>This specifies the alignment for an aggregate type of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1244 <i>size</i>.</dd>
1245
Daniel Dunbar87bde0b2009-06-08 22:17:53 +0000[diff] [blame]1246 <dt><tt>s<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1247 <dd>This specifies the alignment for a stack object of a given bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1248 <i>size</i>.</dd>
Chris Lattnere82bdc42009-11-07 09:35:34 +0000[diff] [blame]1249
1250 <dt><tt>n<i>size1</i>:<i>size2</i>:<i>size3</i>...</tt></dt>
1251 <dd>This specifies a set of native integer widths for the target CPU
1252 in bits. For example, it might contain "n32" for 32-bit PowerPC,
1253 "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]1254 this set are considered to support most general arithmetic
Chris Lattnere82bdc42009-11-07 09:35:34 +0000[diff] [blame]1255 operations efficiently.</dd>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1256</dl>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1257
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1258<p>When constructing the data layout for a given target, LLVM starts with a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1259 default set of specifications which are then (possibly) overriden by the
1260 specifications in the <tt>datalayout</tt> keyword. The default specifications
1261 are given in this list:</p>
1262
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1263<ul>
1264 <li><tt>E</tt> - big endian</li>
1265 <li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
1266 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
1267 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
1268 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
1269 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1270 <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]1271 alignment of 64-bits</li>
1272 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
1273 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
1274 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
1275 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
1276 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
Daniel Dunbar87bde0b2009-06-08 22:17:53 +0000[diff] [blame]1277 <li><tt>s0:64:64</tt> - stack objects are 64-bit aligned</li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1278</ul>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1279
1280<p>When LLVM is determining the alignment for a given type, it uses the
1281 following rules:</p>
1282
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1283<ol>
1284 <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]1285 specification is used.</li>
1286
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1287 <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]1288 smallest integer type that is larger than the bitwidth of the sought type
1289 is used. If none of the specifications are larger than the bitwidth then
1290 the the largest integer type is used. For example, given the default
1291 specifications above, the i7 type will use the alignment of i8 (next
1292 largest) while both i65 and i256 will use the alignment of i64 (largest
1293 specified).</li>
1294
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1295 <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]1296 largest vector type that is smaller than the sought vector type will be
1297 used as a fall back. This happens because &lt;128 x double&gt; can be
1298 implemented in terms of 64 &lt;2 x double&gt;, for example.</li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1299</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1300
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1301</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1302
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1303<!-- ======================================================================= -->
1304<div class="doc_subsection">
1305 <a name="pointeraliasing">Pointer Aliasing Rules</a>
1306</div>
1307
1308<div class="doc_text">
1309
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1310<p>Any memory access must be done through a pointer value associated
Andreas Bolka99a82052009-07-27 20:37:10 +0000[diff] [blame]1311with an address range of the memory access, otherwise the behavior
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1312is undefined. Pointer values are associated with address ranges
1313according to the following rules:</p>
1314
1315<ul>
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1316 <li>A pointer value formed from a
1317 <tt><a href="#i_getelementptr">getelementptr</a></tt> instruction
1318 is associated with the addresses associated with the first operand
1319 of the <tt>getelementptr</tt>.</li>
1320 <li>An address of a global variable is associated with the address
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1321 range of the variable's storage.</li>
1322 <li>The result value of an allocation instruction is associated with
1323 the address range of the allocated storage.</li>
1324 <li>A null pointer in the default address-space is associated with
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1325 no address.</li>
1326 <li>A pointer value formed by an
1327 <tt><a href="#i_inttoptr">inttoptr</a></tt> is associated with all
1328 address ranges of all pointer values that contribute (directly or
1329 indirectly) to the computation of the pointer's value.</li>
1330 <li>The result value of a
1331 <tt><a href="#i_bitcast">bitcast</a></tt> is associated with all
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1332 addresses associated with the operand of the <tt>bitcast</tt>.</li>
1333 <li>An integer constant other than zero or a pointer value returned
1334 from a function not defined within LLVM may be associated with address
1335 ranges allocated through mechanisms other than those provided by
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1336 LLVM. Such ranges shall not overlap with any ranges of addresses
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1337 allocated by mechanisms provided by LLVM.</li>
1338 </ul>
1339
1340<p>LLVM IR does not associate types with memory. The result type of a
Andreas Bolka55e459a2009-07-29 00:02:05 +0000[diff] [blame]1341<tt><a href="#i_load">load</a></tt> merely indicates the size and
1342alignment of the memory from which to load, as well as the
1343interpretation of the value. The first operand of a
1344<tt><a href="#i_store">store</a></tt> similarly only indicates the size
1345and alignment of the store.</p>
Dan Gohman556ca272009-07-27 18:07:55 +0000[diff] [blame]1346
1347<p>Consequently, type-based alias analysis, aka TBAA, aka
1348<tt>-fstrict-aliasing</tt>, is not applicable to general unadorned
1349LLVM IR. <a href="#metadata">Metadata</a> may be used to encode
1350additional information which specialized optimization passes may use
1351to implement type-based alias analysis.</p>
1352
1353</div>
1354
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1355<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1356<div class="doc_section"> <a name="typesystem">Type System</a> </div>
1357<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1358
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1359<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1360
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1361<p>The LLVM type system is one of the most important features of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1362 intermediate representation. Being typed enables a number of optimizations
1363 to be performed on the intermediate representation directly, without having
1364 to do extra analyses on the side before the transformation. A strong type
1365 system makes it easier to read the generated code and enables novel analyses
1366 and transformations that are not feasible to perform on normal three address
1367 code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1368
1369</div>
1370
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1371<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1372<div class="doc_subsection"> <a name="t_classifications">Type
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1373Classifications</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1374
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1375<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1376
1377<p>The types fall into a few useful classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1378
1379<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1380 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1381 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1382 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1383 <td><a href="#t_integer">integer</a></td>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1384 <td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1385 </tr>
1386 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1387 <td><a href="#t_floating">floating point</a></td>
1388 <td><tt>float, double, x86_fp80, fp128, ppc_fp128</tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1389 </tr>
1390 <tr>
1391 <td><a name="t_firstclass">first class</a></td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1392 <td><a href="#t_integer">integer</a>,
1393 <a href="#t_floating">floating point</a>,
1394 <a href="#t_pointer">pointer</a>,
Dan Gohman0066db62008-06-18 18:42:13 +0000[diff] [blame]1395 <a href="#t_vector">vector</a>,
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]1396 <a href="#t_struct">structure</a>,
1397 <a href="#t_array">array</a>,
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1398 <a href="#t_label">label</a>,
1399 <a href="#t_metadata">metadata</a>.
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1400 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1401 </tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1402 <tr>
1403 <td><a href="#t_primitive">primitive</a></td>
1404 <td><a href="#t_label">label</a>,
1405 <a href="#t_void">void</a>,
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1406 <a href="#t_floating">floating point</a>,
1407 <a href="#t_metadata">metadata</a>.</td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1408 </tr>
1409 <tr>
1410 <td><a href="#t_derived">derived</a></td>
1411 <td><a href="#t_integer">integer</a>,
1412 <a href="#t_array">array</a>,
1413 <a href="#t_function">function</a>,
1414 <a href="#t_pointer">pointer</a>,
1415 <a href="#t_struct">structure</a>,
1416 <a href="#t_pstruct">packed structure</a>,
1417 <a href="#t_vector">vector</a>,
1418 <a href="#t_opaque">opaque</a>.
Dan Gohman01ac1012008-10-14 16:32:04 +0000[diff] [blame]1419 </td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1420 </tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1421 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1422</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1423
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1424<p>The <a href="#t_firstclass">first class</a> types are perhaps the most
1425 important. Values of these types are the only ones which can be produced by
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1426 instructions.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1427
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1428</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1429
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1430<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1431<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1432
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1433<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1434
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1435<p>The primitive types are the fundamental building blocks of the LLVM
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1436 system.</p>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1437
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1438</div>
1439
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1440<!-- _______________________________________________________________________ -->
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1441<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
1442
1443<div class="doc_text">
1444
1445<h5>Overview:</h5>
1446<p>The integer type is a very simple type that simply specifies an arbitrary
1447 bit width for the integer type desired. Any bit width from 1 bit to
1448 2<sup>23</sup>-1 (about 8 million) can be specified.</p>
1449
1450<h5>Syntax:</h5>
1451<pre>
1452 iN
1453</pre>
1454
1455<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
1456 value.</p>
1457
1458<h5>Examples:</h5>
1459<table class="layout">
1460 <tr class="layout">
1461 <td class="left"><tt>i1</tt></td>
1462 <td class="left">a single-bit integer.</td>
1463 </tr>
1464 <tr class="layout">
1465 <td class="left"><tt>i32</tt></td>
1466 <td class="left">a 32-bit integer.</td>
1467 </tr>
1468 <tr class="layout">
1469 <td class="left"><tt>i1942652</tt></td>
1470 <td class="left">a really big integer of over 1 million bits.</td>
1471 </tr>
1472</table>
1473
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1474</div>
1475
1476<!-- _______________________________________________________________________ -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1477<div class="doc_subsubsection"> <a name="t_floating">Floating Point Types</a> </div>
1478
1479<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1480
1481<table>
1482 <tbody>
1483 <tr><th>Type</th><th>Description</th></tr>
1484 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
1485 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
1486 <tr><td><tt>fp128</tt></td><td>128-bit floating point value (112-bit mantissa)</td></tr>
1487 <tr><td><tt>x86_fp80</tt></td><td>80-bit floating point value (X87)</td></tr>
1488 <tr><td><tt>ppc_fp128</tt></td><td>128-bit floating point value (two 64-bits)</td></tr>
1489 </tbody>
1490</table>
1491
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1492</div>
1493
1494<!-- _______________________________________________________________________ -->
1495<div class="doc_subsubsection"> <a name="t_void">Void Type</a> </div>
1496
1497<div class="doc_text">
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1498
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1499<h5>Overview:</h5>
1500<p>The void type does not represent any value and has no size.</p>
1501
1502<h5>Syntax:</h5>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1503<pre>
1504 void
1505</pre>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1506
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1507</div>
1508
1509<!-- _______________________________________________________________________ -->
1510<div class="doc_subsubsection"> <a name="t_label">Label Type</a> </div>
1511
1512<div class="doc_text">
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1513
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1514<h5>Overview:</h5>
1515<p>The label type represents code labels.</p>
1516
1517<h5>Syntax:</h5>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1518<pre>
1519 label
1520</pre>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1521
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1522</div>
1523
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1524<!-- _______________________________________________________________________ -->
1525<div class="doc_subsubsection"> <a name="t_metadata">Metadata Type</a> </div>
1526
1527<div class="doc_text">
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1528
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1529<h5>Overview:</h5>
Nick Lewyckyc261df92009-09-27 23:27:42 +0000[diff] [blame]1530<p>The metadata type represents embedded metadata. No derived types may be
1531 created from metadata except for <a href="#t_function">function</a>
1532 arguments.
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1533
1534<h5>Syntax:</h5>
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1535<pre>
1536 metadata
1537</pre>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1538
Nick Lewycky7a0370f2009-05-30 05:06:04 +0000[diff] [blame]1539</div>
1540
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1541
1542<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1543<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1544
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1545<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1546
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1547<p>The real power in LLVM comes from the derived types in the system. This is
1548 what allows a programmer to represent arrays, functions, pointers, and other
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1549 useful types. Each of these types contain one or more element types which
1550 may be a primitive type, or another derived type. For example, it is
1551 possible to have a two dimensional array, using an array as the element type
1552 of another array.</p>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1553
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1554</div>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1555
1556<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1557<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1558
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1559<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1560
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1561<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1562<p>The array type is a very simple derived type that arranges elements
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1563 sequentially in memory. The array type requires a size (number of elements)
1564 and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1565
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1566<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1567<pre>
1568 [&lt;# elements&gt; x &lt;elementtype&gt;]
1569</pre>
1570
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1571<p>The number of elements is a constant integer value; <tt>elementtype</tt> may
1572 be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1573
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1574<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1575<table class="layout">
1576 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1577 <td class="left"><tt>[40 x i32]</tt></td>
1578 <td class="left">Array of 40 32-bit integer values.</td>
1579 </tr>
1580 <tr class="layout">
1581 <td class="left"><tt>[41 x i32]</tt></td>
1582 <td class="left">Array of 41 32-bit integer values.</td>
1583 </tr>
1584 <tr class="layout">
1585 <td class="left"><tt>[4 x i8]</tt></td>
1586 <td class="left">Array of 4 8-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1587 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1588</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1589<p>Here are some examples of multidimensional arrays:</p>
1590<table class="layout">
1591 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1592 <td class="left"><tt>[3 x [4 x i32]]</tt></td>
1593 <td class="left">3x4 array of 32-bit integer values.</td>
1594 </tr>
1595 <tr class="layout">
1596 <td class="left"><tt>[12 x [10 x float]]</tt></td>
1597 <td class="left">12x10 array of single precision floating point values.</td>
1598 </tr>
1599 <tr class="layout">
1600 <td class="left"><tt>[2 x [3 x [4 x i16]]]</tt></td>
1601 <td class="left">2x3x4 array of 16-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1602 </tr>
1603</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1604
Dan Gohman7657f6b2009-11-09 19:01:53 +0000[diff] [blame]1605<p>There is no restriction on indexing beyond the end of the array implied by
1606 a static type (though there are restrictions on indexing beyond the bounds
1607 of an allocated object in some cases). This means that single-dimension
1608 'variable sized array' addressing can be implemented in LLVM with a zero
1609 length array type. An implementation of 'pascal style arrays' in LLVM could
1610 use the type "<tt>{ i32, [0 x float]}</tt>", for example.</p>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1611
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1612</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1613
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1614<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1615<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1616
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1617<div class="doc_text">
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1618
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1619<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1620<p>The function type can be thought of as a function signature. It consists of
1621 a return type and a list of formal parameter types. The return type of a
1622 function type is a scalar type, a void type, or a struct type. If the return
1623 type is a struct type then all struct elements must be of first class types,
1624 and the struct must have at least one element.</p>
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]1625
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1626<h5>Syntax:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1627<pre>
Nick Lewycky51386942009-09-27 07:55:32 +0000[diff] [blame]1628 &lt;returntype&gt; (&lt;parameter list&gt;)
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1629</pre>
1630
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1631<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]1632 specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
1633 which indicates that the function takes a variable number of arguments.
1634 Variable argument functions can access their arguments with
1635 the <a href="#int_varargs">variable argument handling intrinsic</a>
Nick Lewycky51386942009-09-27 07:55:32 +0000[diff] [blame]1636 functions. '<tt>&lt;returntype&gt;</tt>' is a any type except
Nick Lewyckyc261df92009-09-27 23:27:42 +0000[diff] [blame]1637 <a href="#t_label">label</a>.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1638
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1639<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1640<table class="layout">
1641 <tr class="layout">
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1642 <td class="left"><tt>i32 (i32)</tt></td>
1643 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1644 </td>
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1645 </tr><tr class="layout">
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]1646 <td class="left"><tt>float&nbsp;(i16&nbsp;signext,&nbsp;i32&nbsp;*)&nbsp;*
Reid Spencerf17a0b72006-12-31 07:20:23 +0000[diff] [blame]1647 </tt></td>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1648 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
1649 an <tt>i16</tt> that should be sign extended and a
1650 <a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1651 <tt>float</tt>.
1652 </td>
1653 </tr><tr class="layout">
1654 <td class="left"><tt>i32 (i8*, ...)</tt></td>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]1655 <td class="left">A vararg function that takes at least one
1656 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
1657 which returns an integer. This is the signature for <tt>printf</tt> in
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1658 LLVM.
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1659 </td>
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1660 </tr><tr class="layout">
1661 <td class="left"><tt>{i32, i32} (i32)</tt></td>
Nick Lewycky51386942009-09-27 07:55:32 +0000[diff] [blame]1662 <td class="left">A function taking an <tt>i32</tt>, returning a
1663 <a href="#t_struct">structure</a> containing two <tt>i32</tt> values
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1664 </td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1665 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1666</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1667
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1668</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1669
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1670<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1671<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1672
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1673<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1674
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1675<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1676<p>The structure type is used to represent a collection of data members together
1677 in memory. The packing of the field types is defined to match the ABI of the
1678 underlying processor. The elements of a structure may be any type that has a
1679 size.</p>
1680
Jeffrey Yasskin7a088cf2010-01-11 19:19:26 +0000[diff] [blame]1681<p>Structures in memory are accessed using '<tt><a href="#i_load">load</a></tt>'
1682 and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a field
1683 with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.
1684 Structures in registers are accessed using the
1685 '<tt><a href="#i_extractvalue">extractvalue</a></tt>' and
1686 '<tt><a href="#i_insertvalue">insertvalue</a></tt>' instructions.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1687<h5>Syntax:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1688<pre>
1689 { &lt;type list&gt; }
1690</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1691
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1692<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1693<table class="layout">
1694 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1695 <td class="left"><tt>{ i32, i32, i32 }</tt></td>
1696 <td class="left">A triple of three <tt>i32</tt> values</td>
1697 </tr><tr class="layout">
1698 <td class="left"><tt>{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}</tt></td>
1699 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1700 second element is a <a href="#t_pointer">pointer</a> to a
1701 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1702 an <tt>i32</tt>.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1703 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1704</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1705
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1706</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1707
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1708<!-- _______________________________________________________________________ -->
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1709<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1710</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1711
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1712<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1713
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1714<h5>Overview:</h5>
1715<p>The packed structure type is used to represent a collection of data members
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1716 together in memory. There is no padding between fields. Further, the
1717 alignment of a packed structure is 1 byte. The elements of a packed
1718 structure may be any type that has a size.</p>
1719
1720<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt> and
1721 '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a field with
1722 the '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.</p>
1723
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1724<h5>Syntax:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1725<pre>
1726 &lt; { &lt;type list&gt; } &gt;
1727</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1728
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1729<h5>Examples:</h5>
1730<table class="layout">
1731 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1732 <td class="left"><tt>&lt; { i32, i32, i32 } &gt;</tt></td>
1733 <td class="left">A triple of three <tt>i32</tt> values</td>
1734 </tr><tr class="layout">
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1735 <td class="left">
1736<tt>&lt;&nbsp;{&nbsp;float,&nbsp;i32&nbsp;(i32)*&nbsp;}&nbsp;&gt;</tt></td>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1737 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1738 second element is a <a href="#t_pointer">pointer</a> to a
1739 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1740 an <tt>i32</tt>.</td>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1741 </tr>
1742</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1743
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1744</div>
1745
1746<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1747<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Chris Lattner0fd4a272009-02-08 19:53:29 +0000[diff] [blame]1748
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1749<div class="doc_text">
1750
1751<h5>Overview:</h5>
1752<p>As in many languages, the pointer type represents a pointer or reference to
1753 another object, which must live in memory. Pointer types may have an optional
1754 address space attribute defining the target-specific numbered address space
1755 where the pointed-to object resides. The default address space is zero.</p>
1756
1757<p>Note that LLVM does not permit pointers to void (<tt>void*</tt>) nor does it
1758 permit pointers to labels (<tt>label*</tt>). Use <tt>i8*</tt> instead.</p>
Chris Lattner0fd4a272009-02-08 19:53:29 +0000[diff] [blame]1759
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1760<h5>Syntax:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]1761<pre>
1762 &lt;type&gt; *
1763</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1764
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1765<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1766<table class="layout">
1767 <tr class="layout">
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]1768 <td class="left"><tt>[4 x i32]*</tt></td>
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1769 <td class="left">A <a href="#t_pointer">pointer</a> to <a
1770 href="#t_array">array</a> of four <tt>i32</tt> values.</td>
1771 </tr>
1772 <tr class="layout">
1773 <td class="left"><tt>i32 (i32 *) *</tt></td>
1774 <td class="left"> A <a href="#t_pointer">pointer</a> to a <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1775 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1776 <tt>i32</tt>.</td>
1777 </tr>
1778 <tr class="layout">
1779 <td class="left"><tt>i32 addrspace(5)*</tt></td>
1780 <td class="left">A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value
1781 that resides in address space #5.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1782 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1783</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1784
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1785</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1786
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1787<!-- _______________________________________________________________________ -->
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1788<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1789
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1790<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1791
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1792<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1793<p>A vector type is a simple derived type that represents a vector of elements.
1794 Vector types are used when multiple primitive data are operated in parallel
1795 using a single instruction (SIMD). A vector type requires a size (number of
Duncan Sandsd40d14e2009-11-27 13:38:03 +0000[diff] [blame]1796 elements) and an underlying primitive data type. Vector types are considered
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1797 <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1798
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1799<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1800<pre>
1801 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1802</pre>
1803
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1804<p>The number of elements is a constant integer value; elementtype may be any
1805 integer or floating point type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1806
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1807<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1808<table class="layout">
1809 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1810 <td class="left"><tt>&lt;4 x i32&gt;</tt></td>
1811 <td class="left">Vector of 4 32-bit integer values.</td>
1812 </tr>
1813 <tr class="layout">
1814 <td class="left"><tt>&lt;8 x float&gt;</tt></td>
1815 <td class="left">Vector of 8 32-bit floating-point values.</td>
1816 </tr>
1817 <tr class="layout">
1818 <td class="left"><tt>&lt;2 x i64&gt;</tt></td>
1819 <td class="left">Vector of 2 64-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1820 </tr>
1821</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1822
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1823</div>
1824
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1825<!-- _______________________________________________________________________ -->
1826<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1827<div class="doc_text">
1828
1829<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1830<p>Opaque types are used to represent unknown types in the system. This
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1831 corresponds (for example) to the C notion of a forward declared structure
1832 type. In LLVM, opaque types can eventually be resolved to any type (not just
1833 a structure type).</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1834
1835<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1836<pre>
1837 opaque
1838</pre>
1839
1840<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1841<table class="layout">
1842 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1843 <td class="left"><tt>opaque</tt></td>
1844 <td class="left">An opaque type.</td>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1845 </tr>
1846</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1847
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1848</div>
1849
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1850<!-- ======================================================================= -->
1851<div class="doc_subsection">
1852 <a name="t_uprefs">Type Up-references</a>
1853</div>
1854
1855<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1856
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1857<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1858<p>An "up reference" allows you to refer to a lexically enclosing type without
1859 requiring it to have a name. For instance, a structure declaration may
1860 contain a pointer to any of the types it is lexically a member of. Example
1861 of up references (with their equivalent as named type declarations)
1862 include:</p>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1863
1864<pre>
Chris Lattner3060f5b2009-02-09 10:00:56 +0000[diff] [blame]1865 { \2 * } %x = type { %x* }
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1866 { \2 }* %y = type { %y }*
1867 \1* %z = type %z*
1868</pre>
1869
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1870<p>An up reference is needed by the asmprinter for printing out cyclic types
1871 when there is no declared name for a type in the cycle. Because the
1872 asmprinter does not want to print out an infinite type string, it needs a
1873 syntax to handle recursive types that have no names (all names are optional
1874 in llvm IR).</p>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1875
1876<h5>Syntax:</h5>
1877<pre>
1878 \&lt;level&gt;
1879</pre>
1880
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1881<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]1882
1883<h5>Examples:</h5>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1884<table class="layout">
1885 <tr class="layout">
1886 <td class="left"><tt>\1*</tt></td>
1887 <td class="left">Self-referential pointer.</td>
1888 </tr>
1889 <tr class="layout">
1890 <td class="left"><tt>{ { \3*, i8 }, i32 }</tt></td>
1891 <td class="left">Recursive structure where the upref refers to the out-most
1892 structure.</td>
1893 </tr>
1894</table>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame]1895
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1896</div>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1897
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1898<!-- *********************************************************************** -->
1899<div class="doc_section"> <a name="constants">Constants</a> </div>
1900<!-- *********************************************************************** -->
1901
1902<div class="doc_text">
1903
1904<p>LLVM has several different basic types of constants. This section describes
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1905 them all and their syntax.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1906
1907</div>
1908
1909<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1910<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1911
1912<div class="doc_text">
1913
1914<dl>
1915 <dt><b>Boolean constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1916 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]1917 constants of the <tt><a href="#t_integer">i1</a></tt> type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1918
1919 <dt><b>Integer constants</b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1920 <dd>Standard integers (such as '4') are constants of
1921 the <a href="#t_integer">integer</a> type. Negative numbers may be used
1922 with integer types.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1923
1924 <dt><b>Floating point constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1925 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1926 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
1927 notation (see below). The assembler requires the exact decimal value of a
1928 floating-point constant. For example, the assembler accepts 1.25 but
1929 rejects 1.3 because 1.3 is a repeating decimal in binary. Floating point
1930 constants must have a <a href="#t_floating">floating point</a> type. </dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1931
1932 <dt><b>Null pointer constants</b></dt>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1933 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1934 and must be of <a href="#t_pointer">pointer type</a>.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1935</dl>
1936
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1937<p>The one non-intuitive notation for constants is the hexadecimal form of
1938 floating point constants. For example, the form '<tt>double
1939 0x432ff973cafa8000</tt>' is equivalent to (but harder to read than)
1940 '<tt>double 4.5e+15</tt>'. The only time hexadecimal floating point
1941 constants are required (and the only time that they are generated by the
1942 disassembler) is when a floating point constant must be emitted but it cannot
1943 be represented as a decimal floating point number in a reasonable number of
1944 digits. For example, NaN's, infinities, and other special values are
1945 represented in their IEEE hexadecimal format so that assembly and disassembly
1946 do not cause any bits to change in the constants.</p>
1947
Dale Johannesenbd5e5a82009-02-11 22:14:51 +0000[diff] [blame]1948<p>When using the hexadecimal form, constants of types float and double are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1949 represented using the 16-digit form shown above (which matches the IEEE754
1950 representation for double); float values must, however, be exactly
1951 representable as IEE754 single precision. Hexadecimal format is always used
1952 for long double, and there are three forms of long double. The 80-bit format
1953 used by x86 is represented as <tt>0xK</tt> followed by 20 hexadecimal digits.
