Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 242d61d1c974e1c79f063a29fe8448d543f24cb1 / . / docs / LangRef.html
blob: 8d9b60da2a6819c8d8eee11ca198b1b4fceb0c77 [file] [log] [blame]
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3<html>
4<head>
5 <title>LLVM Assembly Language Reference Manual</title>
Reid Spencer3921c742004-08-26 20:44:00 +0000[diff] [blame]6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
7 <meta name="author" content="Chris Lattner">
8 <meta name="description"
9 content="LLVM Assembly Language Reference Manual.">
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]10 <link rel="stylesheet" href="llvm.css" type="text/css">
11</head>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]12
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]13<body>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]14
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]15<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]16<ol>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]17 <li><a href="#abstract">Abstract</a></li>
18 <li><a href="#introduction">Introduction</a></li>
19 <li><a href="#identifiers">Identifiers</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]20 <li><a href="#highlevel">High Level Structure</a>
21 <ol>
22 <li><a href="#modulestructure">Module Structure</a></li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]23 <li><a href="#linkage">Linkage Types</a></li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]24 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]25 <li><a href="#namedtypes">Named Types</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]26 <li><a href="#globalvars">Global Variables</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]27 <li><a href="#functionstructure">Functions</a></li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]28 <li><a href="#aliasstructure">Aliases</a></li>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]29 <li><a href="#paramattrs">Parameter Attributes</a></li>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]30 <li><a href="#fnattrs">Function Attributes</a></li>
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]31 <li><a href="#gc">Garbage Collector Names</a></li>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]32 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]33 <li><a href="#datalayout">Data Layout</a></li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]34 </ol>
35 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]36 <li><a href="#typesystem">Type System</a>
37 <ol>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]38 <li><a href="#t_classifications">Type Classifications</a></li>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]39 <li><a href="#t_primitive">Primitive Types</a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]40 <ol>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]41 <li><a href="#t_floating">Floating Point Types</a></li>
42 <li><a href="#t_void">Void Type</a></li>
43 <li><a href="#t_label">Label Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]44 </ol>
45 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]46 <li><a href="#t_derived">Derived Types</a>
47 <ol>
Chris Lattnerb9488a62007-12-18 06:18:21 +0000[diff] [blame]48 <li><a href="#t_integer">Integer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]49 <li><a href="#t_array">Array Type</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]50 <li><a href="#t_function">Function Type</a></li>
51 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]52 <li><a href="#t_struct">Structure Type</a></li>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]53 <li><a href="#t_pstruct">Packed Structure Type</a></li>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]54 <li><a href="#t_vector">Vector Type</a></li>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]55 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]56 </ol>
57 </li>
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame^]58 <li><a href="#t_uprefs">Type Up-references</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]59 </ol>
60 </li>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]61 <li><a href="#constants">Constants</a>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]62 <ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]63 <li><a href="#simpleconstants">Simple Constants</a></li>
64 <li><a href="#aggregateconstants">Aggregate Constants</a></li>
65 <li><a href="#globalconstants">Global Variable and Function Addresses</a></li>
66 <li><a href="#undefvalues">Undefined Values</a></li>
67 <li><a href="#constantexprs">Constant Expressions</a></li>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]68 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]69 </li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]70 <li><a href="#othervalues">Other Values</a>
71 <ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]72 <li><a href="#inlineasm">Inline Assembler Expressions</a></li>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]73 </ol>
74 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]75 <li><a href="#instref">Instruction Reference</a>
76 <ol>
77 <li><a href="#terminators">Terminator Instructions</a>
78 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]79 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
80 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]81 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
82 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]83 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]84 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]85 </ol>
86 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]87 <li><a href="#binaryops">Binary Operations</a>
88 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]89 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
90 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
91 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]92 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
93 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
94 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]95 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
96 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
97 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]98 </ol>
99 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]100 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
101 <ol>
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]102 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
103 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
104 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]105 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]106 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]107 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]108 </ol>
109 </li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]110 <li><a href="#vectorops">Vector Operations</a>
111 <ol>
112 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
113 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
114 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]115 </ol>
116 </li>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]117 <li><a href="#aggregateops">Aggregate Operations</a>
118 <ol>
119 <li><a href="#i_extractvalue">'<tt>extractvalue</tt>' Instruction</a></li>
120 <li><a href="#i_insertvalue">'<tt>insertvalue</tt>' Instruction</a></li>
121 </ol>
122 </li>
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]123 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]124 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]125 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
126 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
127 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]128 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
129 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
130 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]131 </ol>
132 </li>
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]133 <li><a href="#convertops">Conversion Operations</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]134 <ol>
135 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
136 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
137 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
138 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
139 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]140 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
141 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
142 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
143 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]144 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
145 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]146 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]147 </ol>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]148 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]149 <li><a href="#otherops">Other Operations</a>
150 <ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]151 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
152 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]153 <li><a href="#i_vicmp">'<tt>vicmp</tt>' Instruction</a></li>
154 <li><a href="#i_vfcmp">'<tt>vfcmp</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]155 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]156 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]157 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]158 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]159 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]160 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]161 </ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]162 </li>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]163 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]164 <ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]165 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
166 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]167 <li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
168 <li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
169 <li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]170 </ol>
171 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]172 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
173 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]174 <li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
175 <li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
176 <li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]177 </ol>
178 </li>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]179 <li><a href="#int_codegen">Code Generator Intrinsics</a>
180 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]181 <li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
182 <li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
183 <li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
184 <li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
185 <li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
186 <li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
187 <li><a href="#int_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswell7123e272004-04-09 16:43:20 +0000[diff] [blame]188 </ol>
189 </li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]190 <li><a href="#int_libc">Standard C Library Intrinsics</a>
191 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]192 <li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
193 <li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
194 <li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
195 <li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
196 <li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]197 <li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
198 <li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
199 <li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]200 </ol>
201 </li>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]202 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]203 <ol>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]204 <li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]205 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
206 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
207 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]208 <li><a href="#int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic </a></li>
209 <li><a href="#int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic </a></li>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]210 </ol>
211 </li>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]212 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]213 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]214 <li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]215 <ol>
216 <li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]217 </ol>
218 </li>
Bill Wendling3c44f5b2008-11-18 22:10:53 +0000[diff] [blame]219 <li><a href="#int_atomics">Atomic intrinsics</a>
220 <ol>
221 <li><a href="#int_memory_barrier"><tt>llvm.memory_barrier</tt></a></li>
222 <li><a href="#int_atomic_cmp_swap"><tt>llvm.atomic.cmp.swap</tt></a></li>
223 <li><a href="#int_atomic_swap"><tt>llvm.atomic.swap</tt></a></li>
224 <li><a href="#int_atomic_load_add"><tt>llvm.atomic.load.add</tt></a></li>
225 <li><a href="#int_atomic_load_sub"><tt>llvm.atomic.load.sub</tt></a></li>
226 <li><a href="#int_atomic_load_and"><tt>llvm.atomic.load.and</tt></a></li>
227 <li><a href="#int_atomic_load_nand"><tt>llvm.atomic.load.nand</tt></a></li>
228 <li><a href="#int_atomic_load_or"><tt>llvm.atomic.load.or</tt></a></li>
229 <li><a href="#int_atomic_load_xor"><tt>llvm.atomic.load.xor</tt></a></li>
230 <li><a href="#int_atomic_load_max"><tt>llvm.atomic.load.max</tt></a></li>
231 <li><a href="#int_atomic_load_min"><tt>llvm.atomic.load.min</tt></a></li>
232 <li><a href="#int_atomic_load_umax"><tt>llvm.atomic.load.umax</tt></a></li>
233 <li><a href="#int_atomic_load_umin"><tt>llvm.atomic.load.umin</tt></a></li>
234 </ol>
235 </li>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]236 <li><a href="#int_general">General intrinsics</a>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]237 <ol>
Reid Spencer20677642007-07-20 19:59:11 +0000[diff] [blame]238 <li><a href="#int_var_annotation">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]239 '<tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]240 <li><a href="#int_annotation">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]241 '<tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]242 <li><a href="#int_trap">
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]243 '<tt>llvm.trap</tt>' Intrinsic</a></li>
244 <li><a href="#int_stackprotector">
245 '<tt>llvm.stackprotector</tt>' Intrinsic</a></li>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]246 </ol>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]247 </li>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]248 </ol>
249 </li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]250</ol>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]251
252<div class="doc_author">
253 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
254 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]255</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]256
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]257<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]258<div class="doc_section"> <a name="abstract">Abstract </a></div>
259<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]260
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]261<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]262<p>This document is a reference manual for the LLVM assembly language.
Bill Wendling837f39b2008-08-05 22:29:16 +0000[diff] [blame]263LLVM is a Static Single Assignment (SSA) based representation that provides
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]264type safety, low-level operations, flexibility, and the capability of
265representing 'all' high-level languages cleanly. It is the common code
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]266representation used throughout all phases of the LLVM compilation
267strategy.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]268</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]269
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]270<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]271<div class="doc_section"> <a name="introduction">Introduction</a> </div>
272<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]273
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]274<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]275
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]276<p>The LLVM code representation is designed to be used in three
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]277different forms: as an in-memory compiler IR, as an on-disk bitcode
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]278representation (suitable for fast loading by a Just-In-Time compiler),
279and as a human readable assembly language representation. This allows
280LLVM to provide a powerful intermediate representation for efficient
281compiler transformations and analysis, while providing a natural means
282to debug and visualize the transformations. The three different forms
283of LLVM are all equivalent. This document describes the human readable
284representation and notation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]285
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]286<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]287while being expressive, typed, and extensible at the same time. It
288aims to be a "universal IR" of sorts, by being at a low enough level
289that high-level ideas may be cleanly mapped to it (similar to how
290microprocessors are "universal IR's", allowing many source languages to
291be mapped to them). By providing type information, LLVM can be used as
292the target of optimizations: for example, through pointer analysis, it
293can be proven that a C automatic variable is never accessed outside of
294the current function... allowing it to be promoted to a simple SSA
295value instead of a memory location.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]296
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]297</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]298
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]299<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]300<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]301
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]302<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]303
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]304<p>It is important to note that this document describes 'well formed'
305LLVM assembly language. There is a difference between what the parser
306accepts and what is considered 'well formed'. For example, the
307following instruction is syntactically okay, but not well formed:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]308
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]309<div class="doc_code">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]310<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]311%x = <a href="#i_add">add</a> i32 1, %x
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]312</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]313</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]314
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]315<p>...because the definition of <tt>%x</tt> does not dominate all of
316its uses. The LLVM infrastructure provides a verification pass that may
317be used to verify that an LLVM module is well formed. This pass is
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]318automatically run by the parser after parsing input assembly and by
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]319the optimizer before it outputs bitcode. The violations pointed out
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]320by the verifier pass indicate bugs in transformation passes or input to
321the parser.</p>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]322</div>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]323
Chris Lattnercc689392007-10-03 17:34:29 +0000[diff] [blame]324<!-- Describe the typesetting conventions here. -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]325
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]326<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]327<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]328<!-- *********************************************************************** -->
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]329
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]330<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]331
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]332 <p>LLVM identifiers come in two basic types: global and local. Global
333 identifiers (functions, global variables) begin with the @ character. Local
334 identifiers (register names, types) begin with the % character. Additionally,
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]335 there are three different formats for identifiers, for different purposes:</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]336
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]337<ol>
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]338 <li>Named values are represented as a string of characters with their prefix.
339 For example, %foo, @DivisionByZero, %a.really.long.identifier. The actual
340 regular expression used is '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]341 Identifiers which require other characters in their names can be surrounded
Daniel Dunbar76dea952008-10-14 23:51:43 +0000[diff] [blame]342 with quotes. Special characters may be escaped using "\xx" where xx is the
343 ASCII code for the character in hexadecimal. In this way, any character can
344 be used in a name value, even quotes themselves.
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]345
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]346 <li>Unnamed values are represented as an unsigned numeric value with their
347 prefix. For example, %12, @2, %44.</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]348
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]349 <li>Constants, which are described in a <a href="#constants">section about
350 constants</a>, below.</li>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]351</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]352
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]353<p>LLVM requires that values start with a prefix for two reasons: Compilers
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]354don't need to worry about name clashes with reserved words, and the set of
355reserved words may be expanded in the future without penalty. Additionally,
356unnamed identifiers allow a compiler to quickly come up with a temporary
357variable without having to avoid symbol table conflicts.</p>
358
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]359<p>Reserved words in LLVM are very similar to reserved words in other
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]360languages. There are keywords for different opcodes
361('<tt><a href="#i_add">add</a></tt>',
362 '<tt><a href="#i_bitcast">bitcast</a></tt>',
363 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]364href="#t_void">void</a></tt>', '<tt><a href="#t_primitive">i32</a></tt>', etc...),
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]365and others. These reserved words cannot conflict with variable names, because
Reid Spencer2c452282007-08-07 14:34:28 +0000[diff] [blame]366none of them start with a prefix character ('%' or '@').</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]367
368<p>Here is an example of LLVM code to multiply the integer variable
369'<tt>%X</tt>' by 8:</p>
370
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]371<p>The easy way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]372
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]373<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]374<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]375%result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]376</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]377</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]378
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]379<p>After strength reduction:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]380
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]381<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]382<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]383%result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]384</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]385</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]386
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]387<p>And the hard way:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]388
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]389<div class="doc_code">
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]390<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]391<a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
392<a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
393%result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]394</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]395</div>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]396
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]397<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
398important lexical features of LLVM:</p>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]399
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]400<ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]401
402 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
403 line.</li>
404
405 <li>Unnamed temporaries are created when the result of a computation is not
406 assigned to a named value.</li>
407
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]408 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]409
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]410</ol>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]411
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]412<p>...and it also shows a convention that we follow in this document. When
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]413demonstrating instructions, we will follow an instruction with a comment that
414defines the type and name of value produced. Comments are shown in italic
415text.</p>
416
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]417</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]418
419<!-- *********************************************************************** -->
420<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
421<!-- *********************************************************************** -->
422
423<!-- ======================================================================= -->
424<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
425</div>
426
427<div class="doc_text">
428
429<p>LLVM programs are composed of "Module"s, each of which is a
430translation unit of the input programs. Each module consists of
431functions, global variables, and symbol table entries. Modules may be
432combined together with the LLVM linker, which merges function (and
433global variable) definitions, resolves forward declarations, and merges
434symbol table entries. Here is an example of the "hello world" module:</p>
435
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]436<div class="doc_code">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]437<pre><i>; Declare the string constant as a global constant...</i>
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]438<a href="#identifiers">@.LC0</a> = <a href="#linkage_internal">internal</a> <a
439 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]440
441<i>; External declaration of the puts function</i>
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]442<a href="#functionstructure">declare</a> i32 @puts(i8 *) <i>; i32(i8 *)* </i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]443
444<i>; Definition of main function</i>
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]445define i32 @main() { <i>; i32()* </i>
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]446 <i>; Convert [13 x i8]* to i8 *...</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]447 %cast210 = <a
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]448 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]449
450 <i>; Call puts function to write out the string to stdout...</i>
451 <a
Chris Lattnera89e5f12007-06-12 17:00:26 +0000[diff] [blame]452 href="#i_call">call</a> i32 @puts(i8 * %cast210) <i>; i32</i>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]453 <a
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]454 href="#i_ret">ret</a> i32 0<br>}<br>
455</pre>
456</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]457
458<p>This example is made up of a <a href="#globalvars">global variable</a>
459named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
460function, and a <a href="#functionstructure">function definition</a>
461for "<tt>main</tt>".</p>
462
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]463<p>In general, a module is made up of a list of global values,
464where both functions and global variables are global values. Global values are
465represented by a pointer to a memory location (in this case, a pointer to an
466array of char, and a pointer to a function), and have one of the following <a
467href="#linkage">linkage types</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]468
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]469</div>
470
471<!-- ======================================================================= -->
472<div class="doc_subsection">
473 <a name="linkage">Linkage Types</a>
474</div>
475
476<div class="doc_text">
477
478<p>
479All Global Variables and Functions have one of the following types of linkage:
480</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]481
482<dl>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]483
Rafael Espindolabb46f522009-01-15 20:18:42 +0000[diff] [blame]484 <dt><tt><b><a name="linkage_private">private</a></b></tt>: </dt>
485
486 <dd>Global values with private linkage are only directly accessible by
487 objects in the current module. In particular, linking code into a module with
488 an private global value may cause the private to be renamed as necessary to
489 avoid collisions. Because the symbol is private to the module, all
490 references can be updated. This doesn't show up in any symbol table in the
491 object file.
492 </dd>
493
Dale Johannesen2307a7f2008-05-23 23:13:41 +0000[diff] [blame]494 <dt><tt><b><a name="linkage_internal">internal</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]495
Duncan Sands81d05c22009-01-16 09:29:46 +0000[diff] [blame]496 <dd> Similar to private, but the value shows as a local symbol (STB_LOCAL in
Rafael Espindolabb46f522009-01-15 20:18:42 +0000[diff] [blame]497 the case of ELF) in the object file. This corresponds to the notion of the
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]498 '<tt>static</tt>' keyword in C.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]499 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]500
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]501 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]502
Chris Lattner4887bd82007-01-14 06:51:48 +0000[diff] [blame]503 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
504 the same name when linkage occurs. This is typically used to implement
505 inline functions, templates, or other code which must be generated in each
506 translation unit that uses it. Unreferenced <tt>linkonce</tt> globals are
507 allowed to be discarded.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]508 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]509
Dale Johannesen2307a7f2008-05-23 23:13:41 +0000[diff] [blame]510 <dt><tt><b><a name="linkage_common">common</a></b></tt>: </dt>
511
512 <dd>"<tt>common</tt>" linkage is exactly the same as <tt>linkonce</tt>
513 linkage, except that unreferenced <tt>common</tt> globals may not be
514 discarded. This is used for globals that may be emitted in multiple
515 translation units, but that are not guaranteed to be emitted into every
516 translation unit that uses them. One example of this is tentative
517 definitions in C, such as "<tt>int X;</tt>" at global scope.
518 </dd>
519
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]520 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]521
Dale Johannesen2307a7f2008-05-23 23:13:41 +0000[diff] [blame]522 <dd>"<tt>weak</tt>" linkage is the same as <tt>common</tt> linkage, except
523 that some targets may choose to emit different assembly sequences for them
524 for target-dependent reasons. This is used for globals that are declared
525 "weak" in C source code.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]526 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]527
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]528 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]529
530 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
531 pointer to array type. When two global variables with appending linkage are
532 linked together, the two global arrays are appended together. This is the
533 LLVM, typesafe, equivalent of having the system linker append together
534 "sections" with identical names when .o files are linked.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]535 </dd>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]536
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]537 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt>: </dt>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]538 <dd>The semantics of this linkage follow the ELF object file model: the
539 symbol is weak until linked, if not linked, the symbol becomes null instead
540 of being an undefined reference.
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]541 </dd>
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]542
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]543 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnere5d947b2004-12-09 16:36:40 +0000[diff] [blame]544
545 <dd>If none of the above identifiers are used, the global is externally
546 visible, meaning that it participates in linkage and can be used to resolve
547 external symbol references.
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]548 </dd>
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]549</dl>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]550
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]551 <p>
552 The next two types of linkage are targeted for Microsoft Windows platform
553 only. They are designed to support importing (exporting) symbols from (to)
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]554 DLLs (Dynamic Link Libraries).
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]555 </p>
556
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]557 <dl>
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]558 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt>: </dt>
559
560 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
561 or variable via a global pointer to a pointer that is set up by the DLL
562 exporting the symbol. On Microsoft Windows targets, the pointer name is
Dan Gohman79564122009-01-12 21:35:55 +0000[diff] [blame]563 formed by combining <code>__imp_</code> and the function or variable name.
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]564 </dd>
565
566 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt>: </dt>
567
568 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
569 pointer to a pointer in a DLL, so that it can be referenced with the
570 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
Dan Gohman79564122009-01-12 21:35:55 +0000[diff] [blame]571 name is formed by combining <code>__imp_</code> and the function or variable
Anton Korobeynikovb74ed072006-09-14 18:23:27 +0000[diff] [blame]572 name.
573 </dd>
574
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]575</dl>
576
Dan Gohmanf0032762008-11-24 17:18:39 +0000[diff] [blame]577<p>For example, since the "<tt>.LC0</tt>"
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]578variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
579variable and was linked with this one, one of the two would be renamed,
580preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
581external (i.e., lacking any linkage declarations), they are accessible
Reid Spencerac8d2762007-01-05 00:59:10 +0000[diff] [blame]582outside of the current module.</p>
583<p>It is illegal for a function <i>declaration</i>
584to have any linkage type other than "externally visible", <tt>dllimport</tt>,
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]585or <tt>extern_weak</tt>.</p>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]586<p>Aliases can have only <tt>external</tt>, <tt>internal</tt> and <tt>weak</tt>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]587linkages.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]588</div>
589
590<!-- ======================================================================= -->
591<div class="doc_subsection">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]592 <a name="callingconv">Calling Conventions</a>
593</div>
594
595<div class="doc_text">
596
597<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
598and <a href="#i_invoke">invokes</a> can all have an optional calling convention
599specified for the call. The calling convention of any pair of dynamic
600caller/callee must match, or the behavior of the program is undefined. The
601following calling conventions are supported by LLVM, and more may be added in
602the future:</p>
603
604<dl>
605 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
606
607 <dd>This calling convention (the default if no other calling convention is
608 specified) matches the target C calling conventions. This calling convention
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]609 supports varargs function calls and tolerates some mismatch in the declared
Reid Spencerc28d2bc2006-12-31 21:30:18 +0000[diff] [blame]610 prototype and implemented declaration of the function (as does normal C).
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]611 </dd>
612
613 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
614
615 <dd>This calling convention attempts to make calls as fast as possible
616 (e.g. by passing things in registers). This calling convention allows the
617 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]618 without having to conform to an externally specified ABI (Application Binary
619 Interface). Implementations of this convention should allow arbitrary
Arnold Schwaighofer9097d142008-05-14 09:17:12 +0000[diff] [blame]620 <a href="CodeGenerator.html#tailcallopt">tail call optimization</a> to be
621 supported. This calling convention does not support varargs and requires the
622 prototype of all callees to exactly match the prototype of the function
623 definition.
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]624 </dd>
625
626 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
627
628 <dd>This calling convention attempts to make code in the caller as efficient
629 as possible under the assumption that the call is not commonly executed. As
630 such, these calls often preserve all registers so that the call does not break
631 any live ranges in the caller side. This calling convention does not support
632 varargs and requires the prototype of all callees to exactly match the
633 prototype of the function definition.
634 </dd>
635
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]636 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]637
638 <dd>Any calling convention may be specified by number, allowing
639 target-specific calling conventions to be used. Target specific calling
640 conventions start at 64.
641 </dd>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]642</dl>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]643
644<p>More calling conventions can be added/defined on an as-needed basis, to
645support pascal conventions or any other well-known target-independent
646convention.</p>
647
648</div>
649
650<!-- ======================================================================= -->
651<div class="doc_subsection">
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]652 <a name="visibility">Visibility Styles</a>
653</div>
654
655<div class="doc_text">
656
657<p>
658All Global Variables and Functions have one of the following visibility styles:
659</p>
660
661<dl>
662 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
663
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]664 <dd>On targets that use the ELF object file format, default visibility means
665 that the declaration is visible to other
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]666 modules and, in shared libraries, means that the declared entity may be
667 overridden. On Darwin, default visibility means that the declaration is
668 visible to other modules. Default visibility corresponds to "external
669 linkage" in the language.
670 </dd>
671
672 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
673
674 <dd>Two declarations of an object with hidden visibility refer to the same
675 object if they are in the same shared object. Usually, hidden visibility
676 indicates that the symbol will not be placed into the dynamic symbol table,
677 so no other module (executable or shared library) can reference it
678 directly.
679 </dd>
680
Anton Korobeynikov6f9896f2007-04-29 18:35:00 +0000[diff] [blame]681 <dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
682
683 <dd>On ELF, protected visibility indicates that the symbol will be placed in
684 the dynamic symbol table, but that references within the defining module will
685 bind to the local symbol. That is, the symbol cannot be overridden by another
686 module.
687 </dd>
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]688</dl>
689
690</div>
691
692<!-- ======================================================================= -->
693<div class="doc_subsection">
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]694 <a name="namedtypes">Named Types</a>
695</div>
696
697<div class="doc_text">
698
699<p>LLVM IR allows you to specify name aliases for certain types. This can make
700it easier to read the IR and make the IR more condensed (particularly when
701recursive types are involved). An example of a name specification is:
702</p>
703
704<div class="doc_code">
705<pre>
706%mytype = type { %mytype*, i32 }
707</pre>
708</div>
709
710<p>You may give a name to any <a href="#typesystem">type</a> except "<a
711href="t_void">void</a>". Type name aliases may be used anywhere a type is
712expected with the syntax "%mytype".</p>
713
714<p>Note that type names are aliases for the structural type that they indicate,
715and that you can therefore specify multiple names for the same type. This often
716leads to confusing behavior when dumping out a .ll file. Since LLVM IR uses
717structural typing, the name is not part of the type. When printing out LLVM IR,
718the printer will pick <em>one name</em> to render all types of a particular
719shape. This means that if you have code where two different source types end up
720having the same LLVM type, that the dumper will sometimes print the "wrong" or
721unexpected type. This is an important design point and isn't going to
722change.</p>
723
724</div>
725
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]726<!-- ======================================================================= -->
727<div class="doc_subsection">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]728 <a name="globalvars">Global Variables</a>
729</div>
730
731<div class="doc_text">
732
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]733<p>Global variables define regions of memory allocated at compilation time
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]734instead of run-time. Global variables may optionally be initialized, may have
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]735an explicit section to be placed in, and may have an optional explicit alignment
736specified. A variable may be defined as "thread_local", which means that it
737will not be shared by threads (each thread will have a separated copy of the
738variable). A variable may be defined as a global "constant," which indicates
739that the contents of the variable will <b>never</b> be modified (enabling better
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]740optimization, allowing the global data to be placed in the read-only section of
741an executable, etc). Note that variables that need runtime initialization
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]742cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner3689a342005-02-12 19:30:21 +0000[diff] [blame]743
744<p>
745LLVM explicitly allows <em>declarations</em> of global variables to be marked
746constant, even if the final definition of the global is not. This capability
747can be used to enable slightly better optimization of the program, but requires
748the language definition to guarantee that optimizations based on the
749'constantness' are valid for the translation units that do not include the
750definition.
751</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]752
753<p>As SSA values, global variables define pointer values that are in
754scope (i.e. they dominate) all basic blocks in the program. Global
755variables always define a pointer to their "content" type because they
756describe a region of memory, and all memory objects in LLVM are
757accessed through pointers.</p>
758
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]759<p>A global variable may be declared to reside in a target-specifc numbered
760address space. For targets that support them, address spaces may affect how
761optimizations are performed and/or what target instructions are used to access
Christopher Lambd49e18d2007-12-12 08:44:39 +0000[diff] [blame]762the variable. The default address space is zero. The address space qualifier
763must precede any other attributes.</p>
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]764
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]765<p>LLVM allows an explicit section to be specified for globals. If the target
766supports it, it will emit globals to the section specified.</p>
767
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]768<p>An explicit alignment may be specified for a global. If not present, or if
769the alignment is set to zero, the alignment of the global is set by the target
770to whatever it feels convenient. If an explicit alignment is specified, the
771global is forced to have at least that much alignment. All alignments must be
772a power of 2.</p>
773
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]774<p>For example, the following defines a global in a numbered address space with
775an initializer, section, and alignment:</p>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]776
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]777<div class="doc_code">
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]778<pre>
Dan Gohman398873c2009-01-11 00:40:00 +0000[diff] [blame]779@G = addrspace(5) constant float 1.0, section "foo", align 4
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]780</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]781</div>
Chris Lattner68027ea2007-01-14 00:27:09 +0000[diff] [blame]782
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]783</div>
784
785
786<!-- ======================================================================= -->
787<div class="doc_subsection">
788 <a name="functionstructure">Functions</a>
789</div>
790
791<div class="doc_text">
792
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]793<p>LLVM function definitions consist of the "<tt>define</tt>" keyord,
794an optional <a href="#linkage">linkage type</a>, an optional
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]795<a href="#visibility">visibility style</a>, an optional
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]796<a href="#callingconv">calling convention</a>, a return type, an optional
797<a href="#paramattrs">parameter attribute</a> for the return type, a function
798name, a (possibly empty) argument list (each with optional
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]799<a href="#paramattrs">parameter attributes</a>), optional
800<a href="#fnattrs">function attributes</a>, an optional section,
801an optional alignment, an optional <a href="#gc">garbage collector name</a>,
Chris Lattner0c46a7d2008-10-04 18:10:21 +0000[diff] [blame]802an opening curly brace, a list of basic blocks, and a closing curly brace.