1954 The 128-bit format used by PowerPC (two adjacent doubles) is represented
1955 by <tt>0xM</tt> followed by 32 hexadecimal digits. The IEEE 128-bit format
1956 is represented by <tt>0xL</tt> followed by 32 hexadecimal digits; no
1957 currently supported target uses this format. Long doubles will only work if
1958 they match the long double format on your target. All hexadecimal formats
1959 are big-endian (sign bit at the left).</p>
1960
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1961</div>
1962
1963<!-- ======================================================================= -->
Chris Lattner70882792009-02-28 18:32:25 +0000[diff] [blame]1964<div class="doc_subsection">
Bill Wendlingd9fe2982009-07-20 02:32:41 +0000[diff] [blame]1965<a name="aggregateconstants"></a> <!-- old anchor -->
1966<a name="complexconstants">Complex Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1967</div>
1968
1969<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1970
Chris Lattner70882792009-02-28 18:32:25 +0000[diff] [blame]1971<p>Complex constants are a (potentially recursive) combination of simple
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1972 constants and smaller complex constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1973
1974<dl>
1975 <dt><b>Structure constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1976 <dd>Structure constants are represented with notation similar to structure
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1977 type definitions (a comma separated list of elements, surrounded by braces
1978 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* @G }</tt>",
1979 where "<tt>@G</tt>" is declared as "<tt>@G = external global i32</tt>".
1980 Structure constants must have <a href="#t_struct">structure type</a>, and
1981 the number and types of elements must match those specified by the
1982 type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1983
1984 <dt><b>Array constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1985 <dd>Array constants are represented with notation similar to array type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1986 definitions (a comma separated list of elements, surrounded by square
1987 brackets (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74
1988 ]</tt>". Array constants must have <a href="#t_array">array type</a>, and
1989 the number and types of elements must match those specified by the
1990 type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1991
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1992 <dt><b>Vector constants</b></dt>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1993 <dd>Vector constants are represented with notation similar to vector type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]1994 definitions (a comma separated list of elements, surrounded by
1995 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32
1996 42, i32 11, i32 74, i32 100 &gt;</tt>". Vector constants must
1997 have <a href="#t_vector">vector type</a>, and the number and types of
1998 elements must match those specified by the type.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1999
2000 <dt><b>Zero initialization</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2001 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2002 value to zero of <em>any</em> type, including scalar and aggregate types.
2003 This is often used to avoid having to print large zero initializers
2004 (e.g. for large arrays) and is always exactly equivalent to using explicit
2005 zero initializers.</dd>
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2006
2007 <dt><b>Metadata node</b></dt>
Nick Lewycky1e8c7a62009-05-30 16:08:30 +0000[diff] [blame]2008 <dd>A metadata node is a structure-like constant with
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2009 <a href="#t_metadata">metadata type</a>. For example: "<tt>metadata !{
2010 i32 0, metadata !"test" }</tt>". Unlike other constants that are meant to
2011 be interpreted as part of the instruction stream, metadata is a place to
2012 attach additional information such as debug info.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2013</dl>
2014
2015</div>
2016
2017<!-- ======================================================================= -->
2018<div class="doc_subsection">
2019 <a name="globalconstants">Global Variable and Function Addresses</a>
2020</div>
2021
2022<div class="doc_text">
2023
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2024<p>The addresses of <a href="#globalvars">global variables</a>
2025 and <a href="#functionstructure">functions</a> are always implicitly valid
2026 (link-time) constants. These constants are explicitly referenced when
2027 the <a href="#identifiers">identifier for the global</a> is used and always
2028 have <a href="#t_pointer">pointer</a> type. For example, the following is a
2029 legal LLVM file:</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2030
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2031<div class="doc_code">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2032<pre>
Chris Lattnera18a4242007-06-06 18:28:13 +0000[diff] [blame]2033@X = global i32 17
2034@Y = global i32 42
2035@Z = global [2 x i32*] [ i32* @X, i32* @Y ]
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2036</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2037</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2038
2039</div>
2040
2041<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]2042<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2043<div class="doc_text">
2044
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2045<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]2046 indicates that the user of the value may receive an unspecified bit-pattern.
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2047 Undefined values may be of any type (other than label or void) and be used
2048 anywhere a constant is permitted.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2049
Chris Lattnerc608cb12009-09-11 01:49:31 +0000[diff] [blame]2050<p>Undefined values are useful because they indicate to the compiler that the
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2051 program is well defined no matter what value is used. This gives the
2052 compiler more freedom to optimize. Here are some examples of (potentially
2053 surprising) transformations that are valid (in pseudo IR):</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2054
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2055
2056<div class="doc_code">
2057<pre>
2058 %A = add %X, undef
2059 %B = sub %X, undef
2060 %C = xor %X, undef
2061Safe:
2062 %A = undef
2063 %B = undef
2064 %C = undef
2065</pre>
2066</div>
2067
2068<p>This is safe because all of the output bits are affected by the undef bits.
2069Any output bit can have a zero or one depending on the input bits.</p>
2070
2071<div class="doc_code">
2072<pre>
2073 %A = or %X, undef
2074 %B = and %X, undef
2075Safe:
2076 %A = -1
2077 %B = 0
2078Unsafe:
2079 %A = undef
2080 %B = undef
2081</pre>
2082</div>
2083
2084<p>These logical operations have bits that are not always affected by the input.
2085For example, if "%X" has a zero bit, then the output of the 'and' operation will
2086always be a zero, no matter what the corresponding bit from the undef is. As
Chris Lattnerc608cb12009-09-11 01:49:31 +0000[diff] [blame]2087such, it is unsafe to optimize or assume that the result of the and is undef.
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2088However, it is safe to assume that all bits of the undef could be 0, and
2089optimize the and to 0. Likewise, it is safe to assume that all the bits of
2090the undef operand to the or could be set, allowing the or to be folded to
Chris Lattnerc608cb12009-09-11 01:49:31 +0000[diff] [blame]2091-1.</p>
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2092
2093<div class="doc_code">
2094<pre>
2095 %A = select undef, %X, %Y
2096 %B = select undef, 42, %Y
2097 %C = select %X, %Y, undef
2098Safe:
2099 %A = %X (or %Y)
2100 %B = 42 (or %Y)
2101 %C = %Y
2102Unsafe:
2103 %A = undef
2104 %B = undef
2105 %C = undef
2106</pre>
2107</div>
2108
2109<p>This set of examples show that undefined select (and conditional branch)
2110conditions can go "either way" but they have to come from one of the two
2111operands. In the %A example, if %X and %Y were both known to have a clear low
2112bit, then %A would have to have a cleared low bit. However, in the %C example,
2113the optimizer is allowed to assume that the undef operand could be the same as
2114%Y, allowing the whole select to be eliminated.</p>
2115
2116
2117<div class="doc_code">
2118<pre>
2119 %A = xor undef, undef
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2120
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2121 %B = undef
2122 %C = xor %B, %B
2123
2124 %D = undef
2125 %E = icmp lt %D, 4
2126 %F = icmp gte %D, 4
2127
2128Safe:
2129 %A = undef
2130 %B = undef
2131 %C = undef
2132 %D = undef
2133 %E = undef
2134 %F = undef
2135</pre>
2136</div>
2137
2138<p>This example points out that two undef operands are not necessarily the same.
2139This can be surprising to people (and also matches C semantics) where they
2140assume that "X^X" is always zero, even if X is undef. This isn't true for a
2141number of reasons, but the short answer is that an undef "variable" can
2142arbitrarily change its value over its "live range". This is true because the
2143"variable" doesn't actually <em>have a live range</em>. Instead, the value is
2144logically read from arbitrary registers that happen to be around when needed,
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]2145so the value is not necessarily consistent over time. In fact, %A and %C need
Chris Lattner349eb412009-09-08 15:13:16 +0000[diff] [blame]2146to have the same semantics or the core LLVM "replace all uses with" concept
Chris Lattner48a109c2009-09-07 22:52:39 +0000[diff] [blame]2147would not hold.</p>
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2148
2149<div class="doc_code">
2150<pre>
2151 %A = fdiv undef, %X
2152 %B = fdiv %X, undef
2153Safe:
2154 %A = undef
2155b: unreachable
2156</pre>
2157</div>
2158
2159<p>These examples show the crucial difference between an <em>undefined
2160value</em> and <em>undefined behavior</em>. An undefined value (like undef) is
2161allowed to have an arbitrary bit-pattern. This means that the %A operation
2162can be constant folded to undef because the undef could be an SNaN, and fdiv is
2163not (currently) defined on SNaN's. However, in the second example, we can make
2164a more aggressive assumption: because the undef is allowed to be an arbitrary
2165value, we are allowed to assume that it could be zero. Since a divide by zero
Chris Lattner8e371aa2009-09-08 19:45:34 +0000[diff] [blame]2166has <em>undefined behavior</em>, we are allowed to assume that the operation
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2167does not execute at all. This allows us to delete the divide and all code after
2168it: since the undefined operation "can't happen", the optimizer can assume that
2169it occurs in dead code.
2170</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2171
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2172<div class="doc_code">
2173<pre>
2174a: store undef -> %X
2175b: store %X -> undef
2176Safe:
2177a: &lt;deleted&gt;
2178b: unreachable
2179</pre>
2180</div>
2181
2182<p>These examples reiterate the fdiv example: a store "of" an undefined value
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2183can be assumed to not have any effect: we can assume that the value is
Chris Lattner6e9057b2009-09-07 23:33:52 +0000[diff] [blame]2184overwritten with bits that happen to match what was already there. However, a
2185store "to" an undefined location could clobber arbitrary memory, therefore, it
2186has undefined behavior.</p>
2187
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2188</div>
2189
2190<!-- ======================================================================= -->
Chris Lattnerf9d078e2009-10-27 21:19:13 +0000[diff] [blame]2191<div class="doc_subsection"><a name="blockaddress">Addresses of Basic
2192 Blocks</a></div>
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2193<div class="doc_text">
2194
Chris Lattnercdfc9402009-11-01 01:27:45 +0000[diff] [blame]2195<p><b><tt>blockaddress(@function, %block)</tt></b></p>
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2196
2197<p>The '<tt>blockaddress</tt>' constant computes the address of the specified
Chris Lattner2dfdf2a2009-10-27 21:49:40 +0000[diff] [blame]2198 basic block in the specified function, and always has an i8* type. Taking
Chris Lattnercdfc9402009-11-01 01:27:45 +0000[diff] [blame]2199 the address of the entry block is illegal.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2200
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2201<p>This value only has defined behavior when used as an operand to the
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2202 '<a href="#i_indirectbr"><tt>indirectbr</tt></a>' instruction or for comparisons
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2203 against null. Pointer equality tests between labels addresses is undefined
2204 behavior - though, again, comparison against null is ok, and no label is
Chris Lattnerf9d078e2009-10-27 21:19:13 +0000[diff] [blame]2205 equal to the null pointer. This may also be passed around as an opaque
2206 pointer sized value as long as the bits are not inspected. This allows
Chris Lattner3fd77ce2009-10-27 21:44:20 +0000[diff] [blame]2207 <tt>ptrtoint</tt> and arithmetic to be performed on these values so long as
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2208 the original value is reconstituted before the <tt>indirectbr</tt>.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2209
Chris Lattnerf9d078e2009-10-27 21:19:13 +0000[diff] [blame]2210<p>Finally, some targets may provide defined semantics when
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2211 using the value as the operand to an inline assembly, but that is target
2212 specific.
2213 </p>
2214
2215</div>
2216
2217
2218<!-- ======================================================================= -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2219<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
2220</div>
2221
2222<div class="doc_text">
2223
2224<p>Constant expressions are used to allow expressions involving other constants
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2225 to be used as constants. Constant expressions may be of
2226 any <a href="#t_firstclass">first class</a> type and may involve any LLVM
2227 operation that does not have side effects (e.g. load and call are not
2228 supported). The following is the syntax for constant expressions:</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2229
2230<dl>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2231 <dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2232 <dd>Truncate a constant to another type. The bit size of CST must be larger
2233 than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2234
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2235 <dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2236 <dd>Zero extend a constant to another type. The bit size of CST must be
2237 smaller or equal to the bit size of TYPE. Both types must be
2238 integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2239
2240 <dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2241 <dd>Sign extend a constant to another type. The bit size of CST must be
2242 smaller or equal to the bit size of TYPE. Both types must be
2243 integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2244
2245 <dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2246 <dd>Truncate a floating point constant to another floating point type. The
2247 size of CST must be larger than the size of TYPE. Both types must be
2248 floating point.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2249
2250 <dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2251 <dd>Floating point extend a constant to another type. The size of CST must be
2252 smaller or equal to the size of TYPE. Both types must be floating
2253 point.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2254
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]2255 <dt><b><tt>fptoui ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2256 <dd>Convert a floating point constant to the corresponding unsigned integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2257 constant. TYPE must be a scalar or vector integer type. CST must be of
2258 scalar or vector floating point type. Both CST and TYPE must be scalars,
2259 or vectors of the same number of elements. If the value won't fit in the
2260 integer type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2261
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]2262 <dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2263 <dd>Convert a floating point constant to the corresponding signed integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2264 constant. TYPE must be a scalar or vector integer type. CST must be of
2265 scalar or vector floating point type. Both CST and TYPE must be scalars,
2266 or vectors of the same number of elements. If the value won't fit in the
2267 integer type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2268
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]2269 <dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2270 <dd>Convert an unsigned integer constant to the corresponding floating point
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2271 constant. TYPE must be a scalar or vector floating point type. CST must be
2272 of scalar or vector integer type. Both CST and TYPE must be scalars, or
2273 vectors of the same number of elements. If the value won't fit in the
2274 floating point type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2275
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]2276 <dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2277 <dd>Convert a signed integer constant to the corresponding floating point
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2278 constant. TYPE must be a scalar or vector floating point type. CST must be
2279 of scalar or vector integer type. Both CST and TYPE must be scalars, or
2280 vectors of the same number of elements. If the value won't fit in the
2281 floating point type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]2282
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]2283 <dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
2284 <dd>Convert a pointer typed constant to the corresponding integer constant
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2285 <tt>TYPE</tt> must be an integer type. <tt>CST</tt> must be of pointer
2286 type. The <tt>CST</tt> value is zero extended, truncated, or unchanged to
2287 make it fit in <tt>TYPE</tt>.</dd>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]2288
2289 <dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2290 <dd>Convert a integer constant to a pointer constant. TYPE must be a pointer
2291 type. CST must be of integer type. The CST value is zero extended,
2292 truncated, or unchanged to make it fit in a pointer size. This one is
2293 <i>really</i> dangerous!</dd>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]2294
2295 <dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
Chris Lattner03bbad62009-02-28 18:27:03 +0000[diff] [blame]2296 <dd>Convert a constant, CST, to another TYPE. The constraints of the operands
2297 are the same as those for the <a href="#i_bitcast">bitcast
2298 instruction</a>.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2299
2300 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
Dan Gohmandd8004d2009-07-27 21:53:46 +0000[diff] [blame]2301 <dt><b><tt>getelementptr inbounds ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2302 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2303 constants. As with the <a href="#i_getelementptr">getelementptr</a>
2304 instruction, the index list may have zero or more indexes, which are
2305 required to make sense for the type of "CSTPTR".</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2306
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]2307 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2308 <dd>Perform the <a href="#i_select">select operation</a> on constants.</dd>
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]2309
2310 <dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
2311 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
2312
2313 <dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
2314 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]2315
2316 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2317 <dd>Perform the <a href="#i_extractelement">extractelement operation</a> on
2318 constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]2319
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]2320 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2321 <dd>Perform the <a href="#i_insertelement">insertelement operation</a> on
2322 constants.</dd>
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]2323
2324 <dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2325 <dd>Perform the <a href="#i_shufflevector">shufflevector operation</a> on
2326 constants.</dd>
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]2327
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2328 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2329 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
2330 be any of the <a href="#binaryops">binary</a>
2331 or <a href="#bitwiseops">bitwise binary</a> operations. The constraints
2332 on operands are the same as those for the corresponding instruction
2333 (e.g. no bitwise operations on floating point values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2334</dl>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2335
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2336</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]2337
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2338<!-- ======================================================================= -->
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]2339<div class="doc_subsection"><a name="metadata">Metadata Nodes and Metadata Strings</a>
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2340</div>
2341
2342<div class="doc_text">
2343
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]2344<p>Metadata provides a way to attach arbitrary data to the instruction
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2345 stream without affecting the behaviour of the program. There are two
2346 metadata primitives, strings and nodes. All metadata has the
2347 <tt>metadata</tt> type and is identified in syntax by a preceding exclamation
2348 point ('<tt>!</tt>').</p>
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2349
2350<p>A metadata string is a string surrounded by double quotes. It can contain
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2351 any character by escaping non-printable characters with "\xx" where "xx" is
2352 the two digit hex code. For example: "<tt>!"test\00"</tt>".</p>
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2353
2354<p>Metadata nodes are represented with notation similar to structure constants
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]2355 (a comma separated list of elements, surrounded by braces and preceded by an
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2356 exclamation point). For example: "<tt>!{ metadata !"test\00", i32
2357 10}</tt>".</p>
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2358
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2359<p>A metadata node will attempt to track changes to the values it holds. In the
2360 event that a value is deleted, it will be replaced with a typeless
2361 "<tt>null</tt>", such as "<tt>metadata !{null, i32 10}</tt>".</p>
Nick Lewyckycb337992009-05-10 20:57:05 +0000[diff] [blame]2362
Devang Patelcd1fd252010-01-11 19:35:55 +0000[diff] [blame]2363<p>A <a href="#namedmetadatastructure">named metadata</a> is a collection of
2364 metadata nodes. For example: "<tt>!foo = metadata !{!4, !3}</tt>".
Devang Patel3e30c2a2010-01-05 20:41:31 +0000[diff] [blame]2365
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2366<p>Optimizations may rely on metadata to provide additional information about
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2367 the program that isn't available in the instructions, or that isn't easily
2368 computable. Similarly, the code generator may expect a certain metadata
2369 format to be used to express debugging information.</p>
2370
Nick Lewycky21cc4462009-04-04 07:22:01 +0000[diff] [blame]2371</div>
2372
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2373<!-- *********************************************************************** -->
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2374<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
2375<!-- *********************************************************************** -->
2376
2377<!-- ======================================================================= -->
2378<div class="doc_subsection">
2379<a name="inlineasm">Inline Assembler Expressions</a>
2380</div>
2381
2382<div class="doc_text">
2383
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2384<p>LLVM supports inline assembler expressions (as opposed
2385 to <a href="#moduleasm"> Module-Level Inline Assembly</a>) through the use of
2386 a special value. This value represents the inline assembler as a string
2387 (containing the instructions to emit), a list of operand constraints (stored
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2388 as a string), a flag that indicates whether or not the inline asm
Dale Johannesen8ba2d5b2009-10-21 23:28:00 +0000[diff] [blame]2389 expression has side effects, and a flag indicating whether the function
2390 containing the asm needs to align its stack conservatively. An example
2391 inline assembler expression is:</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2392
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2393<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2394<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2395i32 (i32) asm "bswap $0", "=r,r"
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2396</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2397</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2398
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2399<p>Inline assembler expressions may <b>only</b> be used as the callee operand of
2400 a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we
2401 have:</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2402
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2403<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2404<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2405%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]2406</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2407</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2408
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2409<p>Inline asms with side effects not visible in the constraint list must be
2410 marked as having side effects. This is done through the use of the
2411 '<tt>sideeffect</tt>' keyword, like so:</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2412
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2413<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2414<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2415call void asm sideeffect "eieio", ""()
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2416</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2417</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2418
Dale Johannesen8ba2d5b2009-10-21 23:28:00 +0000[diff] [blame]2419<p>In some cases inline asms will contain code that will not work unless the
2420 stack is aligned in some way, such as calls or SSE instructions on x86,
2421 yet will not contain code that does that alignment within the asm.
2422 The compiler should make conservative assumptions about what the asm might
2423 contain and should generate its usual stack alignment code in the prologue
2424 if the '<tt>alignstack</tt>' keyword is present:</p>
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2425
2426<div class="doc_code">
2427<pre>
Dale Johannesen8ba2d5b2009-10-21 23:28:00 +0000[diff] [blame]2428call void asm alignstack "eieio", ""()
Dale Johannesen09fed252009-10-13 21:56:55 +0000[diff] [blame]2429</pre>
2430</div>
2431
2432<p>If both keywords appear the '<tt>sideeffect</tt>' keyword must come
2433 first.</p>
2434
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2435<p>TODO: The format of the asm and constraints string still need to be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2436 documented here. Constraints on what can be done (e.g. duplication, moving,
2437 etc need to be documented). This is probably best done by reference to
2438 another document that covers inline asm from a holistic perspective.</p>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2439
2440</div>
2441
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2442
2443<!-- *********************************************************************** -->
2444<div class="doc_section">
2445 <a name="intrinsic_globals">Intrinsic Global Variables</a>
2446</div>
2447<!-- *********************************************************************** -->
2448
2449<p>LLVM has a number of "magic" global variables that contain data that affect
2450code generation or other IR semantics. These are documented here. All globals
Chris Lattner401e10c2009-07-20 06:14:25 +0000[diff] [blame]2451of this sort should have a section specified as "<tt>llvm.metadata</tt>". This
2452section and all globals that start with "<tt>llvm.</tt>" are reserved for use
2453by LLVM.</p>
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2454
2455<!-- ======================================================================= -->
2456<div class="doc_subsection">
2457<a name="intg_used">The '<tt>llvm.used</tt>' Global Variable</a>
2458</div>
2459
2460<div class="doc_text">
2461
2462<p>The <tt>@llvm.used</tt> global is an array with i8* element type which has <a
2463href="#linkage_appending">appending linkage</a>. This array contains a list of
2464pointers to global variables and functions which may optionally have a pointer
2465cast formed of bitcast or getelementptr. For example, a legal use of it is:</p>
2466
2467<pre>
2468 @X = global i8 4
2469 @Y = global i32 123
2470
2471 @llvm.used = appending global [2 x i8*] [
2472 i8* @X,
2473 i8* bitcast (i32* @Y to i8*)
2474 ], section "llvm.metadata"
2475</pre>
2476
2477<p>If a global variable appears in the <tt>@llvm.used</tt> list, then the
2478compiler, assembler, and linker are required to treat the symbol as if there is
2479a reference to the global that it cannot see. For example, if a variable has
2480internal linkage and no references other than that from the <tt>@llvm.used</tt>
2481list, it cannot be deleted. This is commonly used to represent references from
2482inline asms and other things the compiler cannot "see", and corresponds to
2483"attribute((used))" in GNU C.</p>
2484
2485<p>On some targets, the code generator must emit a directive to the assembler or
2486object file to prevent the assembler and linker from molesting the symbol.</p>
2487
2488</div>
2489
2490<!-- ======================================================================= -->
2491<div class="doc_subsection">
Chris Lattner401e10c2009-07-20 06:14:25 +0000[diff] [blame]2492<a name="intg_compiler_used">The '<tt>llvm.compiler.used</tt>' Global Variable</a>
2493</div>
2494
2495<div class="doc_text">
2496
2497<p>The <tt>@llvm.compiler.used</tt> directive is the same as the
2498<tt>@llvm.used</tt> directive, except that it only prevents the compiler from
2499touching the symbol. On targets that support it, this allows an intelligent
2500linker to optimize references to the symbol without being impeded as it would be
2501by <tt>@llvm.used</tt>.</p>
2502
2503<p>This is a rare construct that should only be used in rare circumstances, and
2504should not be exposed to source languages.</p>
2505
2506</div>
2507
2508<!-- ======================================================================= -->
2509<div class="doc_subsection">
Chris Lattner857755c2009-07-20 05:55:19 +0000[diff] [blame]2510<a name="intg_global_ctors">The '<tt>llvm.global_ctors</tt>' Global Variable</a>
2511</div>
2512
2513<div class="doc_text">
2514
2515<p>TODO: Describe this.</p>
2516
2517</div>
2518
2519<!-- ======================================================================= -->
2520<div class="doc_subsection">
2521<a name="intg_global_dtors">The '<tt>llvm.global_dtors</tt>' Global Variable</a>
2522</div>
2523
2524<div class="doc_text">
2525
2526<p>TODO: Describe this.</p>
2527
2528</div>
2529
2530
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2531<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2532<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
2533<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2534
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2535<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2536
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2537<p>The LLVM instruction set consists of several different classifications of
2538 instructions: <a href="#terminators">terminator
2539 instructions</a>, <a href="#binaryops">binary instructions</a>,
2540 <a href="#bitwiseops">bitwise binary instructions</a>,
2541 <a href="#memoryops">memory instructions</a>, and
2542 <a href="#otherops">other instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2543
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2544</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2545
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2546<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2547<div class="doc_subsection"> <a name="terminators">Terminator
2548Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2549
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2550<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2551
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2552<p>As mentioned <a href="#functionstructure">previously</a>, every basic block
2553 in a program ends with a "Terminator" instruction, which indicates which
2554 block should be executed after the current block is finished. These
2555 terminator instructions typically yield a '<tt>void</tt>' value: they produce
2556 control flow, not values (the one exception being the
2557 '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
2558
2559<p>There are six different terminator instructions: the
2560 '<a href="#i_ret"><tt>ret</tt></a>' instruction, the
2561 '<a href="#i_br"><tt>br</tt></a>' instruction, the
2562 '<a href="#i_switch"><tt>switch</tt></a>' instruction, the
Bill Wendling21c346e2009-11-02 00:25:26 +0000[diff] [blame]2563 '<a href="#i_indirectbr">'<tt>indirectbr</tt></a>' Instruction, the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2564 '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the
2565 '<a href="#i_unwind"><tt>unwind</tt></a>' instruction, and the
2566 '<a href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2567
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2568</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2569
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2570<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2571<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
2572Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2573
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2574<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2575
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2576<h5>Syntax:</h5>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2577<pre>
2578 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]2579 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2580</pre>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2581
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2582<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2583<p>The '<tt>ret</tt>' instruction is used to return control flow (and optionally
2584 a value) from a function back to the caller.</p>
2585
2586<p>There are two forms of the '<tt>ret</tt>' instruction: one that returns a
2587 value and then causes control flow, and one that just causes control flow to
2588 occur.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2589
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2590<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2591<p>The '<tt>ret</tt>' instruction optionally accepts a single argument, the
2592 return value. The type of the return value must be a
2593 '<a href="#t_firstclass">first class</a>' type.</p>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2594
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2595<p>A function is not <a href="#wellformed">well formed</a> if it it has a
2596 non-void return type and contains a '<tt>ret</tt>' instruction with no return
2597 value or a return value with a type that does not match its type, or if it
2598 has a void return type and contains a '<tt>ret</tt>' instruction with a
2599 return value.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2600
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2601<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2602<p>When the '<tt>ret</tt>' instruction is executed, control flow returns back to
2603 the calling function's context. If the caller is a
2604 "<a href="#i_call"><tt>call</tt></a>" instruction, execution continues at the
2605 instruction after the call. If the caller was an
2606 "<a href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues at
2607 the beginning of the "normal" destination block. If the instruction returns
2608 a value, that value shall set the call or invoke instruction's return
2609 value.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2610
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2611<h5>Example:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2612<pre>
2613 ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2614 ret void <i>; Return from a void function</i>
Bill Wendling0a4bbbf2009-02-28 22:12:54 +0000[diff] [blame]2615 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]2616</pre>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]2617
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2618</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2619<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2620<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2621
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2622<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2623
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2624<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2625<pre>
2626 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]2627</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2628
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2629<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2630<p>The '<tt>br</tt>' instruction is used to cause control flow to transfer to a
2631 different basic block in the current function. There are two forms of this
2632 instruction, corresponding to a conditional branch and an unconditional
2633 branch.</p>
2634
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2635<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2636<p>The conditional branch form of the '<tt>br</tt>' instruction takes a single
2637 '<tt>i1</tt>' value and two '<tt>label</tt>' values. The unconditional form
2638 of the '<tt>br</tt>' instruction takes a single '<tt>label</tt>' value as a
2639 target.</p>
2640
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2641<h5>Semantics:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2642<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2643 argument is evaluated. If the value is <tt>true</tt>, control flows to the
2644 '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
2645 control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
2646
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2647<h5>Example:</h5>
Bill Wendlingc39e3e02009-07-20 02:39:26 +0000[diff] [blame]2648<pre>
2649Test:
2650 %cond = <a href="#i_icmp">icmp</a> eq i32 %a, %b
2651 br i1 %cond, label %IfEqual, label %IfUnequal
2652IfEqual:
2653 <a href="#i_ret">ret</a> i32 1
2654IfUnequal:
2655 <a href="#i_ret">ret</a> i32 0
2656</pre>
2657
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2658</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2659
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2660<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2661<div class="doc_subsubsection">
2662 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
2663</div>
2664
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2665<div class="doc_text">
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2666
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2667<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2668<pre>
2669 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
2670</pre>
2671
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2672<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2673<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]2674 several different places. It is a generalization of the '<tt>br</tt>'
2675 instruction, allowing a branch to occur to one of many possible
2676 destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2677
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2678<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2679<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2680 comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination,
2681 and an array of pairs of comparison value constants and '<tt>label</tt>'s.