Anton Korobeynikov8cea37b2007-01-23 12:35:46 +0000[diff] [blame]803
804LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
805optional <a href="#linkage">linkage type</a>, an optional
806<a href="#visibility">visibility style</a>, an optional
807<a href="#callingconv">calling convention</a>, a return type, an optional
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]808<a href="#paramattrs">parameter attribute</a> for the return type, a function
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]809name, a possibly empty list of arguments, an optional alignment, and an optional
Gordon Henriksene754abe2007-12-10 03:30:21 +0000[diff] [blame]810<a href="#gc">garbage collector name</a>.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]811
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]812<p>A function definition contains a list of basic blocks, forming the CFG
813(Control Flow Graph) for
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]814the function. Each basic block may optionally start with a label (giving the
815basic block a symbol table entry), contains a list of instructions, and ends
816with a <a href="#terminators">terminator</a> instruction (such as a branch or
817function return).</p>
818
Chris Lattner4a3c9012007-06-08 16:52:14 +0000[diff] [blame]819<p>The first basic block in a function is special in two ways: it is immediately
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]820executed on entrance to the function, and it is not allowed to have predecessor
821basic blocks (i.e. there can not be any branches to the entry block of a
822function). Because the block can have no predecessors, it also cannot have any
823<a href="#i_phi">PHI nodes</a>.</p>
824
Chris Lattner88f6c462005-11-12 00:45:07 +0000[diff] [blame]825<p>LLVM allows an explicit section to be specified for functions. If the target
826supports it, it will emit functions to the section specified.</p>
827
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]828<p>An explicit alignment may be specified for a function. If not present, or if
829the alignment is set to zero, the alignment of the function is set by the target
830to whatever it feels convenient. If an explicit alignment is specified, the
831function is forced to have at least that much alignment. All alignments must be
832a power of 2.</p>
833
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]834 <h5>Syntax:</h5>
835
836<div class="doc_code">
Chris Lattner50ad45c2008-10-13 16:55:18 +0000[diff] [blame]837<tt>
838define [<a href="#linkage">linkage</a>] [<a href="#visibility">visibility</a>]
839 [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>]
840 &lt;ResultType&gt; @&lt;FunctionName&gt; ([argument list])
841 [<a href="#fnattrs">fn Attrs</a>] [section "name"] [align N]
842 [<a href="#gc">gc</a>] { ... }
843</tt>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]844</div>
845
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]846</div>
847
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]848
849<!-- ======================================================================= -->
850<div class="doc_subsection">
851 <a name="aliasstructure">Aliases</a>
852</div>
853<div class="doc_text">
854 <p>Aliases act as "second name" for the aliasee value (which can be either
Anton Korobeynikov726d45c2008-03-22 08:36:14 +0000[diff] [blame]855 function, global variable, another alias or bitcast of global value). Aliases
856 may have an optional <a href="#linkage">linkage type</a>, and an
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]857 optional <a href="#visibility">visibility style</a>.</p>
858
859 <h5>Syntax:</h5>
860
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]861<div class="doc_code">
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]862<pre>
Duncan Sands0b23ac12008-09-12 20:48:21 +0000[diff] [blame]863@&lt;Name&gt; = alias [Linkage] [Visibility] &lt;AliaseeTy&gt; @&lt;Aliasee&gt;
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]864</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]865</div>
Anton Korobeynikov8b0a8c82007-04-25 14:27:10 +0000[diff] [blame]866
867</div>
868
869
870
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]871<!-- ======================================================================= -->
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]872<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
873<div class="doc_text">
874 <p>The return type and each parameter of a function type may have a set of
875 <i>parameter attributes</i> associated with them. Parameter attributes are
876 used to communicate additional information about the result or parameters of
Duncan Sandsdc024672007-11-27 13:23:08 +0000[diff] [blame]877 a function. Parameter attributes are considered to be part of the function,
878 not of the function type, so functions with different parameter attributes
879 can have the same function type.</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]880
Reid Spencer950e9f82007-01-15 18:27:39 +0000[diff] [blame]881 <p>Parameter attributes are simple keywords that follow the type specified. If
882 multiple parameter attributes are needed, they are space separated. For
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]883 example:</p>
884
885<div class="doc_code">
886<pre>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]887declare i32 @printf(i8* noalias , ...)
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]888declare i32 @atoi(i8 zeroext)
889declare signext i8 @returns_signed_char()
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]890</pre>
891</div>
892
Duncan Sandsdc024672007-11-27 13:23:08 +0000[diff] [blame]893 <p>Note that any attributes for the function result (<tt>nounwind</tt>,
894 <tt>readonly</tt>) come immediately after the argument list.</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]895
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]896 <p>Currently, only the following parameter attributes are defined:</p>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]897 <dl>
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]898 <dt><tt>zeroext</tt></dt>
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]899 <dd>This indicates to the code generator that the parameter or return value
900 should be zero-extended to a 32-bit value by the caller (for a parameter)
901 or the callee (for a return value).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]902
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]903 <dt><tt>signext</tt></dt>
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]904 <dd>This indicates to the code generator that the parameter or return value
905 should be sign-extended to a 32-bit value by the caller (for a parameter)
906 or the callee (for a return value).</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]907
Anton Korobeynikov34d6dea2007-01-28 14:30:45 +0000[diff] [blame]908 <dt><tt>inreg</tt></dt>
Dale Johannesenc9c6da62008-09-25 20:47:45 +0000[diff] [blame]909 <dd>This indicates that this parameter or return value should be treated
910 in a special target-dependent fashion during while emitting code for a
911 function call or return (usually, by putting it in a register as opposed
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]912 to memory, though some targets use it to distinguish between two different
913 kinds of registers). Use of this attribute is target-specific.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]914
Duncan Sandsedb05df2008-10-06 08:14:18 +0000[diff] [blame]915 <dt><tt><a name="byval">byval</a></tt></dt>
Chris Lattner0747baa2008-01-15 04:34:22 +0000[diff] [blame]916 <dd>This indicates that the pointer parameter should really be passed by
917 value to the function. The attribute implies that a hidden copy of the
918 pointee is made between the caller and the callee, so the callee is unable
Chris Lattnerebec6782008-08-05 18:21:08 +0000[diff] [blame]919 to modify the value in the callee. This attribute is only valid on LLVM
Chris Lattner0747baa2008-01-15 04:34:22 +0000[diff] [blame]920 pointer arguments. It is generally used to pass structs and arrays by
Duncan Sandsedb05df2008-10-06 08:14:18 +0000[diff] [blame]921 value, but is also valid on pointers to scalars. The copy is considered to
922 belong to the caller not the callee (for example,
923 <tt><a href="#readonly">readonly</a></tt> functions should not write to
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]924 <tt>byval</tt> parameters). This is not a valid attribute for return
925 values. </dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]926
Anton Korobeynikov34d6dea2007-01-28 14:30:45 +0000[diff] [blame]927 <dt><tt>sret</tt></dt>
Duncan Sandse26dec62008-02-18 04:19:38 +0000[diff] [blame]928 <dd>This indicates that the pointer parameter specifies the address of a
929 structure that is the return value of the function in the source program.
Chris Lattner66d922c2008-10-04 18:33:34 +0000[diff] [blame]930 This pointer must be guaranteed by the caller to be valid: loads and stores
931 to the structure may be assumed by the callee to not to trap. This may only
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]932 be applied to the first parameter. This is not a valid attribute for
933 return values. </dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]934
Zhou Shengfebca342007-06-05 05:28:26 +0000[diff] [blame]935 <dt><tt>noalias</tt></dt>
Nick Lewycky02ff3082008-11-24 03:41:24 +0000[diff] [blame]936 <dd>This indicates that the pointer does not alias any global or any other
937 parameter. The caller is responsible for ensuring that this is the
Nick Lewyckyb2b32fd2008-11-24 05:00:44 +0000[diff] [blame]938 case. On a function return value, <tt>noalias</tt> additionally indicates
939 that the pointer does not alias any other pointers visible to the
Nick Lewyckyf23d0d32008-12-19 06:39:12 +0000[diff] [blame]940 caller. For further details, please see the discussion of the NoAlias
941 response in
942 <a href="http://llvm.org/docs/AliasAnalysis.html#MustMayNo">alias
943 analysis</a>.</dd>
944
945 <dt><tt>nocapture</tt></dt>
946 <dd>This indicates that the callee does not make any copies of the pointer
947 that outlive the callee itself. This is not a valid attribute for return
948 values.</dd>
Chris Lattner47507de2008-01-11 06:20:47 +0000[diff] [blame]949
Duncan Sands50f19f52007-07-27 19:57:41 +0000[diff] [blame]950 <dt><tt>nest</tt></dt>
Duncan Sands0789b8b2008-07-08 09:27:25 +0000[diff] [blame]951 <dd>This indicates that the pointer parameter can be excised using the
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]952 <a href="#int_trampoline">trampoline intrinsics</a>. This is not a valid
953 attribute for return values.</dd>
Anton Korobeynikov7f705592007-01-12 19:20:47 +0000[diff] [blame]954 </dl>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]955
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]956</div>
957
958<!-- ======================================================================= -->
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]959<div class="doc_subsection">
Gordon Henriksen80a75bf2007-12-10 03:18:06 +0000[diff] [blame]960 <a name="gc">Garbage Collector Names</a>
961</div>
962
963<div class="doc_text">
964<p>Each function may specify a garbage collector name, which is simply a
965string.</p>
966
967<div class="doc_code"><pre
968>define void @f() gc "name" { ...</pre></div>
969
970<p>The compiler declares the supported values of <i>name</i>. Specifying a
971collector which will cause the compiler to alter its output in order to support
972the named garbage collection algorithm.</p>
973</div>
974
975<!-- ======================================================================= -->
976<div class="doc_subsection">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]977 <a name="fnattrs">Function Attributes</a>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]978</div>
979
980<div class="doc_text">
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]981
982<p>Function attributes are set to communicate additional information about
983 a function. Function attributes are considered to be part of the function,
984 not of the function type, so functions with different parameter attributes
985 can have the same function type.</p>
986
987 <p>Function attributes are simple keywords that follow the type specified. If
988 multiple attributes are needed, they are space separated. For
989 example:</p>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]990
991<div class="doc_code">
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]992<pre>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]993define void @f() noinline { ... }
994define void @f() alwaysinline { ... }
995define void @f() alwaysinline optsize { ... }
996define void @f() optsize
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]997</pre>
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]998</div>
999
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1000<dl>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1001<dt><tt>alwaysinline</tt></dt>
Chris Lattner88d4b592008-10-04 18:23:17 +0000[diff] [blame]1002<dd>This attribute indicates that the inliner should attempt to inline this
1003function into callers whenever possible, ignoring any active inlining size
1004threshold for this caller.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1005
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1006<dt><tt>noinline</tt></dt>
Chris Lattner88d4b592008-10-04 18:23:17 +0000[diff] [blame]1007<dd>This attribute indicates that the inliner should never inline this function
Chris Lattner94b5f7d2008-10-05 17:14:59 +0000[diff] [blame]1008in any situation. This attribute may not be used together with the
Chris Lattner88d4b592008-10-04 18:23:17 +0000[diff] [blame]1009<tt>alwaysinline</tt> attribute.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1010
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1011<dt><tt>optsize</tt></dt>
Devang Patel66c6c652008-09-29 18:34:44 +0000[diff] [blame]1012<dd>This attribute suggests that optimization passes and code generator passes
Chris Lattner88d4b592008-10-04 18:23:17 +0000[diff] [blame]1013make choices that keep the code size of this function low, and otherwise do
1014optimizations specifically to reduce code size.</dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1015
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1016<dt><tt>noreturn</tt></dt>
Chris Lattner88d4b592008-10-04 18:23:17 +0000[diff] [blame]1017<dd>This function attribute indicates that the function never returns normally.
1018This produces undefined behavior at runtime if the function ever does
1019dynamically return.</dd>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1020
1021<dt><tt>nounwind</tt></dt>
Chris Lattner88d4b592008-10-04 18:23:17 +0000[diff] [blame]1022<dd>This function attribute indicates that the function never returns with an
1023unwind or exceptional control flow. If the function does unwind, its runtime
1024behavior is undefined.</dd>
1025
1026<dt><tt>readnone</tt></dt>
Duncan Sandsedb05df2008-10-06 08:14:18 +0000[diff] [blame]1027<dd>This attribute indicates that the function computes its result (or the
1028exception it throws) based strictly on its arguments, without dereferencing any
1029pointer arguments or otherwise accessing any mutable state (e.g. memory, control
1030registers, etc) visible to caller functions. It does not write through any
1031pointer arguments (including <tt><a href="#byval">byval</a></tt> arguments) and
1032never changes any state visible to callers.</dd>
Devang Patel2c9c3e72008-09-26 23:51:19 +0000[diff] [blame]1033
Duncan Sandsedb05df2008-10-06 08:14:18 +0000[diff] [blame]1034<dt><tt><a name="readonly">readonly</a></tt></dt>
1035<dd>This attribute indicates that the function does not write through any
1036pointer arguments (including <tt><a href="#byval">byval</a></tt> arguments)
1037or otherwise modify any state (e.g. memory, control registers, etc) visible to
1038caller functions. It may dereference pointer arguments and read state that may
1039be set in the caller. A readonly function always returns the same value (or
1040throws the same exception) when called with the same set of arguments and global
1041state.</dd>
Bill Wendling31359ba2008-11-13 01:02:51 +0000[diff] [blame]1042
1043<dt><tt><a name="ssp">ssp</a></tt></dt>
Bill Wendlingbaa39d82008-11-26 19:19:05 +0000[diff] [blame]1044<dd>This attribute indicates that the function should emit a stack smashing
Bill Wendling31359ba2008-11-13 01:02:51 +0000[diff] [blame]1045protector. It is in the form of a "canary"&mdash;a random value placed on the
1046stack before the local variables that's checked upon return from the function to
1047see if it has been overwritten. A heuristic is used to determine if a function
Bill Wendlingbaa39d82008-11-26 19:19:05 +0000[diff] [blame]1048needs stack protectors or not.
Bill Wendling31359ba2008-11-13 01:02:51 +0000[diff] [blame]1049
Bill Wendlingfbaa7ed2008-11-26 19:07:40 +0000[diff] [blame]1050<p>If a function that has an <tt>ssp</tt> attribute is inlined into a function
1051that doesn't have an <tt>ssp</tt> attribute, then the resulting function will
1052have an <tt>ssp</tt> attribute.</p></dd>
1053
1054<dt><tt>sspreq</tt></dt>
Bill Wendlingbaa39d82008-11-26 19:19:05 +0000[diff] [blame]1055<dd>This attribute indicates that the function should <em>always</em> emit a
Bill Wendling31359ba2008-11-13 01:02:51 +0000[diff] [blame]1056stack smashing protector. This overrides the <tt><a href="#ssp">ssp</a></tt>
Bill Wendlingbaa39d82008-11-26 19:19:05 +0000[diff] [blame]1057function attribute.
Bill Wendlingfbaa7ed2008-11-26 19:07:40 +0000[diff] [blame]1058
1059<p>If a function that has an <tt>sspreq</tt> attribute is inlined into a
1060function that doesn't have an <tt>sspreq</tt> attribute or which has
1061an <tt>ssp</tt> attribute, then the resulting function will have
1062an <tt>sspreq</tt> attribute.</p></dd>
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1063</dl>
1064
Devang Patelf8b94812008-09-04 23:05:13 +0000[diff] [blame]1065</div>
1066
1067<!-- ======================================================================= -->
1068<div class="doc_subsection">
Chris Lattner1eeeb0c2006-04-08 04:40:53 +0000[diff] [blame]1069 <a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1070</div>
1071
1072<div class="doc_text">
1073<p>
1074Modules may contain "module-level inline asm" blocks, which corresponds to the
1075GCC "file scope inline asm" blocks. These blocks are internally concatenated by
1076LLVM and treated as a single unit, but may be separated in the .ll file if
1077desired. The syntax is very simple:
1078</p>
1079
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1080<div class="doc_code">
1081<pre>
1082module asm "inline asm code goes here"
1083module asm "more can go here"
1084</pre>
1085</div>
Chris Lattner4e9aba72006-01-23 23:23:47 +0000[diff] [blame]1086
1087<p>The strings can contain any character by escaping non-printable characters.
1088 The escape sequence used is simply "\xx" where "xx" is the two digit hex code
1089 for the number.
1090</p>
1091
1092<p>
1093 The inline asm code is simply printed to the machine code .s file when
1094 assembly code is generated.
1095</p>
1096</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1097
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1098<!-- ======================================================================= -->
1099<div class="doc_subsection">
1100 <a name="datalayout">Data Layout</a>
1101</div>
1102
1103<div class="doc_text">
1104<p>A module may specify a target specific data layout string that specifies how
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]1105data is to be laid out in memory. The syntax for the data layout is simply:</p>
1106<pre> target datalayout = "<i>layout specification</i>"</pre>
1107<p>The <i>layout specification</i> consists of a list of specifications
1108separated by the minus sign character ('-'). Each specification starts with a
1109letter and may include other information after the letter to define some
1110aspect of the data layout. The specifications accepted are as follows: </p>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1111<dl>
1112 <dt><tt>E</tt></dt>
1113 <dd>Specifies that the target lays out data in big-endian form. That is, the
1114 bits with the most significance have the lowest address location.</dd>
1115 <dt><tt>e</tt></dt>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1116 <dd>Specifies that the target lays out data in little-endian form. That is,
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1117 the bits with the least significance have the lowest address location.</dd>
1118 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1119 <dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
1120 <i>preferred</i> alignments. All sizes are in bits. Specifying the <i>pref</i>
1121 alignment is optional. If omitted, the preceding <tt>:</tt> should be omitted
1122 too.</dd>
1123 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1124 <dd>This specifies the alignment for an integer type of a given bit
1125 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
1126 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1127 <dd>This specifies the alignment for a vector type of a given bit
1128 <i>size</i>.</dd>
1129 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1130 <dd>This specifies the alignment for a floating point type of a given bit
1131 <i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
1132 (double).</dd>
1133 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
1134 <dd>This specifies the alignment for an aggregate type of a given bit
1135 <i>size</i>.</dd>
1136</dl>
1137<p>When constructing the data layout for a given target, LLVM starts with a
1138default set of specifications which are then (possibly) overriden by the
1139specifications in the <tt>datalayout</tt> keyword. The default specifications
1140are given in this list:</p>
1141<ul>
1142 <li><tt>E</tt> - big endian</li>
1143 <li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
1144 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
1145 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
1146 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
1147 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1148 <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]1149 alignment of 64-bits</li>
1150 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
1151 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
1152 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
1153 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
1154 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
1155</ul>
Chris Lattnerebec6782008-08-05 18:21:08 +0000[diff] [blame]1156<p>When LLVM is determining the alignment for a given type, it uses the
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]1157following rules:</p>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1158<ol>
1159 <li>If the type sought is an exact match for one of the specifications, that
1160 specification is used.</li>
1161 <li>If no match is found, and the type sought is an integer type, then the
1162 smallest integer type that is larger than the bitwidth of the sought type is
1163 used. If none of the specifications are larger than the bitwidth then the the
1164 largest integer type is used. For example, given the default specifications
1165 above, the i7 type will use the alignment of i8 (next largest) while both
1166 i65 and i256 will use the alignment of i64 (largest specified).</li>
1167 <li>If no match is found, and the type sought is a vector type, then the
1168 largest vector type that is smaller than the sought vector type will be used
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]1169 as a fall back. This happens because &lt;128 x double&gt; can be implemented
1170 in terms of 64 &lt;2 x double&gt;, for example.</li>
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]1171</ol>
1172</div>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1173
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1174<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1175<div class="doc_section"> <a name="typesystem">Type System</a> </div>
1176<!-- *********************************************************************** -->
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1177
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1178<div class="doc_text">
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1179
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1180<p>The LLVM type system is one of the most important features of the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1181intermediate representation. Being typed enables a number of
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]1182optimizations to be performed on the intermediate representation directly,
1183without having to do
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1184extra analyses on the side before the transformation. A strong type
1185system makes it easier to read the generated code and enables novel
1186analyses and transformations that are not feasible to perform on normal
1187three address code representations.</p>
Chris Lattnerfa730212004-12-09 16:11:40 +0000[diff] [blame]1188
1189</div>
1190
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1191<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1192<div class="doc_subsection"> <a name="t_classifications">Type
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1193Classifications</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1194<div class="doc_text">
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1195<p>The types fall into a few useful
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1196classifications:</p>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1197
1198<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1199 <tbody>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1200 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1201 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1202 <td><a href="#t_integer">integer</a></td>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1203 <td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1204 </tr>
1205 <tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1206 <td><a href="#t_floating">floating point</a></td>
1207 <td><tt>float, double, x86_fp80, fp128, ppc_fp128</tt></td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1208 </tr>
1209 <tr>
1210 <td><a name="t_firstclass">first class</a></td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1211 <td><a href="#t_integer">integer</a>,
1212 <a href="#t_floating">floating point</a>,
1213 <a href="#t_pointer">pointer</a>,
Dan Gohman0066db62008-06-18 18:42:13 +0000[diff] [blame]1214 <a href="#t_vector">vector</a>,
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]1215 <a href="#t_struct">structure</a>,
1216 <a href="#t_array">array</a>,
Dan Gohmanade5faa2008-05-23 22:50:26 +0000[diff] [blame]1217 <a href="#t_label">label</a>.
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1218 </td>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1219 </tr>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1220 <tr>
1221 <td><a href="#t_primitive">primitive</a></td>
1222 <td><a href="#t_label">label</a>,
1223 <a href="#t_void">void</a>,
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1224 <a href="#t_floating">floating point</a>.</td>
1225 </tr>
1226 <tr>
1227 <td><a href="#t_derived">derived</a></td>
1228 <td><a href="#t_integer">integer</a>,
1229 <a href="#t_array">array</a>,
1230 <a href="#t_function">function</a>,
1231 <a href="#t_pointer">pointer</a>,
1232 <a href="#t_struct">structure</a>,
1233 <a href="#t_pstruct">packed structure</a>,
1234 <a href="#t_vector">vector</a>,
1235 <a href="#t_opaque">opaque</a>.
Dan Gohman01ac1012008-10-14 16:32:04 +0000[diff] [blame]1236 </td>
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1237 </tr>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1238 </tbody>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1239</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1240
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1241<p>The <a href="#t_firstclass">first class</a> types are perhaps the
1242most important. Values of these types are the only ones which can be
1243produced by instructions, passed as arguments, or used as operands to
Dan Gohmanc4b49eb2008-05-23 21:53:15 +0000[diff] [blame]1244instructions.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1245</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1246
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1247<!-- ======================================================================= -->
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1248<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1249
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1250<div class="doc_text">
1251<p>The primitive types are the fundamental building blocks of the LLVM
1252system.</p>
1253
Chris Lattner8f8c7b72008-01-04 04:34:14 +0000[diff] [blame]1254</div>
1255
Chris Lattner4f69f462008-01-04 04:32:38 +0000[diff] [blame]1256<!-- _______________________________________________________________________ -->
1257<div class="doc_subsubsection"> <a name="t_floating">Floating Point Types</a> </div>
1258
1259<div class="doc_text">
1260 <table>
1261 <tbody>
1262 <tr><th>Type</th><th>Description</th></tr>
1263 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
1264 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
1265 <tr><td><tt>fp128</tt></td><td>128-bit floating point value (112-bit mantissa)</td></tr>
1266 <tr><td><tt>x86_fp80</tt></td><td>80-bit floating point value (X87)</td></tr>
1267 <tr><td><tt>ppc_fp128</tt></td><td>128-bit floating point value (two 64-bits)</td></tr>
1268 </tbody>
1269 </table>
1270</div>
1271
1272<!-- _______________________________________________________________________ -->
1273<div class="doc_subsubsection"> <a name="t_void">Void Type</a> </div>
1274
1275<div class="doc_text">
1276<h5>Overview:</h5>
1277<p>The void type does not represent any value and has no size.</p>
1278
1279<h5>Syntax:</h5>
1280
1281<pre>
1282 void
1283</pre>
1284</div>
1285
1286<!-- _______________________________________________________________________ -->
1287<div class="doc_subsubsection"> <a name="t_label">Label Type</a> </div>
1288
1289<div class="doc_text">
1290<h5>Overview:</h5>
1291<p>The label type represents code labels.</p>
1292
1293<h5>Syntax:</h5>
1294
1295<pre>
1296 label
1297</pre>
1298</div>
1299
1300
1301<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1302<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1303
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1304<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1305
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1306<p>The real power in LLVM comes from the derived types in the system.
1307This is what allows a programmer to represent arrays, functions,
1308pointers, and other useful types. Note that these derived types may be
1309recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1310
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1311</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1312
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1313<!-- _______________________________________________________________________ -->
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1314<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
1315
1316<div class="doc_text">
1317
1318<h5>Overview:</h5>
1319<p>The integer type is a very simple derived type that simply specifies an
1320arbitrary bit width for the integer type desired. Any bit width from 1 bit to
13212^23-1 (about 8 million) can be specified.</p>
1322
1323<h5>Syntax:</h5>
1324
1325<pre>
1326 iN
1327</pre>
1328
1329<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
1330value.</p>
1331
1332<h5>Examples:</h5>
1333<table class="layout">
Chris Lattnerb9488a62007-12-18 06:18:21 +0000[diff] [blame]1334 <tbody>
1335 <tr>
1336 <td><tt>i1</tt></td>
1337 <td>a single-bit integer.</td>
1338 </tr><tr>
1339 <td><tt>i32</tt></td>
1340 <td>a 32-bit integer.</td>
1341 </tr><tr>
1342 <td><tt>i1942652</tt></td>
1343 <td>a really big integer of over 1 million bits.</td>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1344 </tr>
Chris Lattnerb9488a62007-12-18 06:18:21 +0000[diff] [blame]1345 </tbody>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1346</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1347
1348<p>Note that the code generator does not yet support large integer types
1349to be used as function return types. The specific limit on how large a
1350return type the code generator can currently handle is target-dependent;
1351currently it's often 64 bits for 32-bit targets and 128 bits for 64-bit
1352targets.</p>
1353
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1354</div>
Reid Spencer2b916312007-05-16 18:44:01 +0000[diff] [blame]1355
1356<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1357<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1358
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1359<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1360
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1361<h5>Overview:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1362
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1363<p>The array type is a very simple derived type that arranges elements
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1364sequentially in memory. The array type requires a size (number of
1365elements) and an underlying data type.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1366
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1367<h5>Syntax:</h5>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1368
1369<pre>
1370 [&lt;# elements&gt; x &lt;elementtype&gt;]
1371</pre>
1372
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1373<p>The number of elements is a constant integer value; elementtype may
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1374be any type with a size.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1375
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1376<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1377<table class="layout">
1378 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1379 <td class="left"><tt>[40 x i32]</tt></td>
1380 <td class="left">Array of 40 32-bit integer values.</td>
1381 </tr>
1382 <tr class="layout">
1383 <td class="left"><tt>[41 x i32]</tt></td>
1384 <td class="left">Array of 41 32-bit integer values.</td>
1385 </tr>
1386 <tr class="layout">
1387 <td class="left"><tt>[4 x i8]</tt></td>
1388 <td class="left">Array of 4 8-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1389 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1390</table>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1391<p>Here are some examples of multidimensional arrays:</p>
1392<table class="layout">
1393 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1394 <td class="left"><tt>[3 x [4 x i32]]</tt></td>
1395 <td class="left">3x4 array of 32-bit integer values.</td>
1396 </tr>
1397 <tr class="layout">
1398 <td class="left"><tt>[12 x [10 x float]]</tt></td>
1399 <td class="left">12x10 array of single precision floating point values.</td>
1400 </tr>
1401 <tr class="layout">
1402 <td class="left"><tt>[2 x [3 x [4 x i16]]]</tt></td>
1403 <td class="left">2x3x4 array of 16-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1404 </tr>
1405</table>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1406
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1407<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
1408length array. Normally, accesses past the end of an array are undefined in
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1409LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
1410As a special case, however, zero length arrays are recognized to be variable
1411length. This allows implementation of 'pascal style arrays' with the LLVM
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1412type "{ i32, [0 x float]}", for example.</p>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]1413
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1414<p>Note that the code generator does not yet support large aggregate types
1415to be used as function return types. The specific limit on how large an
1416aggregate return type the code generator can currently handle is
1417target-dependent, and also dependent on the aggregate element types.</p>
1418
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1419</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1420
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1421<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1422<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1423<div class="doc_text">
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1424
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1425<h5>Overview:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1426
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1427<p>The function type can be thought of as a function signature. It
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1428consists of a return type and a list of formal parameter types. The
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1429return type of a function type is a scalar type, a void type, or a struct type.