2682 The table is not allowed to contain duplicate constant entries.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2683
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2684<h5>Semantics:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2685<p>The <tt>switch</tt> instruction specifies a table of values and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2686 destinations. When the '<tt>switch</tt>' instruction is executed, this table
2687 is searched for the given value. If the value is found, control flow is
Benjamin Kramer8040cd32009-10-12 14:46:08 +0000[diff] [blame]2688 transferred to the corresponding destination; otherwise, control flow is
2689 transferred to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2690
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2691<h5>Implementation:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2692<p>Depending on properties of the target machine and the particular
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2693 <tt>switch</tt> instruction, this instruction may be code generated in
2694 different ways. For example, it could be generated as a series of chained
2695 conditional branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2696
2697<h5>Example:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2698<pre>
2699 <i>; Emulate a conditional br instruction</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2700 %Val = <a href="#i_zext">zext</a> i1 %value to i32
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]2701 switch i32 %Val, label %truedest [ i32 0, label %falsedest ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2702
2703 <i>; Emulate an unconditional br instruction</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2704 switch i32 0, label %dest [ ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2705
2706 <i>; Implement a jump table:</i>
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]2707 switch i32 %val, label %otherwise [ i32 0, label %onzero
2708 i32 1, label %onone
2709 i32 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2710</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2711
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2712</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2713
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2714
2715<!-- _______________________________________________________________________ -->
2716<div class="doc_subsubsection">
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2717 <a name="i_indirectbr">'<tt>indirectbr</tt>' Instruction</a>
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2718</div>
2719
2720<div class="doc_text">
2721
2722<h5>Syntax:</h5>
2723<pre>
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2724 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]2725</pre>
2726
2727<h5>Overview:</h5>
2728
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2729<p>The '<tt>indirectbr</tt>' instruction implements an indirect branch to a label
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2730 within the current function, whose address is specified by
Chris Lattnerc6f44362009-10-27 21:01:34 +0000[diff] [blame]2731 "<tt>address</tt>". Address must be derived from a <a
2732 href="#blockaddress">blockaddress</a> constant.</p>
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2733
2734<h5>Arguments:</h5>
2735
2736<p>The '<tt>address</tt>' argument is the address of the label to jump to. The
2737 rest of the arguments indicate the full set of possible destinations that the
2738 address may point to. Blocks are allowed to occur multiple times in the
2739 destination list, though this isn't particularly useful.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]2740
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2741<p>This destination list is required so that dataflow analysis has an accurate
2742 understanding of the CFG.</p>
2743
2744<h5>Semantics:</h5>
2745
2746<p>Control transfers to the block specified in the address argument. All
2747 possible destination blocks must be listed in the label list, otherwise this
2748 instruction has undefined behavior. This implies that jumps to labels
2749 defined in other functions have undefined behavior as well.</p>
2750
2751<h5>Implementation:</h5>
2752
2753<p>This is typically implemented with a jump through a register.</p>
2754
2755<h5>Example:</h5>
2756<pre>
Chris Lattnerab21db72009-10-28 00:19:10 +0000[diff] [blame]2757 indirectbr i8* %Addr, [ label %bb1, label %bb2, label %bb3 ]
Chris Lattnerf9be95f2009-10-27 19:13:16 +0000[diff] [blame]2758</pre>
2759
2760</div>
2761
2762
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2763<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2764<div class="doc_subsubsection">
2765 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
2766</div>
2767
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2768<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2769
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2770<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2771<pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]2772 &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]2773 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2774</pre>
2775
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2776<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2777<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2778 function, with the possibility of control flow transfer to either the
2779 '<tt>normal</tt>' label or the '<tt>exception</tt>' label. If the callee
2780 function returns with the "<tt><a href="#i_ret">ret</a></tt>" instruction,
2781 control flow will return to the "normal" label. If the callee (or any
2782 indirect callees) returns with the "<a href="#i_unwind"><tt>unwind</tt></a>"
2783 instruction, control is interrupted and continued at the dynamically nearest
2784 "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2785
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2786<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2787<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2788
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2789<ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2790 <li>The optional "cconv" marker indicates which <a href="#callingconv">calling
2791 convention</a> the call should use. If none is specified, the call
2792 defaults to using C calling conventions.</li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]2793
2794 <li>The optional <a href="#paramattrs">Parameter Attributes</a> list for
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2795 return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>', and
2796 '<tt>inreg</tt>' attributes are valid here.</li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]2797
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2798 <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]2799 function value being invoked. In most cases, this is a direct function
2800 invocation, but indirect <tt>invoke</tt>s are just as possible, branching
2801 off an arbitrary pointer to function value.</li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2802
2803 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2804 function to be invoked. </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2805
2806 <li>'<tt>function args</tt>': argument list whose types match the function
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2807 signature argument types. If the function signature indicates the
2808 function accepts a variable number of arguments, the extra arguments can
2809 be specified.</li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2810
2811 <li>'<tt>normal label</tt>': the label reached when the called function
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2812 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2813
2814 <li>'<tt>exception label</tt>': the label reached when a callee returns with
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2815 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2816
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]2817 <li>The optional <a href="#fnattrs">function attributes</a> list. Only
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2818 '<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
2819 '<tt>readnone</tt>' attributes are valid here.</li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2820</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2821
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2822<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2823<p>This instruction is designed to operate as a standard
2824 '<tt><a href="#i_call">call</a></tt>' instruction in most regards. The
2825 primary difference is that it establishes an association with a label, which
2826 is used by the runtime library to unwind the stack.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2827
2828<p>This instruction is used in languages with destructors to ensure that proper
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2829 cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
2830 exception. Additionally, this is important for implementation of
2831 '<tt>catch</tt>' clauses in high-level languages that support them.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2832
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2833<p>For the purposes of the SSA form, the definition of the value returned by the
2834 '<tt>invoke</tt>' instruction is deemed to occur on the edge from the current
2835 block to the "normal" label. If the callee unwinds then no return value is
2836 available.</p>
Dan Gohmanf96a4992009-05-22 21:47:08 +0000[diff] [blame]2837
Chris Lattnerdf7a6802010-01-15 18:08:37 +0000[diff] [blame]2838<p>Note that the code generator does not yet completely support unwind, and
2839that the invoke/unwind semantics are likely to change in future versions.</p>
2840
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2841<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2842<pre>
Nick Lewyckyd703f652008-03-16 07:18:12 +0000[diff] [blame]2843 %retval = invoke i32 @Test(i32 15) to label %Continue
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2844 unwind label %TestCleanup <i>; {i32}:retval set</i>
Nick Lewyckyd703f652008-03-16 07:18:12 +0000[diff] [blame]2845 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Testfnptr(i32 15) to label %Continue
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2846 unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2847</pre>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2848
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2849</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2850
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2851<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2852
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2853<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
2854Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2855
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2856<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2857
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2858<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2859<pre>
2860 unwind
2861</pre>
2862
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2863<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2864<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2865 at the first callee in the dynamic call stack which used
2866 an <a href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call.
2867 This is primarily used to implement exception handling.</p>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2868
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2869<h5>Semantics:</h5>
Chris Lattner72ed2002008-04-19 21:01:16 +0000[diff] [blame]2870<p>The '<tt>unwind</tt>' instruction causes execution of the current function to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2871 immediately halt. The dynamic call stack is then searched for the
2872 first <a href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack.
2873 Once found, execution continues at the "exceptional" destination block
2874 specified by the <tt>invoke</tt> instruction. If there is no <tt>invoke</tt>
2875 instruction in the dynamic call chain, undefined behavior results.</p>
2876
Chris Lattnerdf7a6802010-01-15 18:08:37 +0000[diff] [blame]2877<p>Note that the code generator does not yet completely support unwind, and
2878that the invoke/unwind semantics are likely to change in future versions.</p>
2879
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2880</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2881
2882<!-- _______________________________________________________________________ -->
2883
2884<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
2885Instruction</a> </div>
2886
2887<div class="doc_text">
2888
2889<h5>Syntax:</h5>
2890<pre>
2891 unreachable
2892</pre>
2893
2894<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2895<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2896 instruction is used to inform the optimizer that a particular portion of the
2897 code is not reachable. This can be used to indicate that the code after a
2898 no-return function cannot be reached, and other facts.</p>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2899
2900<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2901<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2902
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2903</div>
2904
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2905<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2906<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2907
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2908<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2909
2910<p>Binary operators are used to do most of the computation in a program. They
2911 require two operands of the same type, execute an operation on them, and
2912 produce a single value. The operands might represent multiple data, as is
2913 the case with the <a href="#t_vector">vector</a> data type. The result value
2914 has the same type as its operands.</p>
2915
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2916<p>There are several different binary operators:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]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 Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2921<div class="doc_subsubsection">
2922 <a name="i_add">'<tt>add</tt>' Instruction</a>
2923</div>
2924
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2925<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2926
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2927<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2928<pre>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]2929 &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]2930 &lt;result&gt; = add nuw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
2931 &lt;result&gt; = add nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
2932 &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]2933</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2934
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2935<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2936<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2937
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2938<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2939<p>The two arguments to the '<tt>add</tt>' instruction must
2940 be <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
2941 integer values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2942
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2943<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2944<p>The value produced is the integer sum of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2945
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2946<p>If the sum has unsigned overflow, the result returned is the mathematical
2947 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]2948
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2949<p>Because LLVM integers use a two's complement representation, this instruction
2950 is appropriate for both signed and unsigned integers.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2951
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]2952<p><tt>nuw</tt> and <tt>nsw</tt> stand for &quot;No Unsigned Wrap&quot;
2953 and &quot;No Signed Wrap&quot;, respectively. If the <tt>nuw</tt> and/or
2954 <tt>nsw</tt> keywords are present, the result value of the <tt>add</tt>
2955 is undefined if unsigned and/or signed overflow, respectively, occurs.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]2956
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2957<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2958<pre>
2959 &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2960</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2961
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2962</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2963
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2964<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2965<div class="doc_subsubsection">
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2966 <a name="i_fadd">'<tt>fadd</tt>' Instruction</a>
2967</div>
2968
2969<div class="doc_text">
2970
2971<h5>Syntax:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2972<pre>
2973 &lt;result&gt; = fadd &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
2974</pre>
2975
2976<h5>Overview:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2977<p>The '<tt>fadd</tt>' instruction returns the sum of its two operands.</p>
2978
2979<h5>Arguments:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2980<p>The two arguments to the '<tt>fadd</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2981 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
2982 floating point values. Both arguments must have identical types.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2983
2984<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2985<p>The value produced is the floating point sum of the two operands.</p>
2986
2987<h5>Example:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2988<pre>
2989 &lt;result&gt; = fadd float 4.0, %var <i>; yields {float}:result = 4.0 + %var</i>
2990</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2991
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2992</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]2993
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]2994<!-- _______________________________________________________________________ -->
2995<div class="doc_subsubsection">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2996 <a name="i_sub">'<tt>sub</tt>' Instruction</a>
2997</div>
2998
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2999<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3000
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3001<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3002<pre>
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3003 &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]3004 &lt;result&gt; = sub nuw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3005 &lt;result&gt; = sub nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3006 &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]3007</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3008
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3009<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3010<p>The '<tt>sub</tt>' instruction returns the difference of its two
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3011 operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3012
3013<p>Note that the '<tt>sub</tt>' instruction is used to represent the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3014 '<tt>neg</tt>' instruction present in most other intermediate
3015 representations.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3016
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3017<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3018<p>The two arguments to the '<tt>sub</tt>' instruction must
3019 be <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3020 integer values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3021
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3022<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3023<p>The value produced is the integer difference of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3024
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3025<p>If the difference has unsigned overflow, the result returned is the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3026 mathematical result modulo 2<sup>n</sup>, where n is the bit width of the
3027 result.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3028
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3029<p>Because LLVM integers use a two's complement representation, this instruction
3030 is appropriate for both signed and unsigned integers.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3031
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3032<p><tt>nuw</tt> and <tt>nsw</tt> stand for &quot;No Unsigned Wrap&quot;
3033 and &quot;No Signed Wrap&quot;, respectively. If the <tt>nuw</tt> and/or
3034 <tt>nsw</tt> keywords are present, the result value of the <tt>sub</tt>
3035 is undefined if unsigned and/or signed overflow, respectively, occurs.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3036
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3037<h5>Example:</h5>
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]3038<pre>
3039 &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3040 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3041</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3042
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3043</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3044
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3045<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3046<div class="doc_subsubsection">
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3047 <a name="i_fsub">'<tt>fsub</tt>' Instruction</a>
3048</div>
3049
3050<div class="doc_text">
3051
3052<h5>Syntax:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3053<pre>
3054 &lt;result&gt; = fsub &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3055</pre>
3056
3057<h5>Overview:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3058<p>The '<tt>fsub</tt>' instruction returns the difference of its two
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3059 operands.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3060
3061<p>Note that the '<tt>fsub</tt>' instruction is used to represent the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3062 '<tt>fneg</tt>' instruction present in most other intermediate
3063 representations.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3064
3065<h5>Arguments:</h5>
Bill Wendlingd9fe2982009-07-20 02:32:41 +0000[diff] [blame]3066<p>The two arguments to the '<tt>fsub</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3067 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3068 floating point values. Both arguments must have identical types.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3069
3070<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3071<p>The value produced is the floating point difference of the two operands.</p>
3072
3073<h5>Example:</h5>
3074<pre>
3075 &lt;result&gt; = fsub float 4.0, %var <i>; yields {float}:result = 4.0 - %var</i>
3076 &lt;result&gt; = fsub float -0.0, %val <i>; yields {float}:result = -%var</i>
3077</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3078
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3079</div>
3080
3081<!-- _______________________________________________________________________ -->
3082<div class="doc_subsubsection">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3083 <a name="i_mul">'<tt>mul</tt>' Instruction</a>
3084</div>
3085
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3086<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3087
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3088<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3089<pre>
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3090 &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]3091 &lt;result&gt; = mul nuw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3092 &lt;result&gt; = mul nsw &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3093 &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]3094</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3095
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3096<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3097<p>The '<tt>mul</tt>' instruction returns the product of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3098
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3099<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3100<p>The two arguments to the '<tt>mul</tt>' instruction must
3101 be <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3102 integer values. Both arguments must have identical types.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3103
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3104<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3105<p>The value produced is the integer product of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3106
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3107<p>If the result of the multiplication has unsigned overflow, the result
3108 returned is the mathematical result modulo 2<sup>n</sup>, where n is the bit
3109 width of the result.</p>
3110
3111<p>Because LLVM integers use a two's complement representation, and the result
3112 is the same width as the operands, this instruction returns the correct
3113 result for both signed and unsigned integers. If a full product
3114 (e.g. <tt>i32</tt>x<tt>i32</tt>-><tt>i64</tt>) is needed, the operands should
3115 be sign-extended or zero-extended as appropriate to the width of the full
3116 product.</p>
3117
Dan Gohman08d012e2009-07-22 22:44:56 +0000[diff] [blame]3118<p><tt>nuw</tt> and <tt>nsw</tt> stand for &quot;No Unsigned Wrap&quot;
3119 and &quot;No Signed Wrap&quot;, respectively. If the <tt>nuw</tt> and/or
3120 <tt>nsw</tt> keywords are present, the result value of the <tt>mul</tt>
3121 is undefined if unsigned and/or signed overflow, respectively, occurs.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3122
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3123<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3124<pre>
3125 &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3126</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3127
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3128</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3129
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3130<!-- _______________________________________________________________________ -->
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3131<div class="doc_subsubsection">
3132 <a name="i_fmul">'<tt>fmul</tt>' Instruction</a>
3133</div>
3134
3135<div class="doc_text">
3136
3137<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3138<pre>
3139 &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]3140</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3141
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3142<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3143<p>The '<tt>fmul</tt>' instruction returns the product of its two operands.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3144
3145<h5>Arguments:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3146<p>The two arguments to the '<tt>fmul</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3147 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3148 floating point values. Both arguments must have identical types.</p>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3149
3150<h5>Semantics:</h5>
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3151<p>The value produced is the floating point product of the two operands.</p>
3152
3153<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3154<pre>
3155 &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]3156</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3157
Dan Gohmanae3a0be2009-06-04 22:49:04 +0000[diff] [blame]3158</div>
3159
3160<!-- _______________________________________________________________________ -->
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3161<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
3162</a></div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3163
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3164<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3165
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3166<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3167<pre>
3168 &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]3169</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3170
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3171<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3172<p>The '<tt>udiv</tt>' instruction returns the quotient of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3173
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3174<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3175<p>The two arguments to the '<tt>udiv</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3176 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3177 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3178
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3179<h5>Semantics:</h5>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3180<p>The value produced is the unsigned integer quotient of the two operands.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3181
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3182<p>Note that unsigned integer division and signed integer division are distinct
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3183 operations; for signed integer division, use '<tt>sdiv</tt>'.</p>
3184
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3185<p>Division by zero leads to undefined behavior.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3186
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3187<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3188<pre>
3189 &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3190</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3191
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3192</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3193
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3194<!-- _______________________________________________________________________ -->
3195<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
3196</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3197
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3198<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3199
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3200<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3201<pre>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3202 &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]3203 &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]3204</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3205
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3206<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3207<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3208
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3209<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3210<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3211 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3212 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3213
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3214<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3215<p>The value produced is the signed integer quotient of the two operands rounded
3216 towards zero.</p>
3217
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3218<p>Note that signed integer division and unsigned integer division are distinct
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3219 operations; for unsigned integer division, use '<tt>udiv</tt>'.</p>
3220
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3221<p>Division by zero leads to undefined behavior. Overflow also leads to
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3222 undefined behavior; this is a rare case, but can occur, for example, by doing
3223 a 32-bit division of -2147483648 by -1.</p>
3224
Dan Gohman9c5beed2009-07-22 00:04:19 +0000[diff] [blame]3225<p>If the <tt>exact</tt> keyword is present, the result value of the
3226 <tt>sdiv</tt> is undefined if the result would be rounded or if overflow
3227 would occur.</p>
Dan Gohmancbb38f22009-07-20 22:41:19 +0000[diff] [blame]3228
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3229<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3230<pre>
3231 &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3232</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3233
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3234</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3235
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3236<!-- _______________________________________________________________________ -->
3237<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3238Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3239
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3240<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3241
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3242<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3243<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]3244 &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]3245</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3246
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3247<h5>Overview:</h5>
3248<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3249
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3250<h5>Arguments:</h5>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3251<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3252 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3253 floating point values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3254
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3255<h5>Semantics:</h5>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]3256<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3257
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3258<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3259<pre>
3260 &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]3261</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3262
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3263</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3264
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3265<!-- _______________________________________________________________________ -->
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3266<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
3267</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3268
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3269<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3270
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3271<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3272<pre>
3273 &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]3274</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3275
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3276<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3277<p>The '<tt>urem</tt>' instruction returns the remainder from the unsigned
3278 division of its two arguments.</p>
3279
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3280<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3281<p>The two arguments to the '<tt>urem</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3282 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3283 values. Both arguments must have identical types.</p>
3284
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3285<h5>Semantics:</h5>
3286<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3287 This instruction always performs an unsigned division to get the
3288 remainder.</p>
3289
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3290<p>Note that unsigned integer remainder and signed integer remainder are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3291 distinct operations; for signed integer remainder, use '<tt>srem</tt>'.</p>
3292
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3293<p>Taking the remainder of a division by zero leads to undefined behavior.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3294
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3295<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3296<pre>
3297 &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3298</pre>
3299
3300</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3301
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3302<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3303<div class="doc_subsubsection">
3304 <a name="i_srem">'<tt>srem</tt>' Instruction</a>
3305</div>
3306
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3307<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3308
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3309<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3310<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]3311 &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]3312</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3313
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3314<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3315<p>The '<tt>srem</tt>' instruction returns the remainder from the signed
3316 division of its two operands. This instruction can also take
3317 <a href="#t_vector">vector</a> versions of the values in which case the
3318 elements must be integers.</p>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]3319
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3320<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3321<p>The two arguments to the '<tt>srem</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3322 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3323 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3324
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3325<h5>Semantics:</h5>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3326<p>This instruction returns the <i>remainder</i> of a division (where the result
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3327 has the same sign as the dividend, <tt>op1</tt>), not the <i>modulo</i>
3328 operator (where the result has the same sign as the divisor, <tt>op2</tt>) of
3329 a value. For more information about the difference,
3330 see <a href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
3331 Math Forum</a>. For a table of how this is implemented in various languages,
3332 please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
3333 Wikipedia: modulo operation</a>.</p>
3334
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3335<p>Note that signed integer remainder and unsigned integer remainder are
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3336 distinct operations; for unsigned integer remainder, use '<tt>urem</tt>'.</p>
3337
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]3338<p>Taking the remainder of a division by zero leads to undefined behavior.
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3339 Overflow also leads to undefined behavior; this is a rare case, but can
3340 occur, for example, by taking the remainder of a 32-bit division of
3341 -2147483648 by -1. (The remainder doesn't actually overflow, but this rule
3342 lets srem be implemented using instructions that return both the result of
3343 the division and the remainder.)</p>
3344
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3345<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3346<pre>
3347 &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3348</pre>
3349
3350</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3351
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3352<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3353<div class="doc_subsubsection">
3354 <a name="i_frem">'<tt>frem</tt>' Instruction</a> </div>
3355
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3356<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3357
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3358<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3359<pre>
3360 &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]3361</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3362
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3363<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3364<p>The '<tt>frem</tt>' instruction returns the remainder from the division of
3365 its two operands.</p>
3366
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3367<h5>Arguments:</h5>
3368<p>The two arguments to the '<tt>frem</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3369 <a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a> of
3370 floating point values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3371
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3372<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3373<p>This instruction returns the <i>remainder</i> of a division. The remainder
3374 has the same sign as the dividend.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3375
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]3376<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3377<pre>
3378 &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]3379</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3380
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3381</div>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]3382
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3383<!-- ======================================================================= -->
3384<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
3385Operations</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3386
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3387<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3388
3389<p>Bitwise binary operators are used to do various forms of bit-twiddling in a
3390 program. They are generally very efficient instructions and can commonly be
3391 strength reduced from other instructions. They require two operands of the
3392 same type, execute an operation on them, and produce a single value. The
3393 resulting value is the same type as its operands.</p>
3394
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3395</div>
3396
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3397<!-- _______________________________________________________________________ -->
3398<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
3399Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3400
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3401<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3402
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3403<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3404<pre>
3405 &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]3406</pre>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3407
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3408<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3409<p>The '<tt>shl</tt>' instruction returns the first operand shifted to the left
3410 a specified number of bits.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3411
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3412<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3413<p>Both arguments to the '<tt>shl</tt>' instruction must be the
3414 same <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of
3415 integer type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3416
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3417<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3418<p>The value produced is <tt>op1</tt> * 2<sup><tt>op2</tt></sup> mod
3419 2<sup>n</sup>, where <tt>n</tt> is the width of the result. If <tt>op2</tt>
3420 is (statically or dynamically) negative or equal to or larger than the number
3421 of bits in <tt>op1</tt>, the result is undefined. If the arguments are
3422 vectors, each vector element of <tt>op1</tt> is shifted by the corresponding
3423 shift amount in <tt>op2</tt>.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3424
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3425<h5>Example:</h5>
3426<pre>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3427 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
3428 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
3429 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3430 &lt;result&gt; = shl i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]3431 &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]3432</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3433
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3434</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3435
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3436<!-- _______________________________________________________________________ -->
3437<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
3438Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3439
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3440<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3441
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3442<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3443<pre>
3444 &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]3445</pre>
3446
3447<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3448<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
3449 operand shifted to the right a specified number of bits with zero fill.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3450
3451<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3452<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3453 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3454 type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3455
3456<h5>Semantics:</h5>
3457<p>This instruction always performs a logical shift right operation. The most
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3458 significant bits of the result will be filled with zero bits after the shift.
3459 If <tt>op2</tt> is (statically or dynamically) equal to or larger than the
3460 number of bits in <tt>op1</tt>, the result is undefined. If the arguments are
3461 vectors, each vector element of <tt>op1</tt> is shifted by the corresponding
3462 shift amount in <tt>op2</tt>.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3463
3464<h5>Example:</h5>
3465<pre>
3466 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
3467 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
3468 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
3469 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3470 &lt;result&gt; = lshr i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]3471 &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]3472</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3473
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3474</div>
3475
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]3476<!-- _______________________________________________________________________ -->
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3477<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
3478Instruction</a> </div>
3479<div class="doc_text">
3480
3481<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3482<pre>
3483 &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]3484</pre>
3485
3486<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3487<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
3488 operand shifted to the right a specified number of bits with sign
3489 extension.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3490
3491<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3492<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3493 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3494 type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3495
3496<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3497<p>This instruction always performs an arithmetic shift right operation, The
3498 most significant bits of the result will be filled with the sign bit
3499 of <tt>op1</tt>. If <tt>op2</tt> is (statically or dynamically) equal to or
3500 larger than the number of bits in <tt>op1</tt>, the result is undefined. If
3501 the arguments are vectors, each vector element of <tt>op1</tt> is shifted by
3502 the corresponding shift amount in <tt>op2</tt>.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3503
3504<h5>Example:</h5>
3505<pre>
3506 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
3507 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
3508 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
3509 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]3510 &lt;result&gt; = ashr i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]3511 &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]3512</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3513
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]3514</div>
3515
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3516<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3517<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
3518Instruction</a> </div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3519
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3520<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3521
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3522<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3523<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]3524 &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]3525</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3526
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3527<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3528<p>The '<tt>and</tt>' instruction returns the bitwise logical and of its two
3529 operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3530
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3531<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3532<p>The two arguments to the '<tt>and</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3533 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3534 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3535
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3536<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3537<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3538
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3539<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3540 <tbody>
3541 <tr>
3542 <td>In0</td>
3543 <td>In1</td>
3544 <td>Out</td>
3545 </tr>
3546 <tr>
3547 <td>0</td>
3548 <td>0</td>
3549 <td>0</td>
3550 </tr>
3551 <tr>
3552 <td>0</td>
3553 <td>1</td>
3554 <td>0</td>
3555 </tr>
3556 <tr>
3557 <td>1</td>
3558 <td>0</td>
3559 <td>0</td>
3560 </tr>
3561 <tr>
3562 <td>1</td>
3563 <td>1</td>
3564 <td>1</td>
3565 </tr>
3566 </tbody>
3567</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3568
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3569<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3570<pre>
3571 &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]3572 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
3573 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3574</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3575</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3576<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3577<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3578
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3579<div class="doc_text">
3580
3581<h5>Syntax:</h5>
3582<pre>
3583 &lt;result&gt; = or &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
3584</pre>
3585
3586<h5>Overview:</h5>
3587<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive or of its
3588 two operands.</p>
3589
3590<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3591<p>The two arguments to the '<tt>or</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3592 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3593 values. Both arguments must have identical types.</p>
3594
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3595<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3596<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3597
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3598<table border="1" cellspacing="0" cellpadding="4">
3599 <tbody>
3600 <tr>
3601 <td>In0</td>
3602 <td>In1</td>
3603 <td>Out</td>
3604 </tr>
3605 <tr>
3606 <td>0</td>
3607 <td>0</td>
3608 <td>0</td>
3609 </tr>
3610 <tr>
3611 <td>0</td>
3612 <td>1</td>
3613 <td>1</td>
3614 </tr>
3615 <tr>
3616 <td>1</td>
3617 <td>0</td>
3618 <td>1</td>
3619 </tr>
3620 <tr>
3621 <td>1</td>
3622 <td>1</td>
3623 <td>1</td>
3624 </tr>
3625 </tbody>
3626</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3627
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3628<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3629<pre>
3630 &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3631 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
3632 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3633</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3634
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3635</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3636
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3637<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3638<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
3639Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3640
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3641<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3642
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3643<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3644<pre>
3645 &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]3646</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3647
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3648<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3649<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive or of
3650 its two operands. The <tt>xor</tt> is used to implement the "one's
3651 complement" operation, which is the "~" operator in C.</p>
3652
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3653<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3654<p>The two arguments to the '<tt>xor</tt>' instruction must be
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3655 <a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
3656 values. Both arguments must have identical types.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]3657
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3658<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3659<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3660
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3661<table border="1" cellspacing="0" cellpadding="4">
3662 <tbody>
3663 <tr>
3664 <td>In0</td>
3665 <td>In1</td>
3666 <td>Out</td>
3667 </tr>
3668 <tr>
3669 <td>0</td>
3670 <td>0</td>
3671 <td>0</td>
3672 </tr>
3673 <tr>
3674 <td>0</td>
3675 <td>1</td>
3676 <td>1</td>
3677 </tr>
3678 <tr>
3679 <td>1</td>
3680 <td>0</td>
3681 <td>1</td>
3682 </tr>
3683 <tr>
3684 <td>1</td>
3685 <td>1</td>
3686 <td>0</td>
3687 </tr>
3688 </tbody>
3689</table>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3690
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3691<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3692<pre>
3693 &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3694 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
3695 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
3696 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3697</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3698
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3699</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3700
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3701<!-- ======================================================================= -->
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3702<div class="doc_subsection">
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3703 <a name="vectorops">Vector Operations</a>
3704</div>
3705
3706<div class="doc_text">
3707
3708<p>LLVM supports several instructions to represent vector operations in a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3709 target-independent manner. These instructions cover the element-access and
3710 vector-specific operations needed to process vectors effectively. While LLVM
3711 does directly support these vector operations, many sophisticated algorithms
3712 will want to use target-specific intrinsics to take full advantage of a
3713 specific target.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3714
3715</div>
3716
3717<!-- _______________________________________________________________________ -->
3718<div class="doc_subsubsection">
3719 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
3720</div>
3721
3722<div class="doc_text">
3723
3724<h5>Syntax:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3725<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3726 &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]3727</pre>
3728
3729<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3730<p>The '<tt>extractelement</tt>' instruction extracts a single scalar element
3731 from a vector at a specified index.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3732
3733
3734<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3735<p>The first operand of an '<tt>extractelement</tt>' instruction is a value
3736 of <a href="#t_vector">vector</a> type. The second operand is an index
3737 indicating the position from which to extract the element. The index may be
3738 a variable.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3739
3740<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3741<p>The result is a scalar of the same type as the element type of
3742 <tt>val</tt>. Its value is the value at position <tt>idx</tt> of
3743 <tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
3744 results are undefined.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3745
3746<h5>Example:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3747<pre>
Gabor Greifa5b6f452009-10-28 13:14:50 +0000[diff] [blame]3748 &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]3749</pre>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3750
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3751</div>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3752
3753<!-- _______________________________________________________________________ -->
3754<div class="doc_subsubsection">
3755 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
3756</div>
3757
3758<div class="doc_text">
3759
3760<h5>Syntax:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3761<pre>
Dan Gohmanf3480b92008-05-12 23:38:42 +0000[diff] [blame]3762 &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]3763</pre>
3764
3765<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3766<p>The '<tt>insertelement</tt>' instruction inserts a scalar element into a
3767 vector at a specified index.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3768
3769<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3770<p>The first operand of an '<tt>insertelement</tt>' instruction is a value
3771 of <a href="#t_vector">vector</a> type. The second operand is a scalar value
3772 whose type must equal the element type of the first operand. The third
3773 operand is an index indicating the position at which to insert the value.
3774 The index may be a variable.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3775
3776<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3777<p>The result is a vector of the same type as <tt>val</tt>. Its element values
3778 are those of <tt>val</tt> except at position <tt>idx</tt>, where it gets the
3779 value <tt>elt</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
3780 results are undefined.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3781
3782<h5>Example:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3783<pre>
Gabor Greifa5b6f452009-10-28 13:14:50 +0000[diff] [blame]3784 &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]3785</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3786
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3787</div>
3788
3789<!-- _______________________________________________________________________ -->
3790<div class="doc_subsubsection">
3791 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
3792</div>
3793
3794<div class="doc_text">
3795
3796<h5>Syntax:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3797<pre>
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]3798 &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]3799</pre>
3800
3801<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3802<p>The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
3803 from two input vectors, returning a vector with the same element type as the
3804 input and length that is the same as the shuffle mask.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3805
3806<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3807<p>The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
3808 with types that match each other. The third argument is a shuffle mask whose
3809 element type is always 'i32'. The result of the instruction is a vector
3810 whose length is the same as the shuffle mask and whose element type is the
3811 same as the element type of the first two operands.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3812
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3813<p>The shuffle mask operand is required to be a constant vector with either
3814 constant integer or undef values.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3815
3816<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3817<p>The elements of the two input vectors are numbered from left to right across
3818 both of the vectors. The shuffle mask operand specifies, for each element of
3819 the result vector, which element of the two input vectors the result element
3820 gets. The element selector may be undef (meaning "don't care") and the
3821 second operand may be undef if performing a shuffle from only one vector.</p>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3822
3823<h5>Example:</h5>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3824<pre>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3825 &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]3826 &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]3827 &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]3828 &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]3829 &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]3830 &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]3831 &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]3832 &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]3833</pre>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3834
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3835</div>
Tanya Lattner09474292006-04-14 19:24:33 +0000[diff] [blame]3836
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3837<!-- ======================================================================= -->
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3838<div class="doc_subsection">
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3839 <a name="aggregateops">Aggregate Operations</a>
3840</div>
3841
3842<div class="doc_text">
3843
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3844<p>LLVM supports several instructions for working with aggregate values.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3845
3846</div>
3847
3848<!-- _______________________________________________________________________ -->
3849<div class="doc_subsubsection">
3850 <a name="i_extractvalue">'<tt>extractvalue</tt>' Instruction</a>
3851</div>
3852
3853<div class="doc_text">
3854
3855<h5>Syntax:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3856<pre>
3857 &lt;result&gt; = extractvalue &lt;aggregate type&gt; &lt;val&gt;, &lt;idx&gt;{, &lt;idx&gt;}*
3858</pre>
3859
3860<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3861<p>The '<tt>extractvalue</tt>' instruction extracts the value of a struct field
3862 or array element from an aggregate value.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3863
3864<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3865<p>The first operand of an '<tt>extractvalue</tt>' instruction is a value
3866 of <a href="#t_struct">struct</a> or <a href="#t_array">array</a> type. The
3867 operands are constant indices to specify which value to extract in a similar
3868 manner as indices in a
3869 '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3870
3871<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3872<p>The result is the value at the position in the aggregate specified by the
3873 index operands.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3874
3875<h5>Example:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3876<pre>
Gabor Greifa5b6f452009-10-28 13:14:50 +0000[diff] [blame]3877 &lt;result&gt; = extractvalue {i32, float} %agg, 0 <i>; yields i32</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3878</pre>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3879
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3880</div>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3881
3882<!-- _______________________________________________________________________ -->
3883<div class="doc_subsubsection">
3884 <a name="i_insertvalue">'<tt>insertvalue</tt>' Instruction</a>
3885</div>
3886
3887<div class="doc_text">
3888
3889<h5>Syntax:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3890<pre>
Jeffrey Yasskin7a088cf2010-01-11 19:19:26 +0000[diff] [blame]3891 &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]3892</pre>
3893
3894<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3895<p>The '<tt>insertvalue</tt>' instruction inserts a value into a struct field or
3896 array element in an aggregate.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3897
3898
3899<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3900<p>The first operand of an '<tt>insertvalue</tt>' instruction is a value
3901 of <a href="#t_struct">struct</a> or <a href="#t_array">array</a> type. The
3902 second operand is a first-class value to insert. The following operands are
3903 constant indices indicating the position at which to insert the value in a
3904 similar manner as indices in a
3905 '<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction. The
3906 value to insert must have the same type as the value identified by the
3907 indices.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3908
3909<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3910<p>The result is an aggregate of the same type as <tt>val</tt>. Its value is
3911 that of <tt>val</tt> except that the value at the position specified by the
3912 indices is that of <tt>elt</tt>.</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3913
3914<h5>Example:</h5>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3915<pre>
Jeffrey Yasskin7a088cf2010-01-11 19:19:26 +0000[diff] [blame]3916 %agg1 = insertvalue {i32, float} undef, i32 1, 0 <i>; yields {i32 1, float undef}</i>
3917 %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]3918</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3919
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3920</div>
3921
3922
3923<!-- ======================================================================= -->
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]3924<div class="doc_subsection">
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]3925 <a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3926</div>
3927
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3928<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3929
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3930<p>A key design point of an SSA-based representation is how it represents
3931 memory. In LLVM, no memory locations are in SSA form, which makes things
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]3932 very simple. This section describes how to read, write, and allocate
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3933 memory in LLVM.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3934
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3935</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3936
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3937<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3938<div class="doc_subsubsection">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3939 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
3940</div>
3941
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3942<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3943
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3944<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3945<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3946 &lt;result&gt; = alloca &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3947</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3948
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3949<h5>Overview:</h5>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3950<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3951 currently executing function, to be automatically released when this function
3952 returns to its caller. The object is always allocated in the generic address
3953 space (address space zero).</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3954
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3955<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3956<p>The '<tt>alloca</tt>' instruction
3957 allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt> bytes of memory on the
3958 runtime stack, returning a pointer of the appropriate type to the program.
3959 If "NumElements" is specified, it is the number of elements allocated,
3960 otherwise "NumElements" is defaulted to be one. If a constant alignment is
3961 specified, the value result of the allocation is guaranteed to be aligned to
3962 at least that boundary. If not specified, or if zero, the target can choose
3963 to align the allocation on any convenient boundary compatible with the
3964 type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3965
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3966<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3967
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3968<h5>Semantics:</h5>
Bill Wendling871eb0a2009-05-08 20:49:29 +0000[diff] [blame]3969<p>Memory is allocated; a pointer is returned. The operation is undefined if
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3970 there is insufficient stack space for the allocation. '<tt>alloca</tt>'d
3971 memory is automatically released when the function returns. The
3972 '<tt>alloca</tt>' instruction is commonly used to represent automatic
3973 variables that must have an address available. When the function returns
3974 (either with the <tt><a href="#i_ret">ret</a></tt>
3975 or <tt><a href="#i_unwind">unwind</a></tt> instructions), the memory is
3976 reclaimed. Allocating zero bytes is legal, but the result is undefined.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3977
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3978<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3979<pre>
Dan Gohman81e21672009-01-04 23:49:44 +0000[diff] [blame]3980 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
3981 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
3982 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
3983 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3984</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3985
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3986</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3987
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3988<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3989<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
3990Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3991
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3992<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3993
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3994<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]3995<pre>
3996 &lt;result&gt; = load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;]
3997 &lt;result&gt; = volatile load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;]
3998</pre>
3999
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4000<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4001<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4002
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4003<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4004<p>The argument to the '<tt>load</tt>' instruction specifies the memory address
4005 from which to load. The pointer must point to
4006 a <a href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
4007 marked as <tt>volatile</tt>, then the optimizer is not allowed to modify the
4008 number or order of execution of this <tt>load</tt> with other
4009 volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
4010 instructions. </p>
4011
4012<p>The optional constant "align" argument specifies the alignment of the
4013 operation (that is, the alignment of the memory address). A value of 0 or an
4014 omitted "align" argument means that the operation has the preferential
4015 alignment for the target. It is the responsibility of the code emitter to
4016 ensure that the alignment information is correct. Overestimating the
4017 alignment results in an undefined behavior. Underestimating the alignment may
4018 produce less efficient code. An alignment of 1 is always safe.</p>
4019
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4020<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4021<p>The location of memory pointed to is loaded. If the value being loaded is of
4022 scalar type then the number of bytes read does not exceed the minimum number
4023 of bytes needed to hold all bits of the type. For example, loading an
4024 <tt>i24</tt> reads at most three bytes. When loading a value of a type like
4025 <tt>i20</tt> with a size that is not an integral number of bytes, the result
4026 is undefined if the value was not originally written using a store of the
4027 same type.</p>
4028
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4029<h5>Examples:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4030<pre>
4031 %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
4032 <a href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4033 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4034</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4035
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4036</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4037
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4038<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4039<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
4040Instruction</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4041
Reid Spencer035ab572006-11-09 21:18:01 +0000[diff] [blame]4042<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4043
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4044<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4045<pre>
4046 store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
Christopher Lamb2330e4d2007-04-21 08:16:25 +0000[diff] [blame]4047 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4048</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4049
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4050<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4051<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4052
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4053<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4054<p>There are two arguments to the '<tt>store</tt>' instruction: a value to store
4055 and an address at which to store it. The type of the
4056 '<tt>&lt;pointer&gt;</tt>' operand must be a pointer to
4057 the <a href="#t_firstclass">first class</a> type of the
4058 '<tt>&lt;value&gt;</tt>' operand. If the <tt>store</tt> is marked
4059 as <tt>volatile</tt>, then the optimizer is not allowed to modify the number
4060 or order of execution of this <tt>store</tt> with other
4061 volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
4062 instructions.</p>
4063
4064<p>The optional constant "align" argument specifies the alignment of the
4065 operation (that is, the alignment of the memory address). A value of 0 or an
4066 omitted "align" argument means that the operation has the preferential
4067 alignment for the target. It is the responsibility of the code emitter to
4068 ensure that the alignment information is correct. Overestimating the
4069 alignment results in an undefined behavior. Underestimating the alignment may
4070 produce less efficient code. An alignment of 1 is always safe.</p>
4071
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4072<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4073<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>' at the
4074 location specified by the '<tt>&lt;pointer&gt;</tt>' operand. If
4075 '<tt>&lt;value&gt;</tt>' is of scalar type then the number of bytes written
4076 does not exceed the minimum number of bytes needed to hold all bits of the
4077 type. For example, storing an <tt>i24</tt> writes at most three bytes. When
4078 writing a value of a type like <tt>i20</tt> with a size that is not an
4079 integral number of bytes, it is unspecified what happens to the extra bits
4080 that do not belong to the type, but they will typically be overwritten.</p>
4081
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4082<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4083<pre>
4084 %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling8c6c72d2007-10-22 05:10:05 +0000[diff] [blame]4085 store i32 3, i32* %ptr <i>; yields {void}</i>
4086 %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]4087</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4088
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]4089</div>
4090
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]4091<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4092<div class="doc_subsubsection">
4093 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
4094</div>
4095
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4096<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4097
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4098<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4099<pre>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4100 &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]4101 &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]4102</pre>
4103
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4104<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4105<p>The '<tt>getelementptr</tt>' instruction is used to get the address of a
4106 subelement of an aggregate data structure. It performs address calculation
4107 only and does not access memory.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4108
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4109<h5>Arguments:</h5>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4110<p>The first argument is always a pointer, and forms the basis of the
Chris Lattnerc8eef442009-07-29 06:44:13 +0000[diff] [blame]4111 calculation. The remaining arguments are indices that indicate which of the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4112 elements of the aggregate object are indexed. The interpretation of each
4113 index is dependent on the type being indexed into. The first index always
4114 indexes the pointer value given as the first argument, the second index
4115 indexes a value of the type pointed to (not necessarily the value directly
4116 pointed to, since the first index can be non-zero), etc. The first type
4117 indexed into must be a pointer value, subsequent types can be arrays, vectors
4118 and structs. Note that subsequent types being indexed into can never be
4119 pointers, since that would require loading the pointer before continuing
4120 calculation.</p>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4121
4122<p>The type of each index argument depends on the type it is indexing into.
Chris Lattnerc8eef442009-07-29 06:44:13 +0000[diff] [blame]4123 When indexing into a (optionally packed) structure, only <tt>i32</tt> integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4124 <b>constants</b> are allowed. When indexing into an array, pointer or
Chris Lattnerc8eef442009-07-29 06:44:13 +0000[diff] [blame]4125 vector, integers of any width are allowed, and they are not required to be
4126 constant.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4127
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4128<p>For example, let's consider a C code fragment and how it gets compiled to
4129 LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4130
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4131<div class="doc_code">
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4132<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4133struct RT {
4134 char A;
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]4135 int B[10][20];
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4136 char C;
4137};
4138struct ST {
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]4139 int X;
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4140 double Y;
4141 struct RT Z;
4142};
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4143
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]4144int *foo(struct ST *s) {
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4145 return &amp;s[1].Z.B[5][13];
4146}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4147</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4148</div>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4149
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4150<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4151
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4152<div class="doc_code">
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4153<pre>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]4154%RT = <a href="#namedtypes">type</a> { i8 , [10 x [20 x i32]], i8 }
4155%ST = <a href="#namedtypes">type</a> { i32, double, %RT }
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4156
Dan Gohman4df605b2009-07-25 02:23:48 +0000[diff] [blame]4157define i32* @foo(%ST* %s) {
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4158entry:
4159 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
4160 ret i32* %reg
4161}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4162</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4163</div>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4164
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4165<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4166<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]4167 type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
4168 }</tt>' type, a structure. The second index indexes into the third element
4169 of the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
4170 i8 }</tt>' type, another structure. The third index indexes into the second
4171 element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
4172 array. The two dimensions of the array are subscripted into, yielding an
4173 '<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a
4174 pointer to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4175
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4176<p>Note that it is perfectly legal to index partially through a structure,
4177 returning a pointer to an inner element. Because of this, the LLVM code for
4178 the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4179
4180<pre>
Dan Gohman4df605b2009-07-25 02:23:48 +0000[diff] [blame]4181 define i32* @foo(%ST* %s) {
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4182 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]4183 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
4184 %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]4185 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
4186 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
4187 ret i32* %t5
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4188 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4189</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]4190
Dan Gohmandd8004d2009-07-27 21:53:46 +0000[diff] [blame]4191<p>If the <tt>inbounds</tt> keyword is present, the result value of the
Dan Gohman0a28d182009-07-29 16:00:30 +0000[diff] [blame]4192 <tt>getelementptr</tt> is undefined if the base pointer is not an
4193 <i>in bounds</i> address of an allocated object, or if any of the addresses
Dan Gohmanb255b882009-08-20 17:08:17 +0000[diff] [blame]4194 that would be formed by successive addition of the offsets implied by the
4195 indices to the base address with infinitely precise arithmetic are not an
4196 <i>in bounds</i> address of that allocated object.