Devang Patel7a3ad1a2008-03-24 20:52:42 +0000[diff] [blame]1430If the return type is a struct type then all struct elements must be of first
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1431class types, and the struct must have at least one element.</p>
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]1432
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1433<h5>Syntax:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1434
1435<pre>
1436 &lt;returntype list&gt; (&lt;parameter list&gt;)
1437</pre>
1438
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1439<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
Misha Brukmanc24b7582004-08-12 20:16:08 +0000[diff] [blame]1440specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]1441which indicates that the function takes a variable number of arguments.
1442Variable argument functions can access their arguments with the <a
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]1443 href="#int_varargs">variable argument handling intrinsic</a> functions.
1444'<tt>&lt;returntype list&gt;</tt>' is a comma-separated list of
1445<a href="#t_firstclass">first class</a> type specifiers.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]1446
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1447<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1448<table class="layout">
1449 <tr class="layout">
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1450 <td class="left"><tt>i32 (i32)</tt></td>
1451 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1452 </td>
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1453 </tr><tr class="layout">
Reid Spencer9445e9a2007-07-19 23:13:04 +0000[diff] [blame]1454 <td class="left"><tt>float&nbsp;(i16&nbsp;signext,&nbsp;i32&nbsp;*)&nbsp;*
Reid Spencerf17a0b72006-12-31 07:20:23 +0000[diff] [blame]1455 </tt></td>
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1456 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
1457 an <tt>i16</tt> that should be sign extended and a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1458 <a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1459 <tt>float</tt>.
1460 </td>
1461 </tr><tr class="layout">
1462 <td class="left"><tt>i32 (i8*, ...)</tt></td>
1463 <td class="left">A vararg function that takes at least one
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]1464 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
Reid Spencer92f82302006-12-31 07:18:34 +0000[diff] [blame]1465 which returns an integer. This is the signature for <tt>printf</tt> in
1466 LLVM.
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1467 </td>
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1468 </tr><tr class="layout">
1469 <td class="left"><tt>{i32, i32} (i32)</tt></td>
Misha Brukmanb0a57aa2008-11-27 06:41:20 +0000[diff] [blame]1470 <td class="left">A function taking an <tt>i32</tt>, returning two
1471 <tt>i32</tt> values as an aggregate of type <tt>{ i32, i32 }</tt>
Devang Patela582f402008-03-24 05:35:41 +0000[diff] [blame]1472 </td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1473 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1474</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]1475
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1476</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1477<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1478<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1479<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1480<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1481<p>The structure type is used to represent a collection of data members
1482together in memory. The packing of the field types is defined to match
1483the ABI of the underlying processor. The elements of a structure may
1484be any type that has a size.</p>
1485<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1486and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1487field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1488instruction.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1489<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1490<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1491<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1492<table class="layout">
1493 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1494 <td class="left"><tt>{ i32, i32, i32 }</tt></td>
1495 <td class="left">A triple of three <tt>i32</tt> values</td>
1496 </tr><tr class="layout">
1497 <td class="left"><tt>{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}</tt></td>
1498 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1499 second element is a <a href="#t_pointer">pointer</a> to a
1500 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1501 an <tt>i32</tt>.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1502 </tr>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1503</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1504
1505<p>Note that the code generator does not yet support large aggregate types
1506to be used as function return types. The specific limit on how large an
1507aggregate return type the code generator can currently handle is
1508target-dependent, and also dependent on the aggregate element types.</p>
1509
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1510</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1511
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1512<!-- _______________________________________________________________________ -->
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1513<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1514</div>
1515<div class="doc_text">
1516<h5>Overview:</h5>
1517<p>The packed structure type is used to represent a collection of data members
1518together in memory. There is no padding between fields. Further, the alignment
1519of a packed structure is 1 byte. The elements of a packed structure may
1520be any type that has a size.</p>
1521<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1522and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1523field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1524instruction.</p>
1525<h5>Syntax:</h5>
1526<pre> &lt; { &lt;type list&gt; } &gt; <br></pre>
1527<h5>Examples:</h5>
1528<table class="layout">
1529 <tr class="layout">
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1530 <td class="left"><tt>&lt; { i32, i32, i32 } &gt;</tt></td>
1531 <td class="left">A triple of three <tt>i32</tt> values</td>
1532 </tr><tr class="layout">
Bill Wendlinge36dccc2008-09-07 10:26:33 +0000[diff] [blame]1533 <td class="left">
1534<tt>&lt;&nbsp;{&nbsp;float,&nbsp;i32&nbsp;(i32)*&nbsp;}&nbsp;&gt;</tt></td>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1535 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1536 second element is a <a href="#t_pointer">pointer</a> to a
1537 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1538 an <tt>i32</tt>.</td>
Andrew Lenharth75e10682006-12-08 17:13:00 +0000[diff] [blame]1539 </tr>
1540</table>
1541</div>
1542
1543<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1544<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1545<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1546<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1547<p>As in many languages, the pointer type represents a pointer or
Christopher Lamb284d9922007-12-11 09:31:00 +0000[diff] [blame]1548reference to another object, which must live in memory. Pointer types may have
1549an optional address space attribute defining the target-specific numbered
1550address space where the pointed-to object resides. The default address space is
1551zero.</p>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1552<h5>Syntax:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]1553<pre> &lt;type&gt; *<br></pre>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]1554<h5>Examples:</h5>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1555<table class="layout">
1556 <tr class="layout">
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]1557 <td class="left"><tt>[4 x i32]*</tt></td>
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1558 <td class="left">A <a href="#t_pointer">pointer</a> to <a
1559 href="#t_array">array</a> of four <tt>i32</tt> values.</td>
1560 </tr>
1561 <tr class="layout">
1562 <td class="left"><tt>i32 (i32 *) *</tt></td>
1563 <td class="left"> A <a href="#t_pointer">pointer</a> to a <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1564 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1565 <tt>i32</tt>.</td>
1566 </tr>
1567 <tr class="layout">
1568 <td class="left"><tt>i32 addrspace(5)*</tt></td>
1569 <td class="left">A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value
1570 that resides in address space #5.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1571 </tr>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1572</table>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1573</div>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1574
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1575<!-- _______________________________________________________________________ -->
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1576<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1577<div class="doc_text">
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1578
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1579<h5>Overview:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1580
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1581<p>A vector type is a simple derived type that represents a vector
1582of elements. Vector types are used when multiple primitive data
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1583are operated in parallel using a single instruction (SIMD).
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1584A vector type requires a size (number of
Chris Lattnerb8d172f2005-11-10 01:44:22 +0000[diff] [blame]1585elements) and an underlying primitive data type. Vectors must have a power
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1586of two length (1, 2, 4, 8, 16 ...). Vector types are
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1587considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1588
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1589<h5>Syntax:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1590
1591<pre>
1592 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1593</pre>
1594
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1595<p>The number of elements is a constant integer value; elementtype may
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1596be any integer or floating point type.</p>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1597
Chris Lattnera58561b2004-08-12 19:12:28 +0000[diff] [blame]1598<h5>Examples:</h5>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1599
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1600<table class="layout">
1601 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1602 <td class="left"><tt>&lt;4 x i32&gt;</tt></td>
1603 <td class="left">Vector of 4 32-bit integer values.</td>
1604 </tr>
1605 <tr class="layout">
1606 <td class="left"><tt>&lt;8 x float&gt;</tt></td>
1607 <td class="left">Vector of 8 32-bit floating-point values.</td>
1608 </tr>
1609 <tr class="layout">
1610 <td class="left"><tt>&lt;2 x i64&gt;</tt></td>
1611 <td class="left">Vector of 2 64-bit integer values.</td>
Reid Spencerd3f876c2004-11-01 08:19:36 +0000[diff] [blame]1612 </tr>
1613</table>
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]1614
1615<p>Note that the code generator does not yet support large vector types
1616to be used as function return types. The specific limit on how large a
1617vector return type codegen can currently handle is target-dependent;
1618currently it's often a few times longer than a hardware vector register.</p>
1619
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]1620</div>
1621
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1622<!-- _______________________________________________________________________ -->
1623<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1624<div class="doc_text">
1625
1626<h5>Overview:</h5>
1627
1628<p>Opaque types are used to represent unknown types in the system. This
Gordon Henriksen8ac04ff2007-10-14 00:34:53 +0000[diff] [blame]1629corresponds (for example) to the C notion of a forward declared structure type.
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1630In LLVM, opaque types can eventually be resolved to any type (not just a
1631structure type).</p>
1632
1633<h5>Syntax:</h5>
1634
1635<pre>
1636 opaque
1637</pre>
1638
1639<h5>Examples:</h5>
1640
1641<table class="layout">
1642 <tr class="layout">
Chris Lattner23ff1f92007-12-19 05:04:11 +0000[diff] [blame]1643 <td class="left"><tt>opaque</tt></td>
1644 <td class="left">An opaque type.</td>
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1645 </tr>
1646</table>
1647</div>
1648
Chris Lattner242d61d2009-02-02 07:32:36 +0000[diff] [blame^]1649<!-- ======================================================================= -->
1650<div class="doc_subsection">
1651 <a name="t_uprefs">Type Up-references</a>
1652</div>
1653
1654<div class="doc_text">
1655<h5>Overview:</h5>
1656<p>
1657An "up reference" allows you to refer to a lexically enclosing type without
1658requiring it to have a name. For instance, a structure declaration may contain a
1659pointer to any of the types it is lexically a member of. Example of up
1660references (with their equivalent as named type declarations) include:</p>
1661
1662<pre>
1663 { \2 * } %x = type { %t* }
1664 { \2 }* %y = type { %y }*
1665 \1* %z = type %z*
1666</pre>
1667
1668<p>
1669An up reference is needed by the asmprinter for printing out cyclic types when
1670there is no declared name for a type in the cycle. Because the asmprinter does
1671not want to print out an infinite type string, it needs a syntax to handle
1672recursive types that have no names (all names are optional in llvm IR).
1673</p>
1674
1675<h5>Syntax:</h5>
1676<pre>
1677 \&lt;level&gt;
1678</pre>
1679
1680<p>
1681The level is the count of the lexical type that is being referred to.
1682</p>
1683
1684<h5>Examples:</h5>
1685
1686<table class="layout">
1687 <tr class="layout">
1688 <td class="left"><tt>\1*</tt></td>
1689 <td class="left">Self-referential pointer.</td>
1690 </tr>
1691 <tr class="layout">
1692 <td class="left"><tt>{ { \3*, i8 }, i32 }</tt></td>
1693 <td class="left">Recursive structure where the upref refers to the out-most
1694 structure.</td>
1695 </tr>
1696</table>
1697</div>
1698
1699</div>
1700
Chris Lattner69c11bb2005-04-25 17:34:15 +0000[diff] [blame]1701
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1702<!-- *********************************************************************** -->
1703<div class="doc_section"> <a name="constants">Constants</a> </div>
1704<!-- *********************************************************************** -->
1705
1706<div class="doc_text">
1707
1708<p>LLVM has several different basic types of constants. This section describes
1709them all and their syntax.</p>
1710
1711</div>
1712
1713<!-- ======================================================================= -->
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1714<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1715
1716<div class="doc_text">
1717
1718<dl>
1719 <dt><b>Boolean constants</b></dt>
1720
1721 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]1722 constants of the <tt><a href="#t_primitive">i1</a></tt> type.
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1723 </dd>
1724
1725 <dt><b>Integer constants</b></dt>
1726
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1727 <dd>Standard integers (such as '4') are constants of the <a
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]1728 href="#t_integer">integer</a> type. Negative numbers may be used with
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1729 integer types.
1730 </dd>
1731
1732 <dt><b>Floating point constants</b></dt>
1733
1734 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
1735 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattnera73afe02008-04-01 18:45:27 +0000[diff] [blame]1736 notation (see below). The assembler requires the exact decimal value of
1737 a floating-point constant. For example, the assembler accepts 1.25 but
1738 rejects 1.3 because 1.3 is a repeating decimal in binary. Floating point
1739 constants must have a <a href="#t_floating">floating point</a> type. </dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1740
1741 <dt><b>Null pointer constants</b></dt>
1742
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1743 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1744 and must be of <a href="#t_pointer">pointer type</a>.</dd>
1745
1746</dl>
1747
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1748<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1749of floating point constants. For example, the form '<tt>double
17500x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
17514.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencercc16dc32004-12-09 18:02:53 +0000[diff] [blame]1752(and the only time that they are generated by the disassembler) is when a
1753floating point constant must be emitted but it cannot be represented as a
1754decimal floating point number. For example, NaN's, infinities, and other
1755special values are represented in their IEEE hexadecimal format so that
1756assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1757
1758</div>
1759
1760<!-- ======================================================================= -->
1761<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
1762</div>
1763
1764<div class="doc_text">
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]1765<p>Aggregate constants arise from aggregation of simple constants
1766and smaller aggregate constants.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1767
1768<dl>
1769 <dt><b>Structure constants</b></dt>
1770
1771 <dd>Structure constants are represented with notation similar to structure
1772 type definitions (a comma separated list of elements, surrounded by braces
Chris Lattner64910ee2007-12-25 20:34:52 +0000[diff] [blame]1773 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* @G }</tt>",
1774 where "<tt>@G</tt>" is declared as "<tt>@G = external global i32</tt>". Structure constants
Chris Lattnerd4f6b172005-03-07 22:13:59 +0000[diff] [blame]1775 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1776 types of elements must match those specified by the type.