Dan Gohman0a28d182009-07-29 16:00:30 +0000[diff] [blame]4197 The <i>in bounds</i> addresses for an allocated object are all the addresses
Dan Gohmanb255b882009-08-20 17:08:17 +0000[diff] [blame]4198 that point into the object, plus the address one byte past the end.</p>
Dan Gohmandd8004d2009-07-27 21:53:46 +0000[diff] [blame]4199
4200<p>If the <tt>inbounds</tt> keyword is not present, the offsets are added to
4201 the base address with silently-wrapping two's complement arithmetic, and
4202 the result value of the <tt>getelementptr</tt> may be outside the object
4203 pointed to by the base pointer. The result value may not necessarily be
4204 used to access memory though, even if it happens to point into allocated
4205 storage. See the <a href="#pointeraliasing">Pointer Aliasing Rules</a>
4206 section for more information.</p>
4207
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4208<p>The getelementptr instruction is often confusing. For some more insight into
4209 how it works, see <a href="GetElementPtr.html">the getelementptr FAQ</a>.</p>
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]4210
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]4211<h5>Example:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4212<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4213 <i>; yields [12 x i8]*:aptr</i>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]4214 %aptr = getelementptr {i32, [12 x i8]}* %saptr, i64 0, i32 1
4215 <i>; yields i8*:vptr</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4216 %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]4217 <i>; yields i8*:eptr</i>
4218 %eptr = getelementptr [12 x i8]* %aptr, i64 0, i32 1
Sanjiv Gupta9f805c22009-04-25 07:27:44 +0000[diff] [blame]4219 <i>; yields i32*:iptr</i>
Sanjiv Gupta16ffa802009-04-24 16:38:13 +0000[diff] [blame]4220 %iptr = getelementptr [10 x i32]* @arr, i16 0, i16 0
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4221</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4222
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]4223</div>
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]4224
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4225<!-- ======================================================================= -->
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4226<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4227</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4228
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4229<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4230
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4231<p>The instructions in this category are the conversion instructions (casting)
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4232 which all take a single operand and a type. They perform various bit
4233 conversions on the operand.</p>
4234
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4235</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4236
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4237<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4238<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4239 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
4240</div>
4241<div class="doc_text">
4242
4243<h5>Syntax:</h5>
4244<pre>
4245 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4246</pre>
4247
4248<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4249<p>The '<tt>trunc</tt>' instruction truncates its operand to the
4250 type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4251
4252<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4253<p>The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
4254 be an <a href="#t_integer">integer</a> type, and a type that specifies the
4255 size and type of the result, which must be
4256 an <a href="#t_integer">integer</a> type. The bit size of <tt>value</tt> must
4257 be larger than the bit size of <tt>ty2</tt>. Equal sized types are not
4258 allowed.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4259
4260<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4261<p>The '<tt>trunc</tt>' instruction truncates the high order bits
4262 in <tt>value</tt> and converts the remaining bits to <tt>ty2</tt>. Since the
4263 source size must be larger than the destination size, <tt>trunc</tt> cannot
4264 be a <i>no-op cast</i>. It will always truncate bits.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4265
4266<h5>Example:</h5>
4267<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4268 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4269 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4270 %Z = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4271</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4272
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4273</div>
4274
4275<!-- _______________________________________________________________________ -->
4276<div class="doc_subsubsection">
4277 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
4278</div>
4279<div class="doc_text">
4280
4281<h5>Syntax:</h5>
4282<pre>
4283 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4284</pre>
4285
4286<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4287<p>The '<tt>zext</tt>' instruction zero extends its operand to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4288 <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4289
4290
4291<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4292<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]4293 <a href="#t_integer">integer</a> type, and a type to cast it to, which must
4294 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]4295 <tt>value</tt> must be smaller than the bit size of the destination type,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4296 <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4297
4298<h5>Semantics:</h5>
4299<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]4300 bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4301
Reid Spencerb5929522007-01-12 15:46:11 +0000[diff] [blame]4302<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]4303
4304<h5>Example:</h5>
4305<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4306 %X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4307 %Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4308</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4309
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4310</div>
4311
4312<!-- _______________________________________________________________________ -->
4313<div class="doc_subsubsection">
4314 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
4315</div>
4316<div class="doc_text">
4317
4318<h5>Syntax:</h5>
4319<pre>
4320 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4321</pre>
4322
4323<h5>Overview:</h5>
4324<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
4325
4326<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4327<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]4328 <a href="#t_integer">integer</a> type, and a type to cast it to, which must
4329 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]4330 <tt>value</tt> must be smaller than the bit size of the destination type,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4331 <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4332
4333<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4334<p>The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
4335 bit (highest order bit) of the <tt>value</tt> until it reaches the bit size
4336 of the type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4337
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4338<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]4339
4340<h5>Example:</h5>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4341<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4342 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4343 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4344</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4345
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4346</div>
4347
4348<!-- _______________________________________________________________________ -->
4349<div class="doc_subsubsection">
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4350 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
4351</div>
4352
4353<div class="doc_text">
4354
4355<h5>Syntax:</h5>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4356<pre>
4357 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4358</pre>
4359
4360<h5>Overview:</h5>
4361<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4362 <tt>ty2</tt>.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4363
4364<h5>Arguments:</h5>
4365<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4366 point</a> value to cast and a <a href="#t_floating">floating point</a> type
4367 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]4368 <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]4369 <i>no-op cast</i>.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4370
4371<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4372<p>The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4373 <a href="#t_floating">floating point</a> type to a smaller
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4374 <a href="#t_floating">floating point</a> type. If the value cannot fit
4375 within the destination type, <tt>ty2</tt>, then the results are
4376 undefined.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4377
4378<h5>Example:</h5>
4379<pre>
4380 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
4381 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
4382</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4383
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]4384</div>
4385
4386<!-- _______________________________________________________________________ -->
4387<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4388 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
4389</div>
4390<div class="doc_text">
4391
4392<h5>Syntax:</h5>
4393<pre>
4394 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4395</pre>
4396
4397<h5>Overview:</h5>
4398<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]4399 floating point value.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4400
4401<h5>Arguments:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4402<p>The '<tt>fpext</tt>' instruction takes a
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4403 <a href="#t_floating">floating point</a> <tt>value</tt> to cast, and
4404 a <a href="#t_floating">floating point</a> type to cast it to. The source
4405 type must be smaller than the destination type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4406
4407<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4408<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4409 <a href="#t_floating">floating point</a> type to a larger
4410 <a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
4411 used to make a <i>no-op cast</i> because it always changes bits. Use
4412 <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]4413
4414<h5>Example:</h5>
4415<pre>
4416 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
4417 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
4418</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4419
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4420</div>
4421
4422<!-- _______________________________________________________________________ -->
4423<div class="doc_subsubsection">
Reid Spencer24d6da52007-01-21 00:29:26 +0000[diff] [blame]4424 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4425</div>
4426<div class="doc_text">
4427
4428<h5>Syntax:</h5>
4429<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]4430 &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]4431</pre>
4432
4433<h5>Overview:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]4434<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4435 unsigned integer equivalent of type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4436
4437<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4438<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
4439 scalar or vector <a href="#t_floating">floating point</a> value, and a type
4440 to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
4441 type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
4442 vector integer type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4443
4444<h5>Semantics:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4445<p>The '<tt>fptoui</tt>' instruction converts its
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4446 <a href="#t_floating">floating point</a> operand into the nearest (rounding
4447 towards zero) unsigned integer value. If the value cannot fit
4448 in <tt>ty2</tt>, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4449
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4450<h5>Example:</h5>
4451<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]4452 %X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]4453 %Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4454 %Z = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4455</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4456
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4457</div>
4458
4459<!-- _______________________________________________________________________ -->
4460<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4461 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4462</div>
4463<div class="doc_text">
4464
4465<h5>Syntax:</h5>
4466<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4467 &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]4468</pre>
4469
4470<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4471<p>The '<tt>fptosi</tt>' instruction converts
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4472 <a href="#t_floating">floating point</a> <tt>value</tt> to
4473 type <tt>ty2</tt>.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4474
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4475<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4476<p>The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
4477 scalar or vector <a href="#t_floating">floating point</a> value, and a type
4478 to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
4479 type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
4480 vector integer type with the same number of elements as <tt>ty</tt></p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4481
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4482<h5>Semantics:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4483<p>The '<tt>fptosi</tt>' instruction converts its
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4484 <a href="#t_floating">floating point</a> operand into the nearest (rounding
4485 towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
4486 the results are undefined.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4487
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]4488<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4489<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4490 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]4491 %Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Gabor Greif6a292012009-10-28 09:21:30 +0000[diff] [blame]4492 %Z = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4493</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4494
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4495</div>
4496
4497<!-- _______________________________________________________________________ -->
4498<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4499 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4500</div>
4501<div class="doc_text">
4502
4503<h5>Syntax:</h5>
4504<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4505 &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]4506</pre>
4507
4508<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4509<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4510 integer and converts that value to the <tt>ty2</tt> type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4511
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4512<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]4513<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]4514 scalar or vector <a href="#t_integer">integer</a> value, and a type to cast
4515 it to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
4516 type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
4517 floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4518
4519<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4520<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4521 integer quantity and converts it to the corresponding floating point
4522 value. If the value cannot fit in the floating point value, the results are
4523 undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4524
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4525<h5>Example:</h5>
4526<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4527 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4528 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4529</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4530
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4531</div>
4532
4533<!-- _______________________________________________________________________ -->
4534<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4535 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4536</div>
4537<div class="doc_text">
4538
4539<h5>Syntax:</h5>
4540<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]4541 &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]4542</pre>
4543
4544<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4545<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed integer
4546 and converts that value to the <tt>ty2</tt> type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4547
4548<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]4549<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]4550 scalar or vector <a href="#t_integer">integer</a> value, and a type to cast
4551 it to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
4552 type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
4553 floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4554
4555<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4556<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed integer
4557 quantity and converts it to the corresponding floating point value. If the
4558 value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4559
4560<h5>Example:</h5>
4561<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4562 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4563 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4564</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4565
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4566</div>
4567
4568<!-- _______________________________________________________________________ -->
4569<div class="doc_subsubsection">
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4570 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
4571</div>
4572<div class="doc_text">
4573
4574<h5>Syntax:</h5>
4575<pre>
4576 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4577</pre>
4578
4579<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4580<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
4581 the integer type <tt>ty2</tt>.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4582
4583<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4584<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
4585 must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
4586 <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]4587
4588<h5>Semantics:</h5>
4589<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4590 <tt>ty2</tt> by interpreting the pointer value as an integer and either
4591 truncating or zero extending that value to the size of the integer type. If
4592 <tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
4593 <tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
4594 are the same size, then nothing is done (<i>no-op cast</i>) other than a type
4595 change.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4596
4597<h5>Example:</h5>
4598<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4599 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
4600 %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]4601</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4602
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4603</div>
4604
4605<!-- _______________________________________________________________________ -->
4606<div class="doc_subsubsection">
4607 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
4608</div>
4609<div class="doc_text">
4610
4611<h5>Syntax:</h5>
4612<pre>
4613 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4614</pre>
4615
4616<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4617<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to a
4618 pointer type, <tt>ty2</tt>.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4619
4620<h5>Arguments:</h5>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]4621<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4622 value to cast, and a type to cast it to, which must be a
4623 <a href="#t_pointer">pointer</a> type.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4624
4625<h5>Semantics:</h5>
4626<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4627 <tt>ty2</tt> by applying either a zero extension or a truncation depending on
4628 the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
4629 size of a pointer then a truncation is done. If <tt>value</tt> is smaller
4630 than the size of a pointer then a zero extension is done. If they are the
4631 same size, nothing is done (<i>no-op cast</i>).</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4632
4633<h5>Example:</h5>
4634<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4635 %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]4636 %Y = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
4637 %Z = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4638</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4639
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4640</div>
4641
4642<!-- _______________________________________________________________________ -->
4643<div class="doc_subsubsection">
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4644 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4645</div>
4646<div class="doc_text">
4647
4648<h5>Syntax:</h5>
4649<pre>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4650 &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]4651</pre>
4652
4653<h5>Overview:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4654<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4655 <tt>ty2</tt> without changing any bits.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4656
4657<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4658<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be a
4659 non-aggregate first class value, and a type to cast it to, which must also be
4660 a non-aggregate <a href="#t_firstclass">first class</a> type. The bit sizes
4661 of <tt>value</tt> and the destination type, <tt>ty2</tt>, must be
4662 identical. If the source type is a pointer, the destination type must also be
4663 a pointer. This instruction supports bitwise conversion of vectors to
4664 integers and to vectors of other types (as long as they have the same
4665 size).</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4666
4667<h5>Semantics:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4668<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4669 <tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
4670 this conversion. The conversion is done as if the <tt>value</tt> had been
4671 stored to memory and read back as type <tt>ty2</tt>. Pointer types may only
4672 be converted to other pointer types with this instruction. To convert
4673 pointers to other types, use the <a href="#i_inttoptr">inttoptr</a> or
4674 <a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4675
4676<h5>Example:</h5>
4677<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4678 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4679 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4680 %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]4681</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4682
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4683</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4684
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4685<!-- ======================================================================= -->
4686<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4687
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4688<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4689
4690<p>The instructions in this category are the "miscellaneous" instructions, which
4691 defy better classification.</p>
4692
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4693</div>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4694
4695<!-- _______________________________________________________________________ -->
4696<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
4697</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4698
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4699<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4700
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4701<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4702<pre>
4703 &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]4704</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4705
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4706<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4707<p>The '<tt>icmp</tt>' instruction returns a boolean value or a vector of
4708 boolean values based on comparison of its two integer, integer vector, or
4709 pointer operands.</p>
4710
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4711<h5>Arguments:</h5>
4712<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4713 the condition code indicating the kind of comparison to perform. It is not a
4714 value, just a keyword. The possible condition code are:</p>
4715
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4716<ol>
4717 <li><tt>eq</tt>: equal</li>
4718 <li><tt>ne</tt>: not equal </li>
4719 <li><tt>ugt</tt>: unsigned greater than</li>
4720 <li><tt>uge</tt>: unsigned greater or equal</li>
4721 <li><tt>ult</tt>: unsigned less than</li>
4722 <li><tt>ule</tt>: unsigned less or equal</li>
4723 <li><tt>sgt</tt>: signed greater than</li>
4724 <li><tt>sge</tt>: signed greater or equal</li>
4725 <li><tt>slt</tt>: signed less than</li>
4726 <li><tt>sle</tt>: signed less or equal</li>
4727</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4728
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]4729<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4730 <a href="#t_pointer">pointer</a> or integer <a href="#t_vector">vector</a>
4731 typed. They must also be identical types.</p>
4732
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4733<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4734<p>The '<tt>icmp</tt>' compares <tt>op1</tt> and <tt>op2</tt> according to the
4735 condition code given as <tt>cond</tt>. The comparison performed always yields
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]4736 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]4737 result, as follows:</p>
4738
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4739<ol>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4740 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4741 <tt>false</tt> otherwise. No sign interpretation is necessary or
4742 performed.</li>
4743
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4744 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4745 <tt>false</tt> otherwise. No sign interpretation is necessary or
4746 performed.</li>
4747
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4748 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4749 <tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
4750
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4751 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4752 <tt>true</tt> if <tt>op1</tt> is greater than or equal
4753 to <tt>op2</tt>.</li>
4754
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4755 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4756 <tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
4757
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4758 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4759 <tt>true</tt> if <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
4760
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4761 <li><tt>sgt</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4762 <tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
4763
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4764 <li><tt>sge</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4765 <tt>true</tt> if <tt>op1</tt> is greater than or equal
4766 to <tt>op2</tt>.</li>
4767
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4768 <li><tt>slt</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4769 <tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
4770
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4771 <li><tt>sle</tt>: interprets the operands as signed values and yields
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4772 <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]4773</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4774
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4775<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4776 values are compared as if they were integers.</p>
4777
4778<p>If the operands are integer vectors, then they are compared element by
4779 element. The result is an <tt>i1</tt> vector with the same number of elements
4780 as the values being compared. Otherwise, the result is an <tt>i1</tt>.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4781
4782<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4783<pre>
4784 &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4785 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
4786 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
4787 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
4788 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
4789 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4790</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]4791
4792<p>Note that the code generator does not yet support vector types with
4793 the <tt>icmp</tt> instruction.</p>
4794
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4795</div>
4796
4797<!-- _______________________________________________________________________ -->
4798<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
4799</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4800
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4801<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4802
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4803<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4804<pre>
4805 &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]4806</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4807
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4808<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4809<p>The '<tt>fcmp</tt>' instruction returns a boolean value or vector of boolean
4810 values based on comparison of its operands.</p>
4811
4812<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]4813(<a href="#t_integer"><tt>i1</tt></a>).</p>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4814
4815<p>If the operands are floating point vectors, then the result type is a vector
4816 of boolean with the same number of elements as the operands being
4817 compared.</p>
4818
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4819<h5>Arguments:</h5>
4820<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4821 the condition code indicating the kind of comparison to perform. It is not a
4822 value, just a keyword. The possible condition code are:</p>
4823
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4824<ol>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4825 <li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4826 <li><tt>oeq</tt>: ordered and equal</li>
4827 <li><tt>ogt</tt>: ordered and greater than </li>
4828 <li><tt>oge</tt>: ordered and greater than or equal</li>
4829 <li><tt>olt</tt>: ordered and less than </li>
4830 <li><tt>ole</tt>: ordered and less than or equal</li>
4831 <li><tt>one</tt>: ordered and not equal</li>
4832 <li><tt>ord</tt>: ordered (no nans)</li>
4833 <li><tt>ueq</tt>: unordered or equal</li>
4834 <li><tt>ugt</tt>: unordered or greater than </li>
4835 <li><tt>uge</tt>: unordered or greater than or equal</li>
4836 <li><tt>ult</tt>: unordered or less than </li>
4837 <li><tt>ule</tt>: unordered or less than or equal</li>
4838 <li><tt>une</tt>: unordered or not equal</li>
4839 <li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4840 <li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4841</ol>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4842
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4843<p><i>Ordered</i> means that neither operand is a QNAN while
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4844 <i>unordered</i> means that either operand may be a QNAN.</p>
4845
4846<p>Each of <tt>val1</tt> and <tt>val2</tt> arguments must be either
4847 a <a href="#t_floating">floating point</a> type or
4848 a <a href="#t_vector">vector</a> of floating point type. They must have
4849 identical types.</p>
4850
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4851<h5>Semantics:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4852<p>The '<tt>fcmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4853 according to the condition code given as <tt>cond</tt>. If the operands are
4854 vectors, then the vectors are compared element by element. Each comparison
Nick Lewyckyec38da42009-09-27 00:45:11 +0000[diff] [blame]4855 performed always yields an <a href="#t_integer">i1</a> result, as
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4856 follows:</p>
4857
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4858<ol>
4859 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4860
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4861 <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]4862 <tt>op1</tt> is equal to <tt>op2</tt>.</li>
4863
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4864 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4865 <tt>op1</tt> is greather than <tt>op2</tt>.</li>
4866
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4867 <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]4868 <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
4869
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4870 <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]4871 <tt>op1</tt> is less than <tt>op2</tt>.</li>
4872
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4873 <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]4874 <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
4875
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4876 <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]4877 <tt>op1</tt> is not equal to <tt>op2</tt>.</li>
4878
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4879 <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]4880
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4881 <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]4882 <tt>op1</tt> is equal to <tt>op2</tt>.</li>
4883
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4884 <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]4885 <tt>op1</tt> is greater than <tt>op2</tt>.</li>
4886
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4887 <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]4888 <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
4889
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4890 <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]4891 <tt>op1</tt> is less than <tt>op2</tt>.</li>
4892
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4893 <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]4894 <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
4895
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]4896 <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]4897 <tt>op1</tt> is not equal to <tt>op2</tt>.</li>
4898
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4899 <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]4900
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4901 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
4902</ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4903
4904<h5>Example:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4905<pre>
4906 &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]4907 &lt;result&gt; = fcmp one float 4.0, 5.0 <i>; yields: result=true</i>
4908 &lt;result&gt; = fcmp olt float 4.0, 5.0 <i>; yields: result=true</i>
4909 &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]4910</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]4911
4912<p>Note that the code generator does not yet support vector types with
4913 the <tt>fcmp</tt> instruction.</p>
4914
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4915</div>
4916
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4917<!-- _______________________________________________________________________ -->
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4918<div class="doc_subsubsection">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4919 <a name="i_phi">'<tt>phi</tt>' Instruction</a>
4920</div>
4921
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4922<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4923
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4924<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4925<pre>
4926 &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...
4927</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4928
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4929<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4930<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in the
4931 SSA graph representing the function.</p>
4932
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4933<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4934<p>The type of the incoming values is specified with the first type field. After
4935 this, the '<tt>phi</tt>' instruction takes a list of pairs as arguments, with
4936 one pair for each predecessor basic block of the current block. Only values
4937 of <a href="#t_firstclass">first class</a> type may be used as the value
4938 arguments to the PHI node. Only labels may be used as the label
4939 arguments.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4940
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4941<p>There must be no non-phi instructions between the start of a basic block and
4942 the PHI instructions: i.e. PHI instructions must be first in a basic
4943 block.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4944
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4945<p>For the purposes of the SSA form, the use of each incoming value is deemed to
4946 occur on the edge from the corresponding predecessor block to the current
4947 block (but after any definition of an '<tt>invoke</tt>' instruction's return
4948 value on the same edge).</p>
Jay Foadd2449092009-06-03 10:20:10 +0000[diff] [blame]4949
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4950<h5>Semantics:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4951<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4952 specified by the pair corresponding to the predecessor basic block that
4953 executed just prior to the current block.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4954
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4955<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4956<pre>
4957Loop: ; Infinite loop that counts from 0 on up...
4958 %indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]
4959 %nextindvar = add i32 %indvar, 1
4960 br label %Loop
4961</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4962
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4963</div>
4964
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4965<!-- _______________________________________________________________________ -->
4966<div class="doc_subsubsection">
4967 <a name="i_select">'<tt>select</tt>' Instruction</a>
4968</div>
4969
4970<div class="doc_text">
4971
4972<h5>Syntax:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4973<pre>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4974 &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>
4975
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4976 <i>selty</i> is either i1 or {&lt;N x i1&gt;}
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4977</pre>
4978
4979<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4980<p>The '<tt>select</tt>' instruction is used to choose one value based on a
4981 condition, without branching.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4982
4983
4984<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4985<p>The '<tt>select</tt>' instruction requires an 'i1' value or a vector of 'i1'
4986 values indicating the condition, and two values of the
4987 same <a href="#t_firstclass">first class</a> type. If the val1/val2 are
4988 vectors and the condition is a scalar, then entire vectors are selected, not
4989 individual elements.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4990
4991<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4992<p>If the condition is an i1 and it evaluates to 1, the instruction returns the
4993 first value argument; otherwise, it returns the second value argument.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4994
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]4995<p>If the condition is a vector of i1, then the value arguments must be vectors
4996 of the same size, and the selection is done element by element.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4997
4998<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4999<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]5000 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5001</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]5002
5003<p>Note that the code generator does not yet support conditions
5004 with vector type.</p>
5005
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]5006</div>
5007
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]5008<!-- _______________________________________________________________________ -->
5009<div class="doc_subsubsection">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5010 <a name="i_call">'<tt>call</tt>' Instruction</a>
5011</div>
5012
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5013<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5014
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5015<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5016<pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]5017 &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]5018</pre>
5019
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5020<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5021<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5022
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5023<h5>Arguments:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5024<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5025
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]5026<ol>
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]5027 <li>The optional "tail" marker indicates that the callee function does not
5028 access any allocas or varargs in the caller. Note that calls may be
5029 marked "tail" even if they do not occur before
5030 a <a href="#i_ret"><tt>ret</tt></a> instruction. If the "tail" marker is
5031 present, the function call is eligible for tail call optimization,
5032 but <a href="CodeGenerator.html#tailcallopt">might not in fact be
5033 optimized into a jump</a>. As of this writing, the extra requirements for
5034 a call to actually be optimized are:
5035 <ul>
5036 <li>Caller and callee both have the calling
5037 convention <tt>fastcc</tt>.</li>
5038 <li>The call is in tail position (ret immediately follows call and ret
5039 uses value of call or is void).</li>
5040 <li>Option <tt>-tailcallopt</tt> is enabled,
5041 or <code>llvm::PerformTailCallOpt</code> is <code>true</code>.</li>
5042 <li><a href="CodeGenerator.html#tailcallopt">Platform specific
5043 constraints are met.</a></li>
5044 </ul>
5045 </li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]5046
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5047 <li>The optional "cconv" marker indicates which <a href="#callingconv">calling
5048 convention</a> the call should use. If none is specified, the call
Jeffrey Yasskin95fa80a2010-01-09 19:44:16 +0000[diff] [blame]5049 defaults to using C calling conventions. The calling convention of the
5050 call must match the calling convention of the target function, or else the
5051 behavior is undefined.</li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]5052
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5053 <li>The optional <a href="#paramattrs">Parameter Attributes</a> list for
5054 return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>', and
5055 '<tt>inreg</tt>' attributes are valid here.</li>
5056
5057 <li>'<tt>ty</tt>': the type of the call instruction itself which is also the
5058 type of the return value. Functions that return no value are marked
5059 <tt><a href="#t_void">void</a></tt>.</li>
5060
5061 <li>'<tt>fnty</tt>': shall be the signature of the pointer to function value
5062 being invoked. The argument types must match the types implied by this
5063 signature. This type can be omitted if the function is not varargs and if
5064 the function type does not return a pointer to a function.</li>
5065
5066 <li>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
5067 be invoked. In most cases, this is a direct function invocation, but
5068 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
5069 to function value.</li>
5070
5071 <li>'<tt>function args</tt>': argument list whose types match the function
5072 signature argument types. All arguments must be of
5073 <a href="#t_firstclass">first class</a> type. If the function signature
5074 indicates the function accepts a variable number of arguments, the extra
5075 arguments can be specified.</li>
5076
5077 <li>The optional <a href="#fnattrs">function attributes</a> list. Only
5078 '<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
5079 '<tt>readnone</tt>' attributes are valid here.</li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]5080</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5081
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5082<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5083<p>The '<tt>call</tt>' instruction is used to cause control flow to transfer to
5084 a specified function, with its incoming arguments bound to the specified
5085 values. Upon a '<tt><a href="#i_ret">ret</a></tt>' instruction in the called
5086 function, control flow continues with the instruction after the function
5087 call, and the return value of the function is bound to the result
5088 argument.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5089
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]5090<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5091<pre>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]5092 %retval = call i32 @test(i32 %argc)
Chris Lattner772fccf2008-03-21 17:24:17 +0000[diff] [blame]5093 call i32 (i8 *, ...)* @printf(i8 * %msg, i32 12, i8 42) <i>; yields i32</i>
5094 %X = tail call i32 @foo() <i>; yields i32</i>
5095 %Y = tail call <a href="#callingconv">fastcc</a> i32 @foo() <i>; yields i32</i>
5096 call void %foo(i8 97 signext)
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]5097
5098 %struct.A = type { i32, i8 }
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]5099 %r = call %struct.A @foo() <i>; yields { 32, i8 }</i>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]5100 %gr = extractvalue %struct.A %r, 0 <i>; yields i32</i>
5101 %gr1 = extractvalue %struct.A %r, 1 <i>; yields i8</i>
Chris Lattner85a350f2008-10-08 06:26:11 +0000[diff] [blame]5102 %Z = call void @foo() noreturn <i>; indicates that %foo never returns normally</i>
Matthijs Kooijmancb73d192008-10-07 10:03:45 +0000[diff] [blame]5103 %ZZ = call zeroext i32 @bar() <i>; Return value is %zero extended</i>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]5104</pre>
5105
Dale Johannesen07de8d12009-09-24 18:38:21 +0000[diff] [blame]5106<p>llvm treats calls to some functions with names and arguments that match the
Dale Johannesen9f8380b2009-09-25 17:04:42 +0000[diff] [blame]5107standard C99 library as being the C99 library functions, and may perform
5108optimizations or generate code for them under that assumption. This is
5109something we'd like to change in the future to provide better support for
5110freestanding environments and non-C-based langauges.</p>
Dale Johannesen07de8d12009-09-24 18:38:21 +0000[diff] [blame]5111
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5112</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5113
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5114<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5115<div class="doc_subsubsection">
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]5116 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5117</div>
5118
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5119<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5120
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5121<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5122<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5123 &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]5124</pre>
5125
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5126<h5>Overview:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5127<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5128 the "variable argument" area of a function call. It is used to implement the
5129 <tt>va_arg</tt> macro in C.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5130
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5131<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5132<p>This instruction takes a <tt>va_list*</tt> value and the type of the
5133 argument. It returns a value of the specified argument type and increments
5134 the <tt>va_list</tt> to point to the next argument. The actual type
5135 of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5136
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5137<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5138<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified type
5139 from the specified <tt>va_list</tt> and causes the <tt>va_list</tt> to point
5140 to the next argument. For more information, see the variable argument
5141 handling <a href="#int_varargs">Intrinsic Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5142
5143<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]5144 take a variable number of arguments, for example, the <tt>vfprintf</tt>
5145 function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5146
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5147<p><tt>va_arg</tt> is an LLVM instruction instead of
5148 an <a href="#intrinsics">intrinsic function</a> because it takes a type as an
5149 argument.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5150
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]5151<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]5152<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
5153
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5154<p>Note that the code generator does not yet fully support va_arg on many
5155 targets. Also, it does not currently support va_arg with aggregate types on
5156 any target.</p>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]5157
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5158</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5159
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5160<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]5161<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
5162<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5163
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5164<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5165
5166<p>LLVM supports the notion of an "intrinsic function". These functions have
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5167 well known names and semantics and are required to follow certain
5168 restrictions. Overall, these intrinsics represent an extension mechanism for
5169 the LLVM language that does not require changing all of the transformations
5170 in LLVM when adding to the language (or the bitcode reader/writer, the
5171 parser, etc...).</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5172
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]5173<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5174 prefix is reserved in LLVM for intrinsic names; thus, function names may not
5175 begin with this prefix. Intrinsic functions must always be external
5176 functions: you cannot define the body of intrinsic functions. Intrinsic
5177 functions may only be used in call or invoke instructions: it is illegal to
5178 take the address of an intrinsic function. Additionally, because intrinsic
5179 functions are part of the LLVM language, it is required if any are added that
5180 they be documented here.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5181
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5182<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents a
5183 family of functions that perform the same operation but on different data
5184 types. Because LLVM can represent over 8 million different integer types,
5185 overloading is used commonly to allow an intrinsic function to operate on any
5186 integer type. One or more of the argument types or the result type can be
5187 overloaded to accept any integer type. Argument types may also be defined as
5188 exactly matching a previous argument's type or the result type. This allows
5189 an intrinsic function which accepts multiple arguments, but needs all of them
5190 to be of the same type, to only be overloaded with respect to a single
5191 argument or the result.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5192
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5193<p>Overloaded intrinsics will have the names of its overloaded argument types
5194 encoded into its function name, each preceded by a period. Only those types
5195 which are overloaded result in a name suffix. Arguments whose type is matched
5196 against another type do not. For example, the <tt>llvm.ctpop</tt> function
5197 can take an integer of any width and returns an integer of exactly the same
5198 integer width. This leads to a family of functions such as
5199 <tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29
5200 %val)</tt>. Only one type, the return type, is overloaded, and only one type
5201 suffix is required. Because the argument's type is matched against the return
5202 type, it does not require its own name suffix.</p>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5203
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]5204<p>To learn how to add an intrinsic function, please see the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5205 <a href="ExtendingLLVM.html">Extending LLVM Guide</a>.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5206
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5207</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5208
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5209<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5210<div class="doc_subsection">
5211 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
5212</div>
5213
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5214<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5215
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5216<p>Variable argument support is defined in LLVM with
5217 the <a href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
5218 intrinsic functions. These functions are related to the similarly named
5219 macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5220
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5221<p>All of these functions operate on arguments that use a target-specific value
5222 type "<tt>va_list</tt>". The LLVM assembly language reference manual does
5223 not define what this type is, so all transformations should be prepared to
5224 handle these functions regardless of the type used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5225
Chris Lattner374ab302006-05-15 17:26:46 +0000[diff] [blame]5226<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]5227 instruction and the variable argument handling intrinsic functions are
5228 used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5229
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]5230<div class="doc_code">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5231<pre>
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5232define i32 @test(i32 %X, ...) {
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5233 ; Initialize variable argument processing
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5234 %ap = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5235 %ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5236 call void @llvm.va_start(i8* %ap2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5237
5238 ; Read a single integer argument
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5239 %tmp = va_arg i8** %ap, i32
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5240
5241 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5242 %aq = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5243 %aq2 = bitcast i8** %aq to i8*
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5244 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5245 call void @llvm.va_end(i8* %aq2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5246
5247 ; Stop processing of arguments.
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5248 call void @llvm.va_end(i8* %ap2)
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5249 ret i32 %tmp
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5250}
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]5251
5252declare void @llvm.va_start(i8*)
5253declare void @llvm.va_copy(i8*, i8*)
5254declare void @llvm.va_end(i8*)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5255</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5256</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5257
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]5258</div>
5259
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5260<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5261<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5262 <a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5263</div>
5264
5265
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5266<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5267
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5268<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5269<pre>
5270 declare void %llvm.va_start(i8* &lt;arglist&gt;)
5271</pre>
5272
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5273<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5274<p>The '<tt>llvm.va_start</tt>' intrinsic initializes <tt>*&lt;arglist&gt;</tt>
5275 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]5276
5277<h5>Arguments:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5278<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]5279
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5280<h5>Semantics:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5281<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]5282 macro available in C. In a target-dependent way, it initializes
5283 the <tt>va_list</tt> element to which the argument points, so that the next
5284 call to <tt>va_arg</tt> will produce the first variable argument passed to
5285 the function. Unlike the C <tt>va_start</tt> macro, this intrinsic does not
5286 need to know the last argument of the function as the compiler can figure
5287 that out.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5288
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5289</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5290
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5291<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5292<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5293 <a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5294</div>
5295
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5296<div class="doc_text">
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5297
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5298<h5>Syntax:</h5>
5299<pre>
5300 declare void @llvm.va_end(i8* &lt;arglist&gt;)
5301</pre>
5302
5303<h5>Overview:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5304<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*&lt;arglist&gt;</tt>,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5305 which has been initialized previously
5306 with <tt><a href="#int_va_start">llvm.va_start</a></tt>
5307 or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5308
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5309<h5>Arguments:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5310<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]5311
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5312<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5313<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]5314 macro available in C. In a target-dependent way, it destroys
5315 the <tt>va_list</tt> element to which the argument points. Calls
5316 to <a href="#int_va_start"><tt>llvm.va_start</tt></a>
5317 and <a href="#int_va_copy"> <tt>llvm.va_copy</tt></a> must be matched exactly
5318 with calls to <tt>llvm.va_end</tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]5319
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5320</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5321
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5322<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5323<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5324 <a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5325</div>
5326
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5327<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5328
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5329<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5330<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5331 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5332</pre>
5333
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5334<h5>Overview:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5335<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5336 from the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5337
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5338<h5>Arguments:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]5339<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]5340 The second argument is a pointer to a <tt>va_list</tt> element to copy
5341 from.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5342
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]5343<h5>Semantics:</h5>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]5344<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]5345 macro available in C. In a target-dependent way, it copies the
5346 source <tt>va_list</tt> element into the destination <tt>va_list</tt>
5347 element. This intrinsic is necessary because
5348 the <tt><a href="#int_va_start"> llvm.va_start</a></tt> intrinsic may be
5349 arbitrarily complex and require, for example, memory allocation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5350
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]5351</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5352
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5353<!-- ======================================================================= -->
5354<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5355 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
5356</div>
5357
5358<div class="doc_text">
5359
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5360<p>LLVM support for <a href="GarbageCollection.html">Accurate Garbage
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]5361Collection</a> (GC) requires the implementation and generation of these
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5362intrinsics. These intrinsics allow identification of <a href="#int_gcroot">GC
5363roots on the stack</a>, as well as garbage collector implementations that
5364require <a href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a>
5365barriers. Front-ends for type-safe garbage collected languages should generate
5366these intrinsics to make use of the LLVM garbage collectors. For more details,
5367see <a href="GarbageCollection.html">Accurate Garbage Collection with
5368LLVM</a>.</p>
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]5369
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5370<p>The garbage collection intrinsics only operate on objects in the generic
5371 address space (address space zero).</p>
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]5372
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5373</div>
5374
5375<!-- _______________________________________________________________________ -->
5376<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5377 <a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5378</div>
5379
5380<div class="doc_text">
5381
5382<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5383<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5384 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5385</pre>
5386
5387<h5>Overview:</h5>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]5388<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]5389 the code generator, and allows some metadata to be associated with it.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5390
5391<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5392<p>The first argument specifies the address of a stack object that contains the
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5393 root pointer. The second pointer (which must be either a constant or a
5394 global value address) contains the meta-data to be associated with the
5395 root.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5396
5397<h5>Semantics:</h5>
Chris Lattner05d67092008-04-24 05:59:56 +0000[diff] [blame]5398<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]5399 location. At compile-time, the code generator generates information to allow
5400 the runtime to find the pointer at GC safe points. The '<tt>llvm.gcroot</tt>'
5401 intrinsic may only be used in a function which <a href="#gc">specifies a GC
5402 algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5403
5404</div>
5405
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5406<!-- _______________________________________________________________________ -->
5407<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5408 <a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5409</div>
5410
5411<div class="doc_text">
5412
5413<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5414<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5415 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5416</pre>
5417
5418<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5419<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5420 locations, allowing garbage collector implementations that require read
5421 barriers.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5422
5423<h5>Arguments:</h5>
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]5424<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]5425 allocated from the garbage collector. The first object is a pointer to the
5426 start of the referenced object, if needed by the language runtime (otherwise
5427 null).</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5428
5429<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5430<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5431 instruction, but may be replaced with substantially more complex code by the
5432 garbage collector runtime, as needed. The '<tt>llvm.gcread</tt>' intrinsic
5433 may only be used in a function which <a href="#gc">specifies a GC
5434 algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5435
5436</div>
5437
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5438<!-- _______________________________________________________________________ -->
5439<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5440 <a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5441</div>
5442
5443<div class="doc_text">
5444
5445<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5446<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5447 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5448</pre>
5449
5450<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5451<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5452 locations, allowing garbage collector implementations that require write
5453 barriers (such as generational or reference counting collectors).</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5454
5455<h5>Arguments:</h5>
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]5456<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]5457 object to store it to, and the third is the address of the field of Obj to
5458 store to. If the runtime does not require a pointer to the object, Obj may
5459 be null.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5460
5461<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5462<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5463 instruction, but may be replaced with substantially more complex code by the
5464 garbage collector runtime, as needed. The '<tt>llvm.gcwrite</tt>' intrinsic
5465 may only be used in a function which <a href="#gc">specifies a GC
5466 algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5467
5468</div>
5469
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]5470<!-- ======================================================================= -->
5471<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5472 <a name="int_codegen">Code Generator Intrinsics</a>
5473</div>
5474
5475<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5476
5477<p>These intrinsics are provided by LLVM to expose special features that may
5478 only be implemented with code generator support.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5479
5480</div>
5481
5482<!-- _______________________________________________________________________ -->
5483<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5484 <a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5485</div>
5486
5487<div class="doc_text">
5488
5489<h5>Syntax:</h5>
5490<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5491 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5492</pre>
5493
5494<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5495<p>The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
5496 target-specific value indicating the return address of the current function
5497 or one of its callers.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5498
5499<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5500<p>The argument to this intrinsic indicates which function to return the address
5501 for. Zero indicates the calling function, one indicates its caller, etc.
5502 The argument is <b>required</b> to be a constant integer value.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5503
5504<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5505<p>The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer
5506 indicating the return address of the specified call frame, or zero if it
5507 cannot be identified. The value returned by this intrinsic is likely to be
5508 incorrect or 0 for arguments other than zero, so it should only be used for
5509 debugging purposes.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5510
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5511<p>Note that calling this intrinsic does not prevent function inlining or other
5512 aggressive transformations, so the value returned may not be that of the
5513 obvious source-language caller.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5514
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5515</div>
5516
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5517<!-- _______________________________________________________________________ -->
5518<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5519 <a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5520</div>
5521
5522<div class="doc_text">
5523
5524<h5>Syntax:</h5>
5525<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5526 declare i8 *@llvm.frameaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5527</pre>
5528
5529<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5530<p>The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
5531 target-specific frame pointer value for the specified stack frame.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5532
5533<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5534<p>The argument to this intrinsic indicates which function to return the frame
5535 pointer for. Zero indicates the calling function, one indicates its caller,
5536 etc. The argument is <b>required</b> to be a constant integer value.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5537
5538<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5539<p>The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer
5540 indicating the frame address of the specified call frame, or zero if it
5541 cannot be identified. The value returned by this intrinsic is likely to be
5542 incorrect or 0 for arguments other than zero, so it should only be used for
5543 debugging purposes.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5544
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5545<p>Note that calling this intrinsic does not prevent function inlining or other
5546 aggressive transformations, so the value returned may not be that of the
5547 obvious source-language caller.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5548
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5549</div>
5550
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5551<!-- _______________________________________________________________________ -->
5552<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5553 <a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5554</div>
5555
5556<div class="doc_text">
5557
5558<h5>Syntax:</h5>
5559<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5560 declare i8 *@llvm.stacksave()
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5561</pre>
5562
5563<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5564<p>The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state
5565 of the function stack, for use
5566 with <a href="#int_stackrestore"> <tt>llvm.stackrestore</tt></a>. This is
5567 useful for implementing language features like scoped automatic variable
5568 sized arrays in C99.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5569
5570<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5571<p>This intrinsic returns a opaque pointer value that can be passed
5572 to <a href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When
5573 an <tt>llvm.stackrestore</tt> intrinsic is executed with a value saved
5574 from <tt>llvm.stacksave</tt>, it effectively restores the state of the stack
5575 to the state it was in when the <tt>llvm.stacksave</tt> intrinsic executed.
5576 In practice, this pops any <a href="#i_alloca">alloca</a> blocks from the
5577 stack that were allocated after the <tt>llvm.stacksave</tt> was executed.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5578
5579</div>
5580
5581<!-- _______________________________________________________________________ -->
5582<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5583 <a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5584</div>
5585
5586<div class="doc_text">
5587
5588<h5>Syntax:</h5>
5589<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5590 declare void @llvm.stackrestore(i8 * %ptr)
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5591</pre>
5592
5593<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5594<p>The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
5595 the function stack to the state it was in when the
5596 corresponding <a href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic
5597 executed. This is useful for implementing language features like scoped
5598 automatic variable sized arrays in C99.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5599
5600<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5601<p>See the description
5602 for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.</p>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5603
5604</div>
5605
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5606<!-- _______________________________________________________________________ -->
5607<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5608 <a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5609</div>
5610
5611<div class="doc_text">
5612
5613<h5>Syntax:</h5>
5614<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5615 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]5616</pre>
5617
5618<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5619<p>The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to
5620 insert a prefetch instruction if supported; otherwise, it is a noop.
5621 Prefetches have no effect on the behavior of the program but can change its
5622 performance characteristics.</p>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5623
5624<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5625<p><tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the
5626 specifier determining if the fetch should be for a read (0) or write (1),
5627 and <tt>locality</tt> is a temporal locality specifier ranging from (0) - no
5628 locality, to (3) - extremely local keep in cache. The <tt>rw</tt>
5629 and <tt>locality</tt> arguments must be constant integers.</p>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5630
5631<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5632<p>This intrinsic does not modify the behavior of the program. In particular,
5633 prefetches cannot trap and do not produce a value. On targets that support
5634 this intrinsic, the prefetch can provide hints to the processor cache for
5635 better performance.</p>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5636
5637</div>
5638
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5639<!-- _______________________________________________________________________ -->
5640<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5641 <a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5642</div>
5643
5644<div class="doc_text">
5645
5646<h5>Syntax:</h5>
5647<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5648 declare void @llvm.pcmarker(i32 &lt;id&gt;)
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5649</pre>
5650
5651<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5652<p>The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program
5653 Counter (PC) in a region of code to simulators and other tools. The method
5654 is target specific, but it is expected that the marker will use exported
5655 symbols to transmit the PC of the marker. The marker makes no guarantees
5656 that it will remain with any specific instruction after optimizations. It is
5657 possible that the presence of a marker will inhibit optimizations. The
5658 intended use is to be inserted after optimizations to allow correlations of
5659 simulation runs.</p>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5660
5661<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5662<p><tt>id</tt> is a numerical id identifying the marker.</p>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5663
5664<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5665<p>This intrinsic does not modify the behavior of the program. Backends that do
5666 not support this intrinisic may ignore it.</p>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5667
5668</div>
5669
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5670<!-- _______________________________________________________________________ -->
5671<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5672 <a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5673</div>
5674
5675<div class="doc_text">
5676
5677<h5>Syntax:</h5>
5678<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5679 declare i64 @llvm.readcyclecounter( )
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5680</pre>
5681
5682<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5683<p>The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
5684 counter register (or similar low latency, high accuracy clocks) on those
5685 targets that support it. On X86, it should map to RDTSC. On Alpha, it
5686 should map to RPCC. As the backing counters overflow quickly (on the order
5687 of 9 seconds on alpha), this should only be used for small timings.</p>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5688
5689<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5690<p>When directly supported, reading the cycle counter should not modify any
5691 memory. Implementations are allowed to either return a application specific
5692 value or a system wide value. On backends without support, this is lowered
5693 to a constant 0.</p>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5694
5695</div>
5696
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5697<!-- ======================================================================= -->
5698<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5699 <a name="int_libc">Standard C Library Intrinsics</a>
5700</div>
5701
5702<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5703
5704<p>LLVM provides intrinsics for a few important standard C library functions.
5705 These intrinsics allow source-language front-ends to pass information about
5706 the alignment of the pointer arguments to the code generator, providing
5707 opportunity for more efficient code generation.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5708
5709</div>
5710
5711<!-- _______________________________________________________________________ -->
5712<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5713 <a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5714</div>
5715
5716<div class="doc_text">
5717
5718<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5719<p>This is an overloaded intrinsic. You can use <tt>llvm.memcpy</tt> on any
5720 integer bit width. Not all targets support all bit widths however.</p>
5721
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5722<pre>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5723 declare void @llvm.memcpy.i8(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5724 i8 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5725 declare void @llvm.memcpy.i16(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
5726 i16 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5727 declare void @llvm.memcpy.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5728 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5729 declare void @llvm.memcpy.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5730 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5731</pre>
5732
5733<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5734<p>The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the
5735 source location to the destination location.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5736
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5737<p>Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
5738 intrinsics do not return a value, and takes an extra alignment argument.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5739
5740<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5741<p>The first argument is a pointer to the destination, the second is a pointer
5742 to the source. The third argument is an integer argument specifying the
5743 number of bytes to copy, and the fourth argument is the alignment of the
5744 source and destination locations.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5745
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5746<p>If the call to this intrinisic has an alignment value that is not 0 or 1,
5747 then the caller guarantees that both the source and destination pointers are
5748 aligned to that boundary.</p>
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]5749
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5750<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5751<p>The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the
5752 source location to the destination location, which are not allowed to
5753 overlap. It copies "len" bytes of memory over. If the argument is known to
5754 be aligned to some boundary, this can be specified as the fourth argument,
5755 otherwise it should be set to 0 or 1.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5756
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5757</div>
5758
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5759<!-- _______________________________________________________________________ -->
5760<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5761 <a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5762</div>
5763
5764<div class="doc_text">
5765
5766<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5767<p>This is an overloaded intrinsic. You can use llvm.memmove on any integer bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5768 width. Not all targets support all bit widths however.</p>
5769
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5770<pre>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5771 declare void @llvm.memmove.i8(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5772 i8 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5773 declare void @llvm.memmove.i16(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
5774 i16 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5775 declare void @llvm.memmove.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5776 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5777 declare void @llvm.memmove.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5778 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5779</pre>
5780
5781<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5782<p>The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the
5783 source location to the destination location. It is similar to the
5784 '<tt>llvm.memcpy</tt>' intrinsic but allows the two memory locations to
5785 overlap.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5786
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5787<p>Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
5788 intrinsics do not return a value, and takes an extra alignment argument.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5789
5790<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5791<p>The first argument is a pointer to the destination, the second is a pointer
5792 to the source. The third argument is an integer argument specifying the
5793 number of bytes to copy, and the fourth argument is the alignment of the
5794 source and destination locations.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5795
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5796<p>If the call to this intrinisic has an alignment value that is not 0 or 1,
5797 then the caller guarantees that the source and destination pointers are
5798 aligned to that boundary.</p>
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]5799
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5800<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5801<p>The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the
5802 source location to the destination location, which may overlap. It copies
5803 "len" bytes of memory over. If the argument is known to be aligned to some
5804 boundary, this can be specified as the fourth argument, otherwise it should
5805 be set to 0 or 1.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5806
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5807</div>
5808
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5809<!-- _______________________________________________________________________ -->
5810<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5811 <a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5812</div>
5813
5814<div class="doc_text">
5815
5816<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5817<p>This is an overloaded intrinsic. You can use llvm.memset on any integer bit
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5818 width. Not all targets support all bit widths however.</p>
5819
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5820<pre>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5821 declare void @llvm.memset.i8(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5822 i8 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5823 declare void @llvm.memset.i16(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
5824 i16 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5825 declare void @llvm.memset.i32(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5826 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5827 declare void @llvm.memset.i64(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5828 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5829</pre>
5830
5831<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5832<p>The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a
5833 particular byte value.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5834
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5835<p>Note that, unlike the standard libc function, the <tt>llvm.memset</tt>
5836 intrinsic does not return a value, and takes an extra alignment argument.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5837
5838<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5839<p>The first argument is a pointer to the destination to fill, the second is the
5840 byte value to fill it with, the third argument is an integer argument
5841 specifying the number of bytes to fill, and the fourth argument is the known
5842 alignment of destination location.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5843
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5844<p>If the call to this intrinisic has an alignment value that is not 0 or 1,
5845 then the caller guarantees that the destination pointer is aligned to that
5846 boundary.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5847
5848<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5849<p>The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting
5850 at the destination location. If the argument is known to be aligned to some
5851 boundary, this can be specified as the fourth argument, otherwise it should
5852 be set to 0 or 1.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5853
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5854</div>
5855
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]5856<!-- _______________________________________________________________________ -->
5857<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5858 <a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5859</div>
5860
5861<div class="doc_text">
5862
5863<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5864<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
5865 floating point or vector of floating point type. Not all targets support all
5866 types however.</p>
5867
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5868<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]5869 declare float @llvm.sqrt.f32(float %Val)
5870 declare double @llvm.sqrt.f64(double %Val)
5871 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
5872 declare fp128 @llvm.sqrt.f128(fp128 %Val)
5873 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5874</pre>
5875
5876<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5877<p>The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
5878 returning the same value as the libm '<tt>sqrt</tt>' functions would.