1777 </dd>
1778
1779 <dt><b>Array constants</b></dt>
1780
1781 <dd>Array constants are represented with notation similar to array type
1782 definitions (a comma separated list of elements, surrounded by square brackets
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1783 (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74 ]</tt>". Array
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1784 constants must have <a href="#t_array">array type</a>, and the number and
1785 types of elements must match those specified by the type.
1786 </dd>
1787
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1788 <dt><b>Vector constants</b></dt>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1789
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1790 <dd>Vector constants are represented with notation similar to vector type
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1791 definitions (a comma separated list of elements, surrounded by
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]1792 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32 42,
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]1793 i32 11, i32 74, i32 100 &gt;</tt>". Vector constants must have <a
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1794 href="#t_vector">vector type</a>, and the number and types of elements must
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1795 match those specified by the type.
1796 </dd>
1797
1798 <dt><b>Zero initialization</b></dt>
1799
1800 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
1801 value to zero of <em>any</em> type, including scalar and aggregate types.
1802 This is often used to avoid having to print large zero initializers (e.g. for
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]1803 large arrays) and is always exactly equivalent to using explicit zero
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1804 initializers.
1805 </dd>
1806</dl>
1807
1808</div>
1809
1810<!-- ======================================================================= -->
1811<div class="doc_subsection">
1812 <a name="globalconstants">Global Variable and Function Addresses</a>
1813</div>
1814
1815<div class="doc_text">
1816
1817<p>The addresses of <a href="#globalvars">global variables</a> and <a
1818href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]1819constants. These constants are explicitly referenced when the <a
1820href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1821href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1822file:</p>
1823
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1824<div class="doc_code">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1825<pre>
Chris Lattnera18a4242007-06-06 18:28:13 +0000[diff] [blame]1826@X = global i32 17
1827@Y = global i32 42
1828@Z = global [2 x i32*] [ i32* @X, i32* @Y ]
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1829</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1830</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1831
1832</div>
1833
1834<!-- ======================================================================= -->
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1835<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1836<div class="doc_text">
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1837 <p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1838 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1839 a constant is permitted.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1840
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1841 <p>Undefined values indicate to the compiler that the program is well defined
1842 no matter what value is used, giving the compiler more freedom to optimize.
1843 </p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1844</div>
1845
1846<!-- ======================================================================= -->
1847<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1848</div>
1849
1850<div class="doc_text">
1851
1852<p>Constant expressions are used to allow expressions involving other constants
1853to be used as constants. Constant expressions may be of any <a
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]1854href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1855that does not have side effects (e.g. load and call are not supported). The
1856following is the syntax for constant expressions:</p>
1857
1858<dl>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1859 <dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
1860 <dd>Truncate a constant to another type. The bit size of CST must be larger
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1861 than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1862
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1863 <dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
1864 <dd>Zero extend a constant to another type. The bit size of CST must be
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1865 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1866
1867 <dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
1868 <dd>Sign extend a constant to another type. The bit size of CST must be
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]1869 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1870
1871 <dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
1872 <dd>Truncate a floating point constant to another floating point type. The
1873 size of CST must be larger than the size of TYPE. Both types must be
1874 floating point.</dd>
1875
1876 <dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
1877 <dd>Floating point extend a constant to another type. The size of CST must be
1878 smaller or equal to the size of TYPE. Both types must be floating point.</dd>
1879
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]1880 <dt><b><tt>fptoui ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1881 <dd>Convert a floating point constant to the corresponding unsigned integer
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1882 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1883 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1884 of the same number of elements. If the value won't fit in the integer type,
1885 the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1886
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]1887 <dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1888 <dd>Convert a floating point constant to the corresponding signed integer
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1889 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1890 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1891 of the same number of elements. If the value won't fit in the integer type,
1892 the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1893
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]1894 <dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1895 <dd>Convert an unsigned integer constant to the corresponding floating point
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1896 constant. TYPE must be a scalar or vector floating point type. CST must be of
1897 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1898 of the same number of elements. If the value won't fit in the floating point
1899 type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1900
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]1901 <dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1902 <dd>Convert a signed integer constant to the corresponding floating point
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]1903 constant. TYPE must be a scalar or vector floating point type. CST must be of
1904 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1905 of the same number of elements. If the value won't fit in the floating point
1906 type, the results are undefined.</dd>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1907
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]1908 <dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
1909 <dd>Convert a pointer typed constant to the corresponding integer constant
1910 TYPE must be an integer type. CST must be of pointer type. The CST value is
1911 zero extended, truncated, or unchanged to make it fit in TYPE.</dd>
1912
1913 <dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
1914 <dd>Convert a integer constant to a pointer constant. TYPE must be a
1915 pointer type. CST must be of integer type. The CST value is zero extended,
1916 truncated, or unchanged to make it fit in a pointer size. This one is
1917 <i>really</i> dangerous!</dd>
1918
1919 <dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1920 <dd>Convert a constant, CST, to another TYPE. The size of CST and TYPE must be
1921 identical (same number of bits). The conversion is done as if the CST value
1922 was stored to memory and read back as TYPE. In other words, no bits change
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]1923 with this operator, just the type. This can be used for conversion of
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]1924 vector types to any other type, as long as they have the same bit width. For
Dan Gohman500233a2008-09-08 16:45:59 +0000[diff] [blame]1925 pointers it is only valid to cast to another pointer type. It is not valid
1926 to bitcast to or from an aggregate type.
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]1927 </dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1928
1929 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1930
1931 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1932 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1933 instruction, the index list may have zero or more indexes, which are required
1934 to make sense for the type of "CSTPTR".</dd>
1935
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1936 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
1937
1938 <dd>Perform the <a href="#i_select">select operation</a> on
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]1939 constants.</dd>
1940
1941 <dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
1942 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
1943
1944 <dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
1945 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1946
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]1947 <dt><b><tt>vicmp COND ( VAL1, VAL2 )</tt></b></dt>
1948 <dd>Performs the <a href="#i_vicmp">vicmp operation</a> on constants.</dd>
1949
1950 <dt><b><tt>vfcmp COND ( VAL1, VAL2 )</tt></b></dt>
1951 <dd>Performs the <a href="#i_vfcmp">vfcmp operation</a> on constants.</dd>
1952
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1953 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
1954
1955 <dd>Perform the <a href="#i_extractelement">extractelement
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]1956 operation</a> on constants.</dd>
Robert Bocchino9fbe1452006-01-10 19:31:34 +0000[diff] [blame]1957
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]1958 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
1959
1960 <dd>Perform the <a href="#i_insertelement">insertelement
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]1961 operation</a> on constants.</dd>
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]1962
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]1963
1964 <dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
1965
1966 <dd>Perform the <a href="#i_shufflevector">shufflevector
Reid Spencer01c42592006-12-04 19:23:19 +0000[diff] [blame]1967 operation</a> on constants.</dd>
Chris Lattnerc1989542006-04-08 00:13:41 +0000[diff] [blame]1968
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1969 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1970
Reid Spencer2dc45b82004-12-09 18:13:12 +0000[diff] [blame]1971 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1972 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1973 binary</a> operations. The constraints on operands are the same as those for
1974 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]1975 values are allowed).</dd>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1976</dl>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]1977</div>
Chris Lattner9ee5d222004-03-08 16:49:10 +0000[diff] [blame]1978
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]1979<!-- *********************************************************************** -->
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]1980<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
1981<!-- *********************************************************************** -->
1982
1983<!-- ======================================================================= -->
1984<div class="doc_subsection">
1985<a name="inlineasm">Inline Assembler Expressions</a>
1986</div>
1987
1988<div class="doc_text">
1989
1990<p>
1991LLVM supports inline assembler expressions (as opposed to <a href="#moduleasm">
1992Module-Level Inline Assembly</a>) through the use of a special value. This
1993value represents the inline assembler as a string (containing the instructions
1994to emit), a list of operand constraints (stored as a string), and a flag that
1995indicates whether or not the inline asm expression has side effects. An example
1996inline assembler expression is:
1997</p>
1998
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]1999<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2000<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2001i32 (i32) asm "bswap $0", "=r,r"
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2002</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2003</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2004
2005<p>
2006Inline assembler expressions may <b>only</b> be used as the callee operand of
2007a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we have:
2008</p>
2009
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2010<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2011<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2012%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]2013</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2014</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2015
2016<p>
2017Inline asms with side effects not visible in the constraint list must be marked
2018as having side effects. This is done through the use of the
2019'<tt>sideeffect</tt>' keyword, like so:
2020</p>
2021
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2022<div class="doc_code">
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2023<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2024call void asm sideeffect "eieio", ""()
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2025</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]2026</div>
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2027
2028<p>TODO: The format of the asm and constraints string still need to be
2029documented here. Constraints on what can be done (e.g. duplication, moving, etc
Chris Lattner4f993352008-10-04 18:36:02 +0000[diff] [blame]2030need to be documented). This is probably best done by reference to another
2031document that covers inline asm from a holistic perspective.
Chris Lattnere87d6532006-01-25 23:47:57 +0000[diff] [blame]2032</p>
2033
2034</div>
2035
2036<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2037<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
2038<!-- *********************************************************************** -->
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2039
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2040<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2041
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2042<p>The LLVM instruction set consists of several different
2043classifications of instructions: <a href="#terminators">terminator
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]2044instructions</a>, <a href="#binaryops">binary instructions</a>,
2045<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2046 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
2047instructions</a>.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2048
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2049</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2050
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2051<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2052<div class="doc_subsection"> <a name="terminators">Terminator
2053Instructions</a> </div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2054
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2055<div class="doc_text">
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2056
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2057<p>As mentioned <a href="#functionstructure">previously</a>, every
2058basic block in a program ends with a "Terminator" instruction, which
2059indicates which block should be executed after the current block is
2060finished. These terminator instructions typically yield a '<tt>void</tt>'
2061value: they produce control flow, not values (the one exception being
2062the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2063<p>There are six different terminator instructions: the '<a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2064 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
2065instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2066the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
2067 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
2068 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2069
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2070</div>
Chris Lattnerc3f59762004-12-09 17:30:23 +0000[diff] [blame]2071
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2072<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2073<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
2074Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2075<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2076<h5>Syntax:</h5>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2077<pre>
2078 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]2079 ret void <i>; Return from void function</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2080</pre>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2081
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2082<h5>Overview:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2083
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2084<p>The '<tt>ret</tt>' instruction is used to return control flow (and
2085optionally a value) from a function back to the caller.</p>
John Criswell4457dc92004-04-09 16:48:45 +0000[diff] [blame]2086<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2087returns a value and then causes control flow, and one that just causes
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2088control flow to occur.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2089
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2090<h5>Arguments:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2091
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2092<p>The '<tt>ret</tt>' instruction optionally accepts a single argument,
2093the return value. The type of the return value must be a
2094'<a href="#t_firstclass">first class</a>' type.</p>
2095
2096<p>A function is not <a href="#wellformed">well formed</a> if
2097it it has a non-void return type and contains a '<tt>ret</tt>'
2098instruction with no return value or a return value with a type that
2099does not match its type, or if it has a void return type and contains
2100a '<tt>ret</tt>' instruction with a return value.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2101
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2102<h5>Semantics:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2103
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2104<p>When the '<tt>ret</tt>' instruction is executed, control flow
2105returns back to the calling function's context. If the caller is a "<a
John Criswellfa081872004-06-25 15:16:57 +0000[diff] [blame]2106 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2107the instruction after the call. If the caller was an "<a
2108 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]2109at the beginning of the "normal" destination block. If the instruction
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2110returns a value, that value shall set the call or invoke instruction's
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]2111return value.</p>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2112
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2113<h5>Example:</h5>
Chris Lattnerf4cde4e2008-04-23 04:59:35 +0000[diff] [blame]2114
2115<pre>
2116 ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]2117 ret void <i>; Return from a void function</i>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]2118 ret { i32, i8 } { i32 4, i8 2 } <i>; Return an aggregate of values 4 and 2</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2119</pre>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]2120
Dan Gohmand8791e52009-01-24 15:58:40 +0000[diff] [blame]2121<p>Note that the code generator does not yet fully support large
2122 return values. The specific sizes that are currently supported are
2123 dependent on the target. For integers, on 32-bit targets the limit
2124 is often 64 bits, and on 64-bit targets the limit is often 128 bits.