5879 Unlike <tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined
5880 behavior for negative numbers other than -0.0 (which allows for better
5881 optimization, because there is no need to worry about errno being
5882 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]5883
5884<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5885<p>The argument and return value are floating point numbers of the same
5886 type.</p>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5887
5888<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5889<p>This function returns the sqrt of the specified operand if it is a
5890 nonnegative floating point number.</p>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5891
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5892</div>
5893
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5894<!-- _______________________________________________________________________ -->
5895<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5896 <a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5897</div>
5898
5899<div class="doc_text">
5900
5901<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5902<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
5903 floating point or vector of floating point type. Not all targets support all
5904 types however.</p>
5905
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5906<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]5907 declare float @llvm.powi.f32(float %Val, i32 %power)
5908 declare double @llvm.powi.f64(double %Val, i32 %power)
5909 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
5910 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
5911 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5912</pre>
5913
5914<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5915<p>The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
5916 specified (positive or negative) power. The order of evaluation of
5917 multiplications is not defined. When a vector of floating point type is
5918 used, the second argument remains a scalar integer value.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5919
5920<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5921<p>The second argument is an integer power, and the first is a value to raise to
5922 that power.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5923
5924<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5925<p>This function returns the first value raised to the second power with an
5926 unspecified sequence of rounding operations.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5927
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5928</div>
5929
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5930<!-- _______________________________________________________________________ -->
5931<div class="doc_subsubsection">
5932 <a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
5933</div>
5934
5935<div class="doc_text">
5936
5937<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5938<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
5939 floating point or vector of floating point type. Not all targets support all
5940 types however.</p>
5941
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5942<pre>
5943 declare float @llvm.sin.f32(float %Val)
5944 declare double @llvm.sin.f64(double %Val)
5945 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
5946 declare fp128 @llvm.sin.f128(fp128 %Val)
5947 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
5948</pre>
5949
5950<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5951<p>The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5952
5953<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5954<p>The argument and return value are floating point numbers of the same
5955 type.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5956
5957<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5958<p>This function returns the sine of the specified operand, returning the same
5959 values as the libm <tt>sin</tt> functions would, and handles error conditions
5960 in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5961
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5962</div>
5963
5964<!-- _______________________________________________________________________ -->
5965<div class="doc_subsubsection">
5966 <a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
5967</div>
5968
5969<div class="doc_text">
5970
5971<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5972<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
5973 floating point or vector of floating point type. Not all targets support all
5974 types however.</p>
5975
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5976<pre>
5977 declare float @llvm.cos.f32(float %Val)
5978 declare double @llvm.cos.f64(double %Val)
5979 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
5980 declare fp128 @llvm.cos.f128(fp128 %Val)
5981 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
5982</pre>
5983
5984<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5985<p>The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5986
5987<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5988<p>The argument and return value are floating point numbers of the same
5989 type.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5990
5991<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]5992<p>This function returns the cosine of the specified operand, returning the same
5993 values as the libm <tt>cos</tt> functions would, and handles error conditions
5994 in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5995
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5996</div>
5997
5998<!-- _______________________________________________________________________ -->
5999<div class="doc_subsubsection">
6000 <a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
6001</div>
6002
6003<div class="doc_text">
6004
6005<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6006<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
6007 floating point or vector of floating point type. Not all targets support all
6008 types however.</p>
6009
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6010<pre>
6011 declare float @llvm.pow.f32(float %Val, float %Power)
6012 declare double @llvm.pow.f64(double %Val, double %Power)
6013 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
6014 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
6015 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
6016</pre>
6017
6018<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6019<p>The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
6020 specified (positive or negative) power.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6021
6022<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6023<p>The second argument is a floating point power, and the first is a value to
6024 raise to that power.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6025
6026<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6027<p>This function returns the first value raised to the second power, returning
6028 the same values as the libm <tt>pow</tt> functions would, and handles error
6029 conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6030
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]6031</div>
6032
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6033<!-- ======================================================================= -->
6034<div class="doc_subsection">
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6035 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6036</div>
6037
6038<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6039
6040<p>LLVM provides intrinsics for a few important bit manipulation operations.
6041 These allow efficient code generation for some algorithms.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6042
6043</div>
6044
6045<!-- _______________________________________________________________________ -->
6046<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]6047 <a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6048</div>
6049
6050<div class="doc_text">
6051
6052<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]6053<p>This is an overloaded intrinsic function. You can use bswap on any integer
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6054 type that is an even number of bytes (i.e. BitWidth % 16 == 0).</p>
6055
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6056<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6057 declare i16 @llvm.bswap.i16(i16 &lt;id&gt;)
6058 declare i32 @llvm.bswap.i32(i32 &lt;id&gt;)
6059 declare i64 @llvm.bswap.i64(i64 &lt;id&gt;)
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6060</pre>
6061
6062<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6063<p>The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
6064 values with an even number of bytes (positive multiple of 16 bits). These
6065 are useful for performing operations on data that is not in the target's
6066 native byte order.</p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6067
6068<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6069<p>The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
6070 and low byte of the input i16 swapped. Similarly,
6071 the <tt>llvm.bswap.i32</tt> intrinsic returns an i32 value that has the four
6072 bytes of the input i32 swapped, so that if the input bytes are numbered 0, 1,
6073 2, 3 then the returned i32 will have its bytes in 3, 2, 1, 0 order.
6074 The <tt>llvm.bswap.i48</tt>, <tt>llvm.bswap.i64</tt> and other intrinsics
6075 extend this concept to additional even-byte lengths (6 bytes, 8 bytes and
6076 more, respectively).</p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]6077
6078</div>
6079
6080<!-- _______________________________________________________________________ -->
6081<div class="doc_subsubsection">
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]6082 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6083</div>
6084
6085<div class="doc_text">
6086
6087<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]6088<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]6089 width. Not all targets support all bit widths however.</p>
6090
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6091<pre>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6092 declare i8 @llvm.ctpop.i8(i8 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6093 declare i16 @llvm.ctpop.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]6094 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6095 declare i64 @llvm.ctpop.i64(i64 &lt;src&gt;)
6096 declare i256 @llvm.ctpop.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6097</pre>
6098
6099<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6100<p>The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set
6101 in a value.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6102
6103<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6104<p>The only argument is the value to be counted. The argument may be of any
6105 integer type. The return type must match the argument type.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6106
6107<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6108<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]6109
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6110</div>
6111
6112<!-- _______________________________________________________________________ -->
6113<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]6114 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6115</div>
6116
6117<div class="doc_text">
6118
6119<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6120<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
6121 integer bit width. Not all targets support all bit widths however.</p>
6122
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6123<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6124 declare i8 @llvm.ctlz.i8 (i8 &lt;src&gt;)
6125 declare i16 @llvm.ctlz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]6126 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6127 declare i64 @llvm.ctlz.i64(i64 &lt;src&gt;)
6128 declare i256 @llvm.ctlz.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6129</pre>
6130
6131<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6132<p>The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
6133 leading zeros in a variable.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6134
6135<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6136<p>The only argument is the value to be counted. The argument may be of any
6137 integer type. The return type must match the argument type.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6138
6139<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6140<p>The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant)
6141 zeros in a variable. If the src == 0 then the result is the size in bits of
6142 the type of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6143
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]6144</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]6145
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6146<!-- _______________________________________________________________________ -->
6147<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]6148 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6149</div>
6150
6151<div class="doc_text">
6152
6153<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6154<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
6155 integer bit width. Not all targets support all bit widths however.</p>
6156
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6157<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6158 declare i8 @llvm.cttz.i8 (i8 &lt;src&gt;)
6159 declare i16 @llvm.cttz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]6160 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]6161 declare i64 @llvm.cttz.i64(i64 &lt;src&gt;)
6162 declare i256 @llvm.cttz.i256(i256 &lt;src&gt;)
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6163</pre>
6164
6165<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6166<p>The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
6167 trailing zeros.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6168
6169<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6170<p>The only argument is the value to be counted. The argument may be of any
6171 integer type. The return type must match the argument type.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6172
6173<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6174<p>The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant)
6175 zeros in a variable. If the src == 0 then the result is the size in bits of
6176 the type of src. For example, <tt>llvm.cttz(2) = 1</tt>.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6177
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]6178</div>
6179
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6180<!-- ======================================================================= -->
6181<div class="doc_subsection">
6182 <a name="int_overflow">Arithmetic with Overflow Intrinsics</a>
6183</div>
6184
6185<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6186
6187<p>LLVM provides intrinsics for some arithmetic with overflow operations.</p>
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6188
6189</div>
6190
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6191<!-- _______________________________________________________________________ -->
6192<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6193 <a name="int_sadd_overflow">'<tt>llvm.sadd.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6194</div>
6195
6196<div class="doc_text">
6197
6198<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6199<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]6200 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6201
6202<pre>
6203 declare {i16, i1} @llvm.sadd.with.overflow.i16(i16 %a, i16 %b)
6204 declare {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
6205 declare {i64, i1} @llvm.sadd.with.overflow.i64(i64 %a, i64 %b)
6206</pre>
6207
6208<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6209<p>The '<tt>llvm.sadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6210 a signed addition of the two arguments, and indicate whether an overflow
6211 occurred during the signed summation.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6212
6213<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6214<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]6215 be of integer types of any bit width, but they must have the same bit
6216 width. The second element of the result structure must be of
6217 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6218 undergo signed addition.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6219
6220<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6221<p>The '<tt>llvm.sadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6222 a signed addition of the two variables. They return a structure &mdash; the
6223 first element of which is the signed summation, and the second element of
6224 which is a bit specifying if the signed summation resulted in an
6225 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6226
6227<h5>Examples:</h5>
6228<pre>
6229 %res = call {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
6230 %sum = extractvalue {i32, i1} %res, 0
6231 %obit = extractvalue {i32, i1} %res, 1
6232 br i1 %obit, label %overflow, label %normal
6233</pre>
6234
6235</div>
6236
6237<!-- _______________________________________________________________________ -->
6238<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6239 <a name="int_uadd_overflow">'<tt>llvm.uadd.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6240</div>
6241
6242<div class="doc_text">
6243
6244<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6245<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]6246 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6247
6248<pre>
6249 declare {i16, i1} @llvm.uadd.with.overflow.i16(i16 %a, i16 %b)
6250 declare {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
6251 declare {i64, i1} @llvm.uadd.with.overflow.i64(i64 %a, i64 %b)
6252</pre>
6253
6254<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6255<p>The '<tt>llvm.uadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6256 an unsigned addition of the two arguments, and indicate whether a carry
6257 occurred during the unsigned summation.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6258
6259<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6260<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]6261 be of integer types of any bit width, but they must have the same bit
6262 width. The second element of the result structure must be of
6263 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6264 undergo unsigned addition.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6265
6266<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6267<p>The '<tt>llvm.uadd.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6268 an unsigned addition of the two arguments. They return a structure &mdash;
6269 the first element of which is the sum, and the second element of which is a
6270 bit specifying if the unsigned summation resulted in a carry.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6271
6272<h5>Examples:</h5>
6273<pre>
6274 %res = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
6275 %sum = extractvalue {i32, i1} %res, 0
6276 %obit = extractvalue {i32, i1} %res, 1
6277 br i1 %obit, label %carry, label %normal
6278</pre>
6279
6280</div>
6281
6282<!-- _______________________________________________________________________ -->
6283<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6284 <a name="int_ssub_overflow">'<tt>llvm.ssub.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6285</div>
6286
6287<div class="doc_text">
6288
6289<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6290<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]6291 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6292
6293<pre>
6294 declare {i16, i1} @llvm.ssub.with.overflow.i16(i16 %a, i16 %b)
6295 declare {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
6296 declare {i64, i1} @llvm.ssub.with.overflow.i64(i64 %a, i64 %b)
6297</pre>
6298
6299<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6300<p>The '<tt>llvm.ssub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6301 a signed subtraction of the two arguments, and indicate whether an overflow
6302 occurred during the signed subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6303
6304<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6305<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]6306 be of integer types of any bit width, but they must have the same bit
6307 width. The second element of the result structure must be of
6308 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6309 undergo signed subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6310
6311<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6312<p>The '<tt>llvm.ssub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6313 a signed subtraction of the two arguments. They return a structure &mdash;
6314 the first element of which is the subtraction, and the second element of
6315 which is a bit specifying if the signed subtraction resulted in an
6316 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6317
6318<h5>Examples:</h5>
6319<pre>
6320 %res = call {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
6321 %sum = extractvalue {i32, i1} %res, 0
6322 %obit = extractvalue {i32, i1} %res, 1
6323 br i1 %obit, label %overflow, label %normal
6324</pre>
6325
6326</div>
6327
6328<!-- _______________________________________________________________________ -->
6329<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6330 <a name="int_usub_overflow">'<tt>llvm.usub.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6331</div>
6332
6333<div class="doc_text">
6334
6335<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6336<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]6337 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6338
6339<pre>
6340 declare {i16, i1} @llvm.usub.with.overflow.i16(i16 %a, i16 %b)
6341 declare {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
6342 declare {i64, i1} @llvm.usub.with.overflow.i64(i64 %a, i64 %b)
6343</pre>
6344
6345<h5>Overview:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6346<p>The '<tt>llvm.usub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6347 an unsigned subtraction of the two arguments, and indicate whether an
6348 overflow occurred during the unsigned subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6349
6350<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6351<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]6352 be of integer types of any bit width, but they must have the same bit
6353 width. The second element of the result structure must be of
6354 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6355 undergo unsigned subtraction.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6356
6357<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6358<p>The '<tt>llvm.usub.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6359 an unsigned subtraction of the two arguments. They return a structure &mdash;
6360 the first element of which is the subtraction, and the second element of
6361 which is a bit specifying if the unsigned subtraction resulted in an
6362 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6363
6364<h5>Examples:</h5>
6365<pre>
6366 %res = call {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
6367 %sum = extractvalue {i32, i1} %res, 0
6368 %obit = extractvalue {i32, i1} %res, 1
6369 br i1 %obit, label %overflow, label %normal
6370</pre>
6371
6372</div>
6373
6374<!-- _______________________________________________________________________ -->
6375<div class="doc_subsubsection">
Bill Wendlingda01af72009-02-08 04:04:40 +0000[diff] [blame]6376 <a name="int_smul_overflow">'<tt>llvm.smul.with.overflow.*</tt>' Intrinsics</a>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6377</div>
6378
6379<div class="doc_text">
6380
6381<h5>Syntax:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6382<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]6383 on any integer bit width.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6384
6385<pre>
6386 declare {i16, i1} @llvm.smul.with.overflow.i16(i16 %a, i16 %b)
6387 declare {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
6388 declare {i64, i1} @llvm.smul.with.overflow.i64(i64 %a, i64 %b)
6389</pre>
6390
6391<h5>Overview:</h5>
6392
6393<p>The '<tt>llvm.smul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6394 a signed multiplication of the two arguments, and indicate whether an
6395 overflow occurred during the signed multiplication.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6396
6397<h5>Arguments:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6398<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]6399 be of integer types of any bit width, but they must have the same bit
6400 width. The second element of the result structure must be of
6401 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6402 undergo signed multiplication.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6403
6404<h5>Semantics:</h5>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6405<p>The '<tt>llvm.smul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6406 a signed multiplication of the two arguments. They return a structure &mdash;
6407 the first element of which is the multiplication, and the second element of
6408 which is a bit specifying if the signed multiplication resulted in an
6409 overflow.</p>
Bill Wendlingac1df8e2009-02-08 01:40:31 +0000[diff] [blame]6410
6411<h5>Examples:</h5>
6412<pre>
6413 %res = call {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
6414 %sum = extractvalue {i32, i1} %res, 0
6415 %obit = extractvalue {i32, i1} %res, 1
6416 br i1 %obit, label %overflow, label %normal
6417</pre>
6418
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]6419</div>
6420
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6421<!-- _______________________________________________________________________ -->
6422<div class="doc_subsubsection">
6423 <a name="int_umul_overflow">'<tt>llvm.umul.with.overflow.*</tt>' Intrinsics</a>
6424</div>
6425
6426<div class="doc_text">
6427
6428<h5>Syntax:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6429<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]6430 on any integer bit width.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6431
6432<pre>
6433 declare {i16, i1} @llvm.umul.with.overflow.i16(i16 %a, i16 %b)
6434 declare {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
6435 declare {i64, i1} @llvm.umul.with.overflow.i64(i64 %a, i64 %b)
6436</pre>
6437
6438<h5>Overview:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6439<p>The '<tt>llvm.umul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6440 a unsigned multiplication of the two arguments, and indicate whether an
6441 overflow occurred during the unsigned multiplication.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6442
6443<h5>Arguments:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6444<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]6445 be of integer types of any bit width, but they must have the same bit
6446 width. The second element of the result structure must be of
6447 type <tt>i1</tt>. <tt>%a</tt> and <tt>%b</tt> are the two values that will
6448 undergo unsigned multiplication.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6449
6450<h5>Semantics:</h5>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6451<p>The '<tt>llvm.umul.with.overflow</tt>' family of intrinsic functions perform
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6452 an unsigned multiplication of the two arguments. They return a structure
6453 &mdash; the first element of which is the multiplication, and the second
6454 element of which is a bit specifying if the unsigned multiplication resulted
6455 in an overflow.</p>
Bill Wendling41b485c2009-02-08 23:00:09 +0000[diff] [blame]6456
6457<h5>Examples:</h5>
6458<pre>
6459 %res = call {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
6460 %sum = extractvalue {i32, i1} %res, 0
6461 %obit = extractvalue {i32, i1} %res, 1
6462 br i1 %obit, label %overflow, label %normal
6463</pre>
6464
6465</div>
6466
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6467<!-- ======================================================================= -->
6468<div class="doc_subsection">
6469 <a name="int_debugger">Debugger Intrinsics</a>
6470</div>
6471
6472<div class="doc_text">
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6473
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6474<p>The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt>
6475 prefix), are described in
6476 the <a href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source
6477 Level Debugging</a> document.</p>
6478
6479</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]6480
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]6481<!-- ======================================================================= -->
6482<div class="doc_subsection">
6483 <a name="int_eh">Exception Handling Intrinsics</a>
6484</div>
6485
6486<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6487
6488<p>The LLVM exception handling intrinsics (which all start with
6489 <tt>llvm.eh.</tt> prefix), are described in
6490 the <a href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
6491 Handling</a> document.</p>
6492
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]6493</div>
6494
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]6495<!-- ======================================================================= -->
6496<div class="doc_subsection">
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]6497 <a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6498</div>
6499
6500<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6501
6502<p>This intrinsic makes it possible to excise one parameter, marked with
6503 the <tt>nest</tt> attribute, from a function. The result is a callable
6504 function pointer lacking the nest parameter - the caller does not need to
6505 provide a value for it. Instead, the value to use is stored in advance in a
6506 "trampoline", a block of memory usually allocated on the stack, which also
6507 contains code to splice the nest value into the argument list. This is used
6508 to implement the GCC nested function address extension.</p>
6509
6510<p>For example, if the function is
6511 <tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
6512 pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as
6513 follows:</p>
6514
6515<div class="doc_code">
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6516<pre>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]6517 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
6518 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
6519 %p = call i8* @llvm.init.trampoline( i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8*), i8* %nval )
6520 %fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6521</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6522</div>
6523
6524<p>The call <tt>%val = call i32 %fp( i32 %x, i32 %y )</tt> is then equivalent
6525 to <tt>%val = call i32 %f( i8* %nval, i32 %x, i32 %y )</tt>.</p>
6526
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6527</div>
6528
6529<!-- _______________________________________________________________________ -->
6530<div class="doc_subsubsection">
6531 <a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
6532</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6533
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6534<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6535
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6536<h5>Syntax:</h5>
6537<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6538 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]6539</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6540
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6541<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6542<p>This fills the memory pointed to by <tt>tramp</tt> with code and returns a
6543 function pointer suitable for executing it.</p>
6544
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6545<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6546<p>The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
6547 pointers. The <tt>tramp</tt> argument must point to a sufficiently large and
6548 sufficiently aligned block of memory; this memory is written to by the
6549 intrinsic. Note that the size and the alignment are target-specific - LLVM
6550 currently provides no portable way of determining them, so a front-end that
6551 generates this intrinsic needs to have some target-specific knowledge.
6552 The <tt>func</tt> argument must hold a function bitcast to
6553 an <tt>i8*</tt>.</p>
6554
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6555<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6556<p>The block of memory pointed to by <tt>tramp</tt> is filled with target
6557 dependent code, turning it into a function. A pointer to this function is
6558 returned, but needs to be bitcast to an <a href="#int_trampoline">appropriate
6559 function pointer type</a> before being called. The new function's signature
6560 is the same as that of <tt>func</tt> with any arguments marked with
6561 the <tt>nest</tt> attribute removed. At most one such <tt>nest</tt> argument
6562 is allowed, and it must be of pointer type. Calling the new function is
6563 equivalent to calling <tt>func</tt> with the same argument list, but
6564 with <tt>nval</tt> used for the missing <tt>nest</tt> argument. If, after
6565 calling <tt>llvm.init.trampoline</tt>, the memory pointed to
6566 by <tt>tramp</tt> is modified, then the effect of any later call to the
6567 returned function pointer is undefined.</p>
6568
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]6569</div>
6570
6571<!-- ======================================================================= -->
6572<div class="doc_subsection">
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6573 <a name="int_atomics">Atomic Operations and Synchronization Intrinsics</a>
6574</div>
6575
6576<div class="doc_text">
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6577
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6578<p>These intrinsic functions expand the "universal IR" of LLVM to represent
6579 hardware constructs for atomic operations and memory synchronization. This
6580 provides an interface to the hardware, not an interface to the programmer. It
6581 is aimed at a low enough level to allow any programming models or APIs
6582 (Application Programming Interfaces) which need atomic behaviors to map
6583 cleanly onto it. It is also modeled primarily on hardware behavior. Just as
6584 hardware provides a "universal IR" for source languages, it also provides a
6585 starting point for developing a "universal" atomic operation and
6586 synchronization IR.</p>
6587
6588<p>These do <em>not</em> form an API such as high-level threading libraries,
6589 software transaction memory systems, atomic primitives, and intrinsic
6590 functions as found in BSD, GNU libc, atomic_ops, APR, and other system and
6591 application libraries. The hardware interface provided by LLVM should allow
6592 a clean implementation of all of these APIs and parallel programming models.
6593 No one model or paradigm should be selected above others unless the hardware
6594 itself ubiquitously does so.</p>
6595
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6596</div>
6597
6598<!-- _______________________________________________________________________ -->
6599<div class="doc_subsubsection">
6600 <a name="int_memory_barrier">'<tt>llvm.memory.barrier</tt>' Intrinsic</a>
6601</div>
6602<div class="doc_text">
6603<h5>Syntax:</h5>
6604<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6605 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]6606</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6607
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6608<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6609<p>The <tt>llvm.memory.barrier</tt> intrinsic guarantees ordering between
6610 specific pairs of memory access types.</p>
6611
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6612<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6613<p>The <tt>llvm.memory.barrier</tt> intrinsic requires five boolean arguments.
6614 The first four arguments enables a specific barrier as listed below. The
6615 fith argument specifies that the barrier applies to io or device or uncached
6616 memory.</p>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6617
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6618<ul>
6619 <li><tt>ll</tt>: load-load barrier</li>
6620 <li><tt>ls</tt>: load-store barrier</li>
6621 <li><tt>sl</tt>: store-load barrier</li>
6622 <li><tt>ss</tt>: store-store barrier</li>
6623 <li><tt>device</tt>: barrier applies to device and uncached memory also.</li>
6624</ul>
6625
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6626<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6627<p>This intrinsic causes the system to enforce some ordering constraints upon
6628 the loads and stores of the program. This barrier does not
6629 indicate <em>when</em> any events will occur, it only enforces
6630 an <em>order</em> in which they occur. For any of the specified pairs of load
6631 and store operations (f.ex. load-load, or store-load), all of the first
6632 operations preceding the barrier will complete before any of the second
6633 operations succeeding the barrier begin. Specifically the semantics for each
6634 pairing is as follows:</p>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6635
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6636<ul>
6637 <li><tt>ll</tt>: All loads before the barrier must complete before any load
6638 after the barrier begins.</li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]6639 <li><tt>ls</tt>: All loads before the barrier must complete before any
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6640 store after the barrier begins.</li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]6641 <li><tt>ss</tt>: All stores before the barrier must complete before any
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6642 store after the barrier begins.</li>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]6643 <li><tt>sl</tt>: All stores before the barrier must complete before any
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6644 load after the barrier begins.</li>
6645</ul>
6646
6647<p>These semantics are applied with a logical "and" behavior when more than one
6648 is enabled in a single memory barrier intrinsic.</p>
6649
6650<p>Backends may implement stronger barriers than those requested when they do
6651 not support as fine grained a barrier as requested. Some architectures do
6652 not need all types of barriers and on such architectures, these become
6653 noops.</p>
6654
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6655<h5>Example:</h5>
6656<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]6657%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
6658%ptr = bitcast i8* %mallocP to i32*
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6659 store i32 4, %ptr
6660
6661%result1 = load i32* %ptr <i>; yields {i32}:result1 = 4</i>
6662 call void @llvm.memory.barrier( i1 false, i1 true, i1 false, i1 false )
6663 <i>; guarantee the above finishes</i>
6664 store i32 8, %ptr <i>; before this begins</i>
6665</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6666
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6667</div>
6668
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6669<!-- _______________________________________________________________________ -->
6670<div class="doc_subsubsection">
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6671 <a name="int_atomic_cmp_swap">'<tt>llvm.atomic.cmp.swap.*</tt>' Intrinsic</a>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6672</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6673
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6674<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6675
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6676<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6677<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.cmp.swap</tt> on
6678 any integer bit width and for different address spaces. Not all targets
6679 support all bit widths however.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6680
6681<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6682 declare i8 @llvm.atomic.cmp.swap.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;cmp&gt;, i8 &lt;val&gt; )
6683 declare i16 @llvm.atomic.cmp.swap.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;cmp&gt;, i16 &lt;val&gt; )
6684 declare i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;cmp&gt;, i32 &lt;val&gt; )
6685 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]6686</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6687
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6688<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6689<p>This loads a value in memory and compares it to a given value. If they are
6690 equal, it stores a new value into the memory.</p>
6691
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6692<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6693<p>The <tt>llvm.atomic.cmp.swap</tt> intrinsic takes three arguments. The result
6694 as well as both <tt>cmp</tt> and <tt>val</tt> must be integer values with the
6695 same bit width. The <tt>ptr</tt> argument must be a pointer to a value of
6696 this integer type. While any bit width integer may be used, targets may only
6697 lower representations they support in hardware.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6698
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6699<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6700<p>This entire intrinsic must be executed atomically. It first loads the value
6701 in memory pointed to by <tt>ptr</tt> and compares it with the
6702 value <tt>cmp</tt>. If they are equal, <tt>val</tt> is stored into the
6703 memory. The loaded value is yielded in all cases. This provides the
6704 equivalent of an atomic compare-and-swap operation within the SSA
6705 framework.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6706
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6707<h5>Examples:</h5>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6708<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]6709%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
6710%ptr = bitcast i8* %mallocP to i32*
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6711 store i32 4, %ptr
6712
6713%val1 = add i32 4, 4
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6714%result1 = call i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* %ptr, i32 4, %val1 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6715 <i>; yields {i32}:result1 = 4</i>
6716%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
6717%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
6718
6719%val2 = add i32 1, 1
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6720%result2 = call i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* %ptr, i32 5, %val2 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6721 <i>; yields {i32}:result2 = 8</i>
6722%stored2 = icmp eq i32 %result2, 5 <i>; yields {i1}:stored2 = false</i>
6723
6724%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 8</i>
6725</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6726
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6727</div>
6728
6729<!-- _______________________________________________________________________ -->
6730<div class="doc_subsubsection">
6731 <a name="int_atomic_swap">'<tt>llvm.atomic.swap.*</tt>' Intrinsic</a>
6732</div>
6733<div class="doc_text">
6734<h5>Syntax:</h5>
6735
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6736<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.swap</tt> on any
6737 integer bit width. Not all targets support all bit widths however.</p>
6738
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6739<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6740 declare i8 @llvm.atomic.swap.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;val&gt; )
6741 declare i16 @llvm.atomic.swap.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;val&gt; )
6742 declare i32 @llvm.atomic.swap.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;val&gt; )
6743 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]6744</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6745
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6746<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6747<p>This intrinsic loads the value stored in memory at <tt>ptr</tt> and yields
6748 the value from memory. It then stores the value in <tt>val</tt> in the memory
6749 at <tt>ptr</tt>.</p>
6750
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6751<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6752<p>The <tt>llvm.atomic.swap</tt> intrinsic takes two arguments. Both
6753 the <tt>val</tt> argument and the result must be integers of the same bit
6754 width. The first argument, <tt>ptr</tt>, must be a pointer to a value of this
6755 integer type. The targets may only lower integer representations they
6756 support.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6757
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6758<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6759<p>This intrinsic loads the value pointed to by <tt>ptr</tt>, yields it, and
6760 stores <tt>val</tt> back into <tt>ptr</tt> atomically. This provides the
6761 equivalent of an atomic swap operation within the SSA framework.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6762
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6763<h5>Examples:</h5>
6764<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]6765%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
6766%ptr = bitcast i8* %mallocP to i32*
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6767 store i32 4, %ptr
6768
6769%val1 = add i32 4, 4
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6770%result1 = call i32 @llvm.atomic.swap.i32.p0i32( i32* %ptr, i32 %val1 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6771 <i>; yields {i32}:result1 = 4</i>
6772%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
6773%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
6774
6775%val2 = add i32 1, 1
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6776%result2 = call i32 @llvm.atomic.swap.i32.p0i32( i32* %ptr, i32 %val2 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6777 <i>; yields {i32}:result2 = 8</i>
6778
6779%stored2 = icmp eq i32 %result2, 8 <i>; yields {i1}:stored2 = true</i>
6780%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 2</i>
6781</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6782
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6783</div>
6784
6785<!-- _______________________________________________________________________ -->
6786<div class="doc_subsubsection">
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6787 <a name="int_atomic_load_add">'<tt>llvm.atomic.load.add.*</tt>' Intrinsic</a>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6788
6789</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6790
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6791<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6792
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6793<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6794<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.add</tt> on
6795 any integer bit width. Not all targets support all bit widths however.</p>
6796
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6797<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6798 declare i8 @llvm.atomic.load.add.i8..p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6799 declare i16 @llvm.atomic.load.add.i16..p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6800 declare i32 @llvm.atomic.load.add.i32..p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6801 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]6802</pre>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6803
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6804<h5>Overview:</h5>
6805<p>This intrinsic adds <tt>delta</tt> to the value stored in memory
6806 at <tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.</p>
6807
6808<h5>Arguments:</h5>
6809<p>The intrinsic takes two arguments, the first a pointer to an integer value
6810 and the second an integer value. The result is also an integer value. These
6811 integer types can have any bit width, but they must all have the same bit
6812 width. The targets may only lower integer representations they support.</p>
6813
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6814<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6815<p>This intrinsic does a series of operations atomically. It first loads the
6816 value stored at <tt>ptr</tt>. It then adds <tt>delta</tt>, stores the result
6817 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]6818
6819<h5>Examples:</h5>
6820<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]6821%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
6822%ptr = bitcast i8* %mallocP to i32*
6823 store i32 4, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6824%result1 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 4 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6825 <i>; yields {i32}:result1 = 4</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6826%result2 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 2 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6827 <i>; yields {i32}:result2 = 8</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6828%result3 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 5 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6829 <i>; yields {i32}:result3 = 10</i>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6830%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 15</i>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6831</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6832
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6833</div>
6834
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6835<!-- _______________________________________________________________________ -->
6836<div class="doc_subsubsection">
6837 <a name="int_atomic_load_sub">'<tt>llvm.atomic.load.sub.*</tt>' Intrinsic</a>
6838
6839</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6840
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6841<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6842
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6843<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6844<p>This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.sub</tt> on
6845 any integer bit width and for different address spaces. Not all targets
6846 support all bit widths however.</p>
6847
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6848<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6849 declare i8 @llvm.atomic.load.sub.i8.p0i32( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6850 declare i16 @llvm.atomic.load.sub.i16.p0i32( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6851 declare i32 @llvm.atomic.load.sub.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6852 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]6853</pre>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6854
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6855<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]6856<p>This intrinsic subtracts <tt>delta</tt> to the value stored in memory at
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6857 <tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.</p>
6858
6859<h5>Arguments:</h5>
6860<p>The intrinsic takes two arguments, the first a pointer to an integer value
6861 and the second an integer value. The result is also an integer value. These
6862 integer types can have any bit width, but they must all have the same bit
6863 width. The targets may only lower integer representations they support.</p>
6864
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6865<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6866<p>This intrinsic does a series of operations atomically. It first loads the
6867 value stored at <tt>ptr</tt>. It then subtracts <tt>delta</tt>, stores the
6868 result to <tt>ptr</tt>. It yields the original value stored
6869 at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6870
6871<h5>Examples:</h5>
6872<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]6873%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
6874%ptr = bitcast i8* %mallocP to i32*
6875 store i32 8, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6876%result1 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 4 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6877 <i>; yields {i32}:result1 = 8</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6878%result2 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 2 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6879 <i>; yields {i32}:result2 = 4</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6880%result3 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 5 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6881 <i>; yields {i32}:result3 = 2</i>
6882%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = -3</i>
6883</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6884
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6885</div>
6886
6887<!-- _______________________________________________________________________ -->
6888<div class="doc_subsubsection">
6889 <a name="int_atomic_load_and">'<tt>llvm.atomic.load.and.*</tt>' Intrinsic</a><br>
6890 <a name="int_atomic_load_nand">'<tt>llvm.atomic.load.nand.*</tt>' Intrinsic</a><br>
6891 <a name="int_atomic_load_or">'<tt>llvm.atomic.load.or.*</tt>' Intrinsic</a><br>
6892 <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]6893</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6894
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6895<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6896
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6897<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6898<p>These are overloaded intrinsics. You can
6899 use <tt>llvm.atomic.load_and</tt>, <tt>llvm.atomic.load_nand</tt>,
6900 <tt>llvm.atomic.load_or</tt>, and <tt>llvm.atomic.load_xor</tt> on any integer
6901 bit width and for different address spaces. Not all targets support all bit
6902 widths however.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6903
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6904<pre>
6905 declare i8 @llvm.atomic.load.and.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6906 declare i16 @llvm.atomic.load.and.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6907 declare i32 @llvm.atomic.load.and.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6908 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]6909</pre>
6910
6911<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6912 declare i8 @llvm.atomic.load.or.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6913 declare i16 @llvm.atomic.load.or.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6914 declare i32 @llvm.atomic.load.or.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6915 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]6916</pre>
6917
6918<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6919 declare i8 @llvm.atomic.load.nand.i8.p0i32( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6920 declare i16 @llvm.atomic.load.nand.i16.p0i32( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6921 declare i32 @llvm.atomic.load.nand.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6922 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]6923</pre>
6924
6925<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6926 declare i8 @llvm.atomic.load.xor.i8.p0i32( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6927 declare i16 @llvm.atomic.load.xor.i16.p0i32( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6928 declare i32 @llvm.atomic.load.xor.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6929 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]6930</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6931
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6932<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6933<p>These intrinsics bitwise the operation (and, nand, or, xor) <tt>delta</tt> to
6934 the value stored in memory at <tt>ptr</tt>. It yields the original value
6935 at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6936
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6937<h5>Arguments:</h5>
6938<p>These intrinsics take two arguments, the first a pointer to an integer value
6939 and the second an integer value. The result is also an integer value. These
6940 integer types can have any bit width, but they must all have the same bit
6941 width. The targets may only lower integer representations they support.</p>
6942
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6943<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6944<p>These intrinsics does a series of operations atomically. They first load the
6945 value stored at <tt>ptr</tt>. They then do the bitwise
6946 operation <tt>delta</tt>, store the result to <tt>ptr</tt>. They yield the
6947 original value stored at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6948
6949<h5>Examples:</h5>
6950<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]6951%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
6952%ptr = bitcast i8* %mallocP to i32*
6953 store i32 0x0F0F, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6954%result0 = call i32 @llvm.atomic.load.nand.i32.p0i32( i32* %ptr, i32 0xFF )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6955 <i>; yields {i32}:result0 = 0x0F0F</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6956%result1 = call i32 @llvm.atomic.load.and.i32.p0i32( i32* %ptr, i32 0xFF )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6957 <i>; yields {i32}:result1 = 0xFFFFFFF0</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6958%result2 = call i32 @llvm.atomic.load.or.i32.p0i32( i32* %ptr, i32 0F )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6959 <i>; yields {i32}:result2 = 0xF0</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6960%result3 = call i32 @llvm.atomic.load.xor.i32.p0i32( i32* %ptr, i32 0F )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6961 <i>; yields {i32}:result3 = FF</i>
6962%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = F0</i>
6963</pre>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6964
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6965</div>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6966
6967<!-- _______________________________________________________________________ -->
6968<div class="doc_subsubsection">
6969 <a name="int_atomic_load_max">'<tt>llvm.atomic.load.max.*</tt>' Intrinsic</a><br>
6970 <a name="int_atomic_load_min">'<tt>llvm.atomic.load.min.*</tt>' Intrinsic</a><br>
6971 <a name="int_atomic_load_umax">'<tt>llvm.atomic.load.umax.*</tt>' Intrinsic</a><br>
6972 <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]6973</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6974
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6975<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6976
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6977<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6978<p>These are overloaded intrinsics. You can use <tt>llvm.atomic.load_max</tt>,
6979 <tt>llvm.atomic.load_min</tt>, <tt>llvm.atomic.load_umax</tt>, and
6980 <tt>llvm.atomic.load_umin</tt> on any integer bit width and for different
6981 address spaces. Not all targets support all bit widths however.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6982
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6983<pre>
6984 declare i8 @llvm.atomic.load.max.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6985 declare i16 @llvm.atomic.load.max.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6986 declare i32 @llvm.atomic.load.max.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6987 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]6988</pre>
6989
6990<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6991 declare i8 @llvm.atomic.load.min.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6992 declare i16 @llvm.atomic.load.min.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6993 declare i32 @llvm.atomic.load.min.i32..p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6994 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]6995</pre>
6996
6997<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]6998 declare i8 @llvm.atomic.load.umax.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6999 declare i16 @llvm.atomic.load.umax.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
7000 declare i32 @llvm.atomic.load.umax.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
7001 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]7002</pre>
7003
7004<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7005 declare i8 @llvm.atomic.load.umin.i8..p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
7006 declare i16 @llvm.atomic.load.umin.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
7007 declare i32 @llvm.atomic.load.umin.i32..p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
7008 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]7009</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7010
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7011<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7012<p>These intrinsics takes the signed or unsigned minimum or maximum of
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7013 <tt>delta</tt> and the value stored in memory at <tt>ptr</tt>. It yields the
7014 original value at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7015
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7016<h5>Arguments:</h5>
7017<p>These intrinsics take two arguments, the first a pointer to an integer value
7018 and the second an integer value. The result is also an integer value. These
7019 integer types can have any bit width, but they must all have the same bit
7020 width. The targets may only lower integer representations they support.</p>
7021
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7022<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7023<p>These intrinsics does a series of operations atomically. They first load the
7024 value stored at <tt>ptr</tt>. They then do the signed or unsigned min or
7025 max <tt>delta</tt> and the value, store the result to <tt>ptr</tt>. They
7026 yield the original value stored at <tt>ptr</tt>.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7027
7028<h5>Examples:</h5>
7029<pre>
Victor Hernandez2fee2942009-10-26 23:44:29 +0000[diff] [blame]7030%mallocP = tail call i8* @malloc(i32 ptrtoint (i32* getelementptr (i32* null, i32 1) to i32))
7031%ptr = bitcast i8* %mallocP to i32*
7032 store i32 7, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]7033%result0 = call i32 @llvm.atomic.load.min.i32.p0i32( i32* %ptr, i32 -2 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7034 <i>; yields {i32}:result0 = 7</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]7035%result1 = call i32 @llvm.atomic.load.max.i32.p0i32( i32* %ptr, i32 8 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7036 <i>; yields {i32}:result1 = -2</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]7037%result2 = call i32 @llvm.atomic.load.umin.i32.p0i32( i32* %ptr, i32 10 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7038 <i>; yields {i32}:result2 = 8</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]7039%result3 = call i32 @llvm.atomic.load.umax.i32.p0i32( i32* %ptr, i32 30 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7040 <i>; yields {i32}:result3 = 8</i>
7041%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 30</i>
7042</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7043
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]7044</div>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7045
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7046
7047<!-- ======================================================================= -->
7048<div class="doc_subsection">
7049 <a name="int_memorymarkers">Memory Use Markers</a>
7050</div>
7051
7052<div class="doc_text">
7053
7054<p>This class of intrinsics exists to information about the lifetime of memory
7055 objects and ranges where variables are immutable.</p>
7056
7057</div>
7058
7059<!-- _______________________________________________________________________ -->
7060<div class="doc_subsubsection">
7061 <a name="int_lifetime_start">'<tt>llvm.lifetime.start</tt>' Intrinsic</a>
7062</div>
7063
7064<div class="doc_text">
7065
7066<h5>Syntax:</h5>
7067<pre>
7068 declare void @llvm.lifetime.start(i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
7069</pre>
7070
7071<h5>Overview:</h5>
7072<p>The '<tt>llvm.lifetime.start</tt>' intrinsic specifies the start of a memory
7073 object's lifetime.</p>
7074
7075<h5>Arguments:</h5>
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7076<p>The first argument is a constant integer representing the size of the
7077 object, or -1 if it is variable sized. The second argument is a pointer to
7078 the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7079
7080<h5>Semantics:</h5>
7081<p>This intrinsic indicates that before this point in the code, the value of the
7082 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]7083 never be used and has an undefined value. A load from the pointer that
7084 precedes this intrinsic can be replaced with
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7085 <tt>'<a href="#undefvalues">undef</a>'</tt>.</p>
7086
7087</div>
7088
7089<!-- _______________________________________________________________________ -->
7090<div class="doc_subsubsection">
7091 <a name="int_lifetime_end">'<tt>llvm.lifetime.end</tt>' Intrinsic</a>
7092</div>
7093
7094<div class="doc_text">
7095
7096<h5>Syntax:</h5>
7097<pre>
7098 declare void @llvm.lifetime.end(i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
7099</pre>
7100
7101<h5>Overview:</h5>
7102<p>The '<tt>llvm.lifetime.end</tt>' intrinsic specifies the end of a memory
7103 object's lifetime.</p>
7104
7105<h5>Arguments:</h5>
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7106<p>The first argument is a constant integer representing the size of the
7107 object, or -1 if it is variable sized. The second argument is a pointer to
7108 the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7109
7110<h5>Semantics:</h5>
7111<p>This intrinsic indicates that after this point in the code, the value of the
7112 memory pointed to by <tt>ptr</tt> is dead. This means that it is known to
7113 never be used and has an undefined value. Any stores into the memory object
7114 following this intrinsic may be removed as dead.
7115
7116</div>
7117
7118<!-- _______________________________________________________________________ -->
7119<div class="doc_subsubsection">
7120 <a name="int_invariant_start">'<tt>llvm.invariant.start</tt>' Intrinsic</a>
7121</div>
7122
7123<div class="doc_text">
7124
7125<h5>Syntax:</h5>
7126<pre>
7127 declare {}* @llvm.invariant.start(i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;) readonly
7128</pre>
7129
7130<h5>Overview:</h5>
7131<p>The '<tt>llvm.invariant.start</tt>' intrinsic specifies that the contents of
7132 a memory object will not change.</p>
7133
7134<h5>Arguments:</h5>
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7135<p>The first argument is a constant integer representing the size of the
7136 object, or -1 if it is variable sized. The second argument is a pointer to
7137 the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7138
7139<h5>Semantics:</h5>
7140<p>This intrinsic indicates that until an <tt>llvm.invariant.end</tt> that uses
7141 the return value, the referenced memory location is constant and
7142 unchanging.</p>
7143
7144</div>
7145
7146<!-- _______________________________________________________________________ -->
7147<div class="doc_subsubsection">
7148 <a name="int_invariant_end">'<tt>llvm.invariant.end</tt>' Intrinsic</a>
7149</div>
7150
7151<div class="doc_text">
7152
7153<h5>Syntax:</h5>
7154<pre>
7155 declare void @llvm.invariant.end({}* &lt;start&gt;, i64 &lt;size&gt;, i8* nocapture &lt;ptr&gt;)
7156</pre>
7157
7158<h5>Overview:</h5>
7159<p>The '<tt>llvm.invariant.end</tt>' intrinsic specifies that the contents of
7160 a memory object are mutable.</p>
7161
7162<h5>Arguments:</h5>
7163<p>The first argument is the matching <tt>llvm.invariant.start</tt> intrinsic.
Nick Lewycky321333e2009-10-13 07:57:33 +0000[diff] [blame]7164 The second argument is a constant integer representing the size of the
7165 object, or -1 if it is variable sized and the third argument is a pointer
7166 to the object.</p>
Nick Lewyckycc271862009-10-13 07:03:23 +0000[diff] [blame]7167
7168<h5>Semantics:</h5>
7169<p>This intrinsic indicates that the memory is mutable again.</p>
7170
7171</div>
7172
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]7173<!-- ======================================================================= -->
7174<div class="doc_subsection">
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7175 <a name="int_general">General Intrinsics</a>
7176</div>
7177
7178<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7179
7180<p>This class of intrinsics is designed to be generic and has no specific
7181 purpose.</p>
7182
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7183</div>
7184
7185<!-- _______________________________________________________________________ -->
7186<div class="doc_subsubsection">
7187 <a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
7188</div>
7189
7190<div class="doc_text">
7191
7192<h5>Syntax:</h5>
7193<pre>
Tanya Lattnerd2e84422007-06-18 23:42:37 +0000[diff] [blame]7194 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]7195</pre>
7196
7197<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7198<p>The '<tt>llvm.var.annotation</tt>' intrinsic.</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7199
7200<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7201<p>The first argument is a pointer to a value, the second is a pointer to a
7202 global string, the third is a pointer to a global string which is the source
7203 file name, and the last argument is the line number.</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7204
7205<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7206<p>This intrinsic allows annotation of local variables with arbitrary strings.
7207 This can be useful for special purpose optimizations that want to look for
7208 these annotations. These have no other defined use, they are ignored by code
7209 generation and optimization.</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7210
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]7211</div>
7212
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7213<!-- _______________________________________________________________________ -->
7214<div class="doc_subsubsection">
Tanya Lattnere1a8da02007-09-21 23:57:59 +0000[diff] [blame]7215 <a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7216</div>
7217
7218<div class="doc_text">
7219
7220<h5>Syntax:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7221<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
7222 any integer bit width.</p>
7223
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7224<pre>
Tanya Lattnerd3989a82007-09-22 00:03:01 +0000[diff] [blame]7225 declare i8 @llvm.annotation.i8(i8 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
7226 declare i16 @llvm.annotation.i16(i16 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
7227 declare i32 @llvm.annotation.i32(i32 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
7228 declare i64 @llvm.annotation.i64(i64 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
7229 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]7230</pre>
7231
7232<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7233<p>The '<tt>llvm.annotation</tt>' intrinsic.</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7234
7235<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7236<p>The first argument is an integer value (result of some expression), the
7237 second is a pointer to a global string, the third is a pointer to a global
7238 string which is the source file name, and the last argument is the line
7239 number. It returns the value of the first argument.</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7240
7241<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7242<p>This intrinsic allows annotations to be put on arbitrary expressions with
7243 arbitrary strings. This can be useful for special purpose optimizations that
7244 want to look for these annotations. These have no other defined use, they
7245 are ignored by code generation and optimization.</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7246
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]7247</div>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]7248
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7249<!-- _______________________________________________________________________ -->
7250<div class="doc_subsubsection">
7251 <a name="int_trap">'<tt>llvm.trap</tt>' Intrinsic</a>
7252</div>
7253
7254<div class="doc_text">
7255
7256<h5>Syntax:</h5>
7257<pre>
7258 declare void @llvm.trap()
7259</pre>
7260
7261<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7262<p>The '<tt>llvm.trap</tt>' intrinsic.</p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7263
7264<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7265<p>None.</p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7266
7267<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7268<p>This intrinsics is lowered to the target dependent trap instruction. If the
7269 target does not have a trap instruction, this intrinsic will be lowered to
7270 the call of the <tt>abort()</tt> function.</p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7271
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]7272</div>
7273
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7274<!-- _______________________________________________________________________ -->
7275<div class="doc_subsubsection">
Misha Brukmandccb0252008-11-22 23:55:29 +0000[diff] [blame]7276 <a name="int_stackprotector">'<tt>llvm.stackprotector</tt>' Intrinsic</a>
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7277</div>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7278
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7279<div class="doc_text">
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7280
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7281<h5>Syntax:</h5>
7282<pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7283 declare void @llvm.stackprotector( i8* &lt;guard&gt;, i8** &lt;slot&gt; )
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7284</pre>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7285
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7286<h5>Overview:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7287<p>The <tt>llvm.stackprotector</tt> intrinsic takes the <tt>guard</tt> and
7288 stores it onto the stack at <tt>slot</tt>. The stack slot is adjusted to
7289 ensure that it is placed on the stack before local variables.</p>
7290
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7291<h5>Arguments:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7292<p>The <tt>llvm.stackprotector</tt> intrinsic requires two pointer
7293 arguments. The first argument is the value loaded from the stack
7294 guard <tt>@__stack_chk_guard</tt>. The second variable is an <tt>alloca</tt>
7295 that has enough space to hold the value of the guard.</p>
7296
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7297<h5>Semantics:</h5>
Bill Wendlinge910b4c2009-07-20 02:29:24 +0000[diff] [blame]7298<p>This intrinsic causes the prologue/epilogue inserter to force the position of
7299 the <tt>AllocaInst</tt> stack slot to be before local variables on the
7300 stack. This is to ensure that if a local variable on the stack is
7301 overwritten, it will destroy the value of the guard. When the function exits,
7302 the guard on the stack is checked against the original guard. If they're
7303 different, then the program aborts by calling the <tt>__stack_chk_fail()</tt>
7304 function.</p>
7305
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]7306</div>
7307
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7308<!-- _______________________________________________________________________ -->
7309<div class="doc_subsubsection">
7310 <a name="int_objectsize">'<tt>llvm.objectsize</tt>' Intrinsic</a>
7311</div>
7312
7313<div class="doc_text">
7314
7315<h5>Syntax:</h5>
7316<pre>
Eric Christopher8295a0a2009-12-23 00:29:49 +0000[diff] [blame]7317 declare i32 @llvm.objectsize.i32( i8* &lt;object&gt;, i1 &lt;type&gt; )
7318 declare i64 @llvm.objectsize.i64( i8* &lt;object&gt;, i1 &lt;type&gt; )
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7319</pre>
7320
7321<h5>Overview:</h5>
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7322<p>The <tt>llvm.objectsize</tt> intrinsic is designed to provide information
Eric Christopherd003c5b2010-01-08 21:42:39 +0000[diff] [blame]7323 to the optimizers to discover at compile time either a) when an
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7324 operation like memcpy will either overflow a buffer that corresponds to
7325 an object, or b) to determine that a runtime check for overflow isn't
7326 necessary. An object in this context means an allocation of a
Eric Christopher8295a0a2009-12-23 00:29:49 +0000[diff] [blame]7327 specific class, structure, array, or other object.</p>
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7328
7329<h5>Arguments:</h5>
7330<p>The <tt>llvm.objectsize</tt> intrinsic takes two arguments. The first
Eric Christopher8295a0a2009-12-23 00:29:49 +0000[diff] [blame]7331 argument is a pointer to or into the <tt>object</tt>. The second argument
7332 is a boolean 0 or 1. This argument determines whether you want the
7333 maximum (0) or minimum (1) bytes remaining. This needs to be a literal 0 or
7334 1, variables are not allowed.</p>
7335
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7336<h5>Semantics:</h5>
7337<p>The <tt>llvm.objectsize</tt> intrinsic is lowered to either a constant
Eric Christopher6c7e8a02009-12-05 02:46:03 +0000[diff] [blame]7338 representing the size of the object concerned or <tt>i32/i64 -1 or 0</tt>
7339 (depending on the <tt>type</tt> argument if the size cannot be determined
7340 at compile time.</p>
Eric Christopher0e671492009-11-30 08:03:53 +0000[diff] [blame]7341
7342</div>
7343
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]7344<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]7345<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7346<address>
7347 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]7348 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7349 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]7350 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]7351
7352 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencer05fe4b02006-03-14 05:39:39 +0000[diff] [blame]7353 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]7354 Last modified: $Date$
7355</address>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]7356
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]7357</body>
7358</html>