2125 For aggregate types, the current limits are dependent on the element
2126 types; for example targets are often limited to 2 total integer
2127 elements and 2 total floating-point elements.</p>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]2128
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2129</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2130<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2131<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2132<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2133<h5>Syntax:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2134<pre> br i1 &lt;cond&gt;, label &lt;iftrue&gt;, label &lt;iffalse&gt;<br> br label &lt;dest&gt; <i>; Unconditional branch</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2135</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2136<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2137<p>The '<tt>br</tt>' instruction is used to cause control flow to
2138transfer to a different basic block in the current function. There are
2139two forms of this instruction, corresponding to a conditional branch
2140and an unconditional branch.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2141<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2142<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2143single '<tt>i1</tt>' value and two '<tt>label</tt>' values. The
Reid Spencerde151942007-02-19 23:54:10 +0000[diff] [blame]2144unconditional form of the '<tt>br</tt>' instruction takes a single
2145'<tt>label</tt>' value as a target.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2146<h5>Semantics:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2147<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2148argument is evaluated. If the value is <tt>true</tt>, control flows
2149to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
2150control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2151<h5>Example:</h5>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2152<pre>Test:<br> %cond = <a href="#i_icmp">icmp</a> eq, i32 %a, %b<br> br i1 %cond, label %IfEqual, label %IfUnequal<br>IfEqual:<br> <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2153 href="#i_ret">ret</a> i32 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> i32 0<br></pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2154</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2155<!-- _______________________________________________________________________ -->
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2156<div class="doc_subsubsection">
2157 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
2158</div>
2159
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2160<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2161<h5>Syntax:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2162
2163<pre>
2164 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
2165</pre>
2166
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2167<h5>Overview:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2168
2169<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
2170several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2171instruction, allowing a branch to occur to one of many possible
2172destinations.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2173
2174
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2175<h5>Arguments:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2176
2177<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
2178comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
2179an array of pairs of comparison value constants and '<tt>label</tt>'s. The
2180table is not allowed to contain duplicate constant entries.</p>
2181
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2182<h5>Semantics:</h5>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2183
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2184<p>The <tt>switch</tt> instruction specifies a table of values and
2185destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]2186table is searched for the given value. If the value is found, control flow is
2187transfered to the corresponding destination; otherwise, control flow is
2188transfered to the default destination.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2189
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2190<h5>Implementation:</h5>
2191
2192<p>Depending on properties of the target machine and the particular
2193<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell84114752004-06-25 16:05:06 +0000[diff] [blame]2194ways. For example, it could be generated as a series of chained conditional
2195branches or with a lookup table.</p>
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2196
2197<h5>Example:</h5>
2198
2199<pre>
2200 <i>; Emulate a conditional br instruction</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]2201 %Val = <a href="#i_zext">zext</a> i1 %value to i32
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]2202 switch i32 %Val, label %truedest [ i32 0, label %falsedest ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2203
2204 <i>; Emulate an unconditional br instruction</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2205 switch i32 0, label %dest [ ]
Chris Lattnerc88c17b2004-02-24 04:54:45 +0000[diff] [blame]2206
2207 <i>; Implement a jump table:</i>
Dan Gohman2a08c532009-01-04 23:44:43 +0000[diff] [blame]2208 switch i32 %val, label %otherwise [ i32 0, label %onzero
2209 i32 1, label %onone
2210 i32 2, label %ontwo ]
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2211</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2212</div>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2213
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2214<!-- _______________________________________________________________________ -->
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2215<div class="doc_subsubsection">
2216 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
2217</div>
2218
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2219<div class="doc_text">
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2220
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2221<h5>Syntax:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2222
2223<pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]2224 &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]2225 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2226</pre>
2227
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]2228<h5>Overview:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2229
2230<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
2231function, with the possibility of control flow transfer to either the
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]2232'<tt>normal</tt>' label or the
2233'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2234"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
2235"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswelle4c57cc2005-05-12 16:52:32 +0000[diff] [blame]2236href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]2237continued at the dynamically nearest "exception" label.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2238
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2239<h5>Arguments:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2240
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2241<p>This instruction requires several arguments:</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2242
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2243<ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2244 <li>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]2245 The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2246 convention</a> the call should use. If none is specified, the call defaults
2247 to using C calling conventions.
2248 </li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]2249
2250 <li>The optional <a href="#paramattrs">Parameter Attributes</a> list for
2251 return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>',
2252 and '<tt>inreg</tt>' attributes are valid here.</li>
2253
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2254 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
2255 function value being invoked. In most cases, this is a direct function
2256 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
2257 an arbitrary pointer to function value.
2258 </li>
2259
2260 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
2261 function to be invoked. </li>
2262
2263 <li>'<tt>function args</tt>': argument list whose types match the function
2264 signature argument types. If the function signature indicates the function
2265 accepts a variable number of arguments, the extra arguments can be
2266 specified. </li>
2267
2268 <li>'<tt>normal label</tt>': the label reached when the called function
2269 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
2270
2271 <li>'<tt>exception label</tt>': the label reached when a callee returns with
2272 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
2273
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]2274 <li>The optional <a href="#fnattrs">function attributes</a> list. Only
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]2275 '<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
2276 '<tt>readnone</tt>' attributes are valid here.</li>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2277</ol>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2278
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2279<h5>Semantics:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2280
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2281<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2282href="#i_call">call</a></tt>' instruction in most regards. The primary
2283difference is that it establishes an association with a label, which is used by
2284the runtime library to unwind the stack.</p>
2285
2286<p>This instruction is used in languages with destructors to ensure that proper
2287cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
2288exception. Additionally, this is important for implementation of
2289'<tt>catch</tt>' clauses in high-level languages that support them.</p>
2290
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2291<h5>Example:</h5>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]2292<pre>
Nick Lewyckyd703f652008-03-16 07:18:12 +0000[diff] [blame]2293 %retval = invoke i32 @Test(i32 15) to label %Continue
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2294 unwind label %TestCleanup <i>; {i32}:retval set</i>
Nick Lewyckyd703f652008-03-16 07:18:12 +0000[diff] [blame]2295 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Testfnptr(i32 15) to label %Continue
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2296 unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2297</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2298</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2299
2300
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2301<!-- _______________________________________________________________________ -->
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2302
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2303<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
2304Instruction</a> </div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2305
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2306<div class="doc_text">
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2307
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2308<h5>Syntax:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2309<pre>
2310 unwind
2311</pre>
2312
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2313<h5>Overview:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2314
2315<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
2316at the first callee in the dynamic call stack which used an <a
2317href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
2318primarily used to implement exception handling.</p>
2319
Chris Lattner27f71f22003-09-03 00:41:47 +0000[diff] [blame]2320<h5>Semantics:</h5>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2321
Chris Lattner72ed2002008-04-19 21:01:16 +0000[diff] [blame]2322<p>The '<tt>unwind</tt>' instruction causes execution of the current function to
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2323immediately halt. The dynamic call stack is then searched for the first <a
2324href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
2325execution continues at the "exceptional" destination block specified by the
2326<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
2327dynamic call chain, undefined behavior results.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2328</div>
Chris Lattner35eca582004-10-16 18:04:13 +0000[diff] [blame]2329
2330<!-- _______________________________________________________________________ -->
2331
2332<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
2333Instruction</a> </div>
2334
2335<div class="doc_text">
2336
2337<h5>Syntax:</h5>
2338<pre>
2339 unreachable
2340</pre>
2341
2342<h5>Overview:</h5>
2343
2344<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
2345instruction is used to inform the optimizer that a particular portion of the
2346code is not reachable. This can be used to indicate that the code after a
2347no-return function cannot be reached, and other facts.</p>
2348
2349<h5>Semantics:</h5>
2350
2351<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
2352</div>
2353
2354
2355
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2356<!-- ======================================================================= -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2357<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2358<div class="doc_text">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2359<p>Binary operators are used to do most of the computation in a
Chris Lattner5a158142008-04-01 18:47:32 +0000[diff] [blame]2360program. They require two operands of the same type, execute an operation on them, and
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]2361produce a single value. The operands might represent
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]2362multiple data, as is the case with the <a href="#t_vector">vector</a> data type.
Chris Lattner5a158142008-04-01 18:47:32 +0000[diff] [blame]2363The result value has the same type as its operands.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2364<p>There are several different binary operators:</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2365</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2366<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2367<div class="doc_subsubsection">
2368 <a name="i_add">'<tt>add</tt>' Instruction</a>
2369</div>
2370
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2371<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2372
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2373<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2374
2375<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2376 &lt;result&gt; = add &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2377</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2378
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2379<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2380
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2381<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2382
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2383<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2384
2385<p>The two arguments to the '<tt>add</tt>' instruction must be <a
2386 href="#t_integer">integer</a>, <a href="#t_floating">floating point</a>, or
2387 <a href="#t_vector">vector</a> values. Both arguments must have identical
2388 types.</p>
2389
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2390<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2391
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2392<p>The value produced is the integer or floating point sum of the two
2393operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2394
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2395<p>If an integer sum has unsigned overflow, the result returned is the
2396mathematical result modulo 2<sup>n</sup>, where n is the bit width of
2397the result.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2398
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2399<p>Because LLVM integers use a two's complement representation, this
2400instruction is appropriate for both signed and unsigned integers.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2401
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2402<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2403
2404<pre>
2405 &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2406</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2407</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2408<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2409<div class="doc_subsubsection">
2410 <a name="i_sub">'<tt>sub</tt>' Instruction</a>
2411</div>
2412
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2413<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2414
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2415<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2416
2417<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2418 &lt;result&gt; = sub &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2419</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2420
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2421<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2422
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2423<p>The '<tt>sub</tt>' instruction returns the difference of its two
2424operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2425
2426<p>Note that the '<tt>sub</tt>' instruction is used to represent the
2427'<tt>neg</tt>' instruction present in most other intermediate
2428representations.</p>
2429
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2430<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2431
2432<p>The two arguments to the '<tt>sub</tt>' instruction must be <a
2433 href="#t_integer">integer</a>, <a href="#t_floating">floating point</a>,
2434 or <a href="#t_vector">vector</a> values. Both arguments must have identical
2435 types.</p>
2436
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2437<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2438
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2439<p>The value produced is the integer or floating point difference of
2440the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2441
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2442<p>If an integer difference has unsigned overflow, the result returned is the
2443mathematical result modulo 2<sup>n</sup>, where n is the bit width of
2444the result.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2445
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2446<p>Because LLVM integers use a two's complement representation, this
2447instruction is appropriate for both signed and unsigned integers.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2448
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2449<h5>Example:</h5>
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]2450<pre>
2451 &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2452 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2453</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2454</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2455
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2456<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2457<div class="doc_subsubsection">
2458 <a name="i_mul">'<tt>mul</tt>' Instruction</a>
2459</div>
2460
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2461<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2462
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2463<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2464<pre> &lt;result&gt; = mul &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2465</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2466<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2467<p>The '<tt>mul</tt>' instruction returns the product of its two
2468operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2469
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2470<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2471
2472<p>The two arguments to the '<tt>mul</tt>' instruction must be <a
2473href="#t_integer">integer</a>, <a href="#t_floating">floating point</a>,
2474or <a href="#t_vector">vector</a> values. Both arguments must have identical
2475types.</p>
2476
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2477<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2478
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2479<p>The value produced is the integer or floating point product of the
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2480two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2481
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2482<p>If the result of an integer multiplication has unsigned overflow,
2483the result returned is the mathematical result modulo
24842<sup>n</sup>, where n is the bit width of the result.</p>
2485<p>Because LLVM integers use a two's complement representation, and the
2486result is the same width as the operands, this instruction returns the
2487correct result for both signed and unsigned integers. If a full product
2488(e.g. <tt>i32</tt>x<tt>i32</tt>-><tt>i64</tt>) is needed, the operands
2489should be sign-extended or zero-extended as appropriate to the
2490width of the full product.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2491<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2492<pre> &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2493</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2494</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2495
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2496<!-- _______________________________________________________________________ -->
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2497<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
2498</a></div>
2499<div class="doc_text">
2500<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2501<pre> &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]2502</pre>
2503<h5>Overview:</h5>
2504<p>The '<tt>udiv</tt>' instruction returns the quotient of its two
2505operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2506
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2507<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2508
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2509<p>The two arguments to the '<tt>udiv</tt>' instruction must be
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2510<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2511values. Both arguments must have identical types.</p>
2512
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2513<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2514
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2515<p>The value produced is the unsigned integer quotient of the two operands.</p>
2516<p>Note that unsigned integer division and signed integer division are distinct
2517operations; for signed integer division, use '<tt>sdiv</tt>'.</p>
2518<p>Division by zero leads to undefined behavior.</p>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2519<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2520<pre> &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2521</pre>
2522</div>
2523<!-- _______________________________________________________________________ -->
2524<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
2525</a> </div>
2526<div class="doc_text">
2527<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2528<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2529 &lt;result&gt; = sdiv &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2530</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2531
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2532<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2533
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2534<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two
2535operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2536
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2537<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2538
2539<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
2540<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2541values. Both arguments must have identical types.</p>
2542
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2543<h5>Semantics:</h5>
Chris Lattnera73afe02008-04-01 18:45:27 +0000[diff] [blame]2544<p>The value produced is the signed integer quotient of the two operands rounded towards zero.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2545<p>Note that signed integer division and unsigned integer division are distinct
2546operations; for unsigned integer division, use '<tt>udiv</tt>'.</p>
2547<p>Division by zero leads to undefined behavior. Overflow also leads to
2548undefined behavior; this is a rare case, but can occur, for example,
2549by doing a 32-bit division of -2147483648 by -1.</p>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2550<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2551<pre> &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2552</pre>
2553</div>
2554<!-- _______________________________________________________________________ -->
2555<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2556Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2557<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2558<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2559<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2560 &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]2561</pre>
2562<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2563
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2564<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2565operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2566
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2567<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2568
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2569<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2570<a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a>
2571of floating point values. Both arguments must have identical types.</p>
2572
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2573<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2574
Reid Spencer1628cec2006-10-26 06:15:43 +0000[diff] [blame]2575<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2576
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2577<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2578
2579<pre>
2580 &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]2581</pre>
2582</div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2583
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2584<!-- _______________________________________________________________________ -->
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2585<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
2586</div>
2587<div class="doc_text">
2588<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2589<pre> &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]2590</pre>
2591<h5>Overview:</h5>
2592<p>The '<tt>urem</tt>' instruction returns the remainder from the
2593unsigned division of its two arguments.</p>
2594<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2595<p>The two arguments to the '<tt>urem</tt>' instruction must be
2596<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2597values. Both arguments must have identical types.</p>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2598<h5>Semantics:</h5>
2599<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
Chris Lattnera73afe02008-04-01 18:45:27 +0000[diff] [blame]2600This instruction always performs an unsigned division to get the remainder.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2601<p>Note that unsigned integer remainder and signed integer remainder are
2602distinct operations; for signed integer remainder, use '<tt>srem</tt>'.</p>
2603<p>Taking the remainder of a division by zero leads to undefined behavior.</p>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2604<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2605<pre> &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2606</pre>
2607
2608</div>
2609<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2610<div class="doc_subsubsection">
2611 <a name="i_srem">'<tt>srem</tt>' Instruction</a>
2612</div>
2613
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2614<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2615
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2616<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2617
2618<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2619 &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]2620</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2621
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2622<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2623
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2624<p>The '<tt>srem</tt>' instruction returns the remainder from the
Dan Gohman80176312007-11-05 23:35:22 +0000[diff] [blame]2625signed division of its two operands. This instruction can also take
2626<a href="#t_vector">vector</a> versions of the values in which case
2627the elements must be integers.</p>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]2628
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2629<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2630
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2631<p>The two arguments to the '<tt>srem</tt>' instruction must be
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2632<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2633values. Both arguments must have identical types.</p>
2634
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2635<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2636
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2637<p>This instruction returns the <i>remainder</i> of a division (where the result
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2638has the same sign as the dividend, <tt>op1</tt>), not the <i>modulo</i>
2639operator (where the result has the same sign as the divisor, <tt>op2</tt>) of
Reid Spencerc9fdfc82007-03-24 22:23:39 +0000[diff] [blame]2640a value. For more information about the difference, see <a
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2641 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
Reid Spencerc9fdfc82007-03-24 22:23:39 +0000[diff] [blame]2642Math Forum</a>. For a table of how this is implemented in various languages,
Reid Spencer64f5c6c2007-03-24 22:40:44 +0000[diff] [blame]2643please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
Reid Spencerc9fdfc82007-03-24 22:23:39 +0000[diff] [blame]2644Wikipedia: modulo operation</a>.</p>
Chris Lattner5ec89832008-01-28 00:36:27 +0000[diff] [blame]2645<p>Note that signed integer remainder and unsigned integer remainder are
2646distinct operations; for unsigned integer remainder, use '<tt>urem</tt>'.</p>
2647<p>Taking the remainder of a division by zero leads to undefined behavior.
2648Overflow also leads to undefined behavior; this is a rare case, but can occur,
2649for example, by taking the remainder of a 32-bit division of -2147483648 by -1.
2650(The remainder doesn't actually overflow, but this rule lets srem be
2651implemented using instructions that return both the result of the division
2652and the remainder.)</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2653<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2654<pre> &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2655</pre>
2656
2657</div>
2658<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2659<div class="doc_subsubsection">
2660 <a name="i_frem">'<tt>frem</tt>' Instruction</a> </div>
2661
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2662<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2663
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2664<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2665<pre> &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]2666</pre>
2667<h5>Overview:</h5>
2668<p>The '<tt>frem</tt>' instruction returns the remainder from the
2669division of its two operands.</p>
2670<h5>Arguments:</h5>
2671<p>The two arguments to the '<tt>frem</tt>' instruction must be
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2672<a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a>
2673of floating point values. Both arguments must have identical types.</p>
2674
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2675<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2676
Chris Lattnera73afe02008-04-01 18:45:27 +0000[diff] [blame]2677<p>This instruction returns the <i>remainder</i> of a division.
2678The remainder has the same sign as the dividend.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2679
Reid Spencer0a783f72006-11-02 01:53:59 +0000[diff] [blame]2680<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2681
2682<pre>
2683 &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]2684</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2685</div>
Robert Bocchino7b81c752006-02-17 21:18:08 +0000[diff] [blame]2686
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]2687<!-- ======================================================================= -->
2688<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
2689Operations</a> </div>
2690<div class="doc_text">
2691<p>Bitwise binary operators are used to do various forms of
2692bit-twiddling in a program. They are generally very efficient
2693instructions and can commonly be strength reduced from other
Chris Lattnera73afe02008-04-01 18:45:27 +0000[diff] [blame]2694instructions. They require two operands of the same type, execute an operation on them,
2695and produce a single value. The resulting value is the same type as its operands.</p>
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]2696</div>
2697
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2698<!-- _______________________________________________________________________ -->
2699<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
2700Instruction</a> </div>
2701<div class="doc_text">
2702<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2703<pre> &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]2704</pre>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2705
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2706<h5>Overview:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2707
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2708<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
2709the left a specified number of bits.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2710
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2711<h5>Arguments:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2712
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2713<p>Both arguments to the '<tt>shl</tt>' instruction must be the same <a
Nate Begeman5bc1ea02008-07-29 15:49:41 +0000[diff] [blame]2714 href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2715type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2716
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2717<h5>Semantics:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2718
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2719<p>The value produced is <tt>op1</tt> * 2<sup><tt>op2</tt></sup> mod 2<sup>n</sup>,
2720where n is the width of the result. If <tt>op2</tt> is (statically or dynamically) negative or
Mon P Wang01f8d092008-12-10 08:55:09 +0000[diff] [blame]2721equal to or larger than the number of bits in <tt>op1</tt>, the result is undefined.
2722If the arguments are vectors, each vector element of <tt>op1</tt> is shifted by the
2723corresponding shift amount in <tt>op2</tt>.</p>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2724
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2725<h5>Example:</h5><pre>
2726 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
2727 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
2728 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2729 &lt;result&gt; = shl i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]2730 &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]2731</pre>
2732</div>
2733<!-- _______________________________________________________________________ -->
2734<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
2735Instruction</a> </div>
2736<div class="doc_text">
2737<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2738<pre> &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]2739</pre>
2740
2741<h5>Overview:</h5>
2742<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2743operand shifted to the right a specified number of bits with zero fill.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2744
2745<h5>Arguments:</h5>
2746<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
Nate Begeman5bc1ea02008-07-29 15:49:41 +0000[diff] [blame]2747<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2748type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2749
2750<h5>Semantics:</h5>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2751
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2752<p>This instruction always performs a logical shift right operation. The most
2753significant bits of the result will be filled with zero bits after the
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2754shift. If <tt>op2</tt> is (statically or dynamically) equal to or larger than
Mon P Wang01f8d092008-12-10 08:55:09 +0000[diff] [blame]2755the number of bits in <tt>op1</tt>, the result is undefined. If the arguments are
2756vectors, each vector element of <tt>op1</tt> is shifted by the corresponding shift
2757amount in <tt>op2</tt>.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2758
2759<h5>Example:</h5>
2760<pre>
2761 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
2762 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
2763 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
2764 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2765 &lt;result&gt; = lshr i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]2766 &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]2767</pre>
2768</div>
2769
Reid Spencer8e11bf82007-02-02 13:57:07 +0000[diff] [blame]2770<!-- _______________________________________________________________________ -->
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2771<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
2772Instruction</a> </div>
2773<div class="doc_text">
2774
2775<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2776<pre> &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]2777</pre>
2778
2779<h5>Overview:</h5>
2780<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2781operand shifted to the right a specified number of bits with sign extension.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2782
2783<h5>Arguments:</h5>
2784<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
Nate Begeman5bc1ea02008-07-29 15:49:41 +0000[diff] [blame]2785<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2786type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2787
2788<h5>Semantics:</h5>
2789<p>This instruction always performs an arithmetic shift right operation,
2790The most significant bits of the result will be filled with the sign bit
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2791of <tt>op1</tt>. If <tt>op2</tt> is (statically or dynamically) equal to or
Mon P Wang01f8d092008-12-10 08:55:09 +0000[diff] [blame]2792larger than the number of bits in <tt>op1</tt>, the result is undefined. If the
2793arguments are vectors, each vector element of <tt>op1</tt> is shifted by the
2794corresponding shift amount in <tt>op2</tt>.</p>
Reid Spencer569f2fa2007-01-31 21:39:12 +0000[diff] [blame]2795
2796<h5>Example:</h5>
2797<pre>
2798 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
2799 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
2800 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
2801 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattner6ccc2d52007-10-03 21:01:14 +0000[diff] [blame]2802 &lt;result&gt; = ashr i32 1, 32 <i>; undefined</i>
Mon P Wange9f10152008-12-09 05:46:39 +0000[diff] [blame]2803 &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]2804</pre>
2805</div>
2806
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2807<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2808<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
2809Instruction</a> </div>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2810
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2811<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2812
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2813<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2814
2815<pre>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2816 &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]2817</pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2818
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2819<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2820
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2821<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
2822its two operands.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2823
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2824<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2825
2826<p>The two arguments to the '<tt>and</tt>' instruction must be
2827<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2828values. Both arguments must have identical types.</p>
2829
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2830<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2831<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2832<p> </p>
Bill Wendlingc7e4c4d2008-09-07 10:29:20 +0000[diff] [blame]2833<div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2834<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2835 <tbody>
2836 <tr>
2837 <td>In0</td>
2838 <td>In1</td>
2839 <td>Out</td>
2840 </tr>
2841 <tr>
2842 <td>0</td>
2843 <td>0</td>
2844 <td>0</td>
2845 </tr>
2846 <tr>
2847 <td>0</td>
2848 <td>1</td>
2849 <td>0</td>
2850 </tr>
2851 <tr>
2852 <td>1</td>
2853 <td>0</td>
2854 <td>0</td>
2855 </tr>
2856 <tr>
2857 <td>1</td>
2858 <td>1</td>
2859 <td>1</td>
2860 </tr>
2861 </tbody>
2862</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2863</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2864<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2865<pre>
2866 &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]2867 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
2868 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2869</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2870</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2871<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2872<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2873<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2874<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2875<pre> &lt;result&gt; = or &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2876</pre>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2877<h5>Overview:</h5>
2878<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
2879or of its two operands.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2880<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2881
2882<p>The two arguments to the '<tt>or</tt>' instruction must be
2883<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2884values. Both arguments must have identical types.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2885<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2886<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2887<p> </p>
Bill Wendlingc7e4c4d2008-09-07 10:29:20 +0000[diff] [blame]2888<div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2889<table border="1" cellspacing="0" cellpadding="4">
2890 <tbody>
2891 <tr>
2892 <td>In0</td>
2893 <td>In1</td>
2894 <td>Out</td>
2895 </tr>
2896 <tr>
2897 <td>0</td>
2898 <td>0</td>
2899 <td>0</td>
2900 </tr>
2901 <tr>
2902 <td>0</td>
2903 <td>1</td>
2904 <td>1</td>
2905 </tr>
2906 <tr>
2907 <td>1</td>
2908 <td>0</td>
2909 <td>1</td>
2910 </tr>
2911 <tr>
2912 <td>1</td>
2913 <td>1</td>
2914 <td>1</td>
2915 </tr>
2916 </tbody>
2917</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2918</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2919<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2920<pre> &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
2921 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
2922 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2923</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2924</div>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2925<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2926<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
2927Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2928<div class="doc_text">
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2929<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]2930<pre> &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]2931</pre>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2932<h5>Overview:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2933<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
2934or of its two operands. The <tt>xor</tt> is used to implement the
2935"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2936<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2937<p>The two arguments to the '<tt>xor</tt>' instruction must be
2938<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
2939values. Both arguments must have identical types.</p>
2940
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2941<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]2942
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2943<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2944<p> </p>
Bill Wendlingc7e4c4d2008-09-07 10:29:20 +0000[diff] [blame]2945<div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2946<table border="1" cellspacing="0" cellpadding="4">
2947 <tbody>
2948 <tr>
2949 <td>In0</td>
2950 <td>In1</td>
2951 <td>Out</td>
2952 </tr>
2953 <tr>
2954 <td>0</td>
2955 <td>0</td>
2956 <td>0</td>
2957 </tr>
2958 <tr>
2959 <td>0</td>
2960 <td>1</td>
2961 <td>1</td>
2962 </tr>
2963 <tr>
2964 <td>1</td>
2965 <td>0</td>
2966 <td>1</td>
2967 </tr>
2968 <tr>
2969 <td>1</td>
2970 <td>1</td>
2971 <td>0</td>
2972 </tr>
2973 </tbody>
2974</table>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]2975</div>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]2976<p> </p>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2977<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]2978<pre> &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
2979 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
2980 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
2981 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2982</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]2983</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2984
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]2985<!-- ======================================================================= -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]2986<div class="doc_subsection">
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2987 <a name="vectorops">Vector Operations</a>
2988</div>
2989
2990<div class="doc_text">
2991
2992<p>LLVM supports several instructions to represent vector operations in a
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]2993target-independent manner. These instructions cover the element-access and
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]2994vector-specific operations needed to process vectors effectively. While LLVM
2995does directly support these vector operations, many sophisticated algorithms
2996will want to use target-specific intrinsics to take full advantage of a specific
2997target.</p>
2998
2999</div>
3000
3001<!-- _______________________________________________________________________ -->
3002<div class="doc_subsubsection">
3003 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
3004</div>
3005
3006<div class="doc_text">
3007
3008<h5>Syntax:</h5>
3009
3010<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3011 &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]3012</pre>
3013
3014<h5>Overview:</h5>
3015
3016<p>
3017The '<tt>extractelement</tt>' instruction extracts a single scalar
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]3018element from a vector at a specified index.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3019</p>
3020
3021
3022<h5>Arguments:</h5>
3023
3024<p>
3025The first operand of an '<tt>extractelement</tt>' instruction is a
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]3026value of <a href="#t_vector">vector</a> type. The second operand is
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3027an index indicating the position from which to extract the element.
3028The index may be a variable.</p>
3029
3030<h5>Semantics:</h5>
3031
3032<p>
3033The result is a scalar of the same type as the element type of
3034<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
3035<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
3036results are undefined.
3037</p>
3038
3039<h5>Example:</h5>
3040
3041<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3042 %result = extractelement &lt;4 x i32&gt; %vec, i32 0 <i>; yields i32</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3043</pre>
3044</div>
3045
3046
3047<!-- _______________________________________________________________________ -->
3048<div class="doc_subsubsection">
3049 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
3050</div>
3051
3052<div class="doc_text">
3053
3054<h5>Syntax:</h5>
3055
3056<pre>
Dan Gohmanf3480b92008-05-12 23:38:42 +0000[diff] [blame]3057 &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]3058</pre>
3059
3060<h5>Overview:</h5>
3061
3062<p>
3063The '<tt>insertelement</tt>' instruction inserts a scalar
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]3064element into a vector at a specified index.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3065</p>
3066
3067
3068<h5>Arguments:</h5>
3069
3070<p>
3071The first operand of an '<tt>insertelement</tt>' instruction is a
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]3072value of <a href="#t_vector">vector</a> type. The second operand is a
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3073scalar value whose type must equal the element type of the first
3074operand. The third operand is an index indicating the position at
3075which to insert the value. The index may be a variable.</p>
3076
3077<h5>Semantics:</h5>
3078
3079<p>
Reid Spencer485bad12007-02-15 03:07:05 +0000[diff] [blame]3080The result is a vector of the same type as <tt>val</tt>. Its
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3081element values are those of <tt>val</tt> except at position
3082<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
3083exceeds the length of <tt>val</tt>, the results are undefined.
3084</p>
3085
3086<h5>Example:</h5>
3087
3088<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3089 %result = insertelement &lt;4 x i32&gt; %vec, i32 1, i32 0 <i>; yields &lt;4 x i32&gt;</i>
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3090</pre>
3091</div>
3092
3093<!-- _______________________________________________________________________ -->
3094<div class="doc_subsubsection">
3095 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
3096</div>
3097
3098<div class="doc_text">
3099
3100<h5>Syntax:</h5>
3101
3102<pre>
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]3103 &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]3104</pre>
3105
3106<h5>Overview:</h5>
3107
3108<p>
3109The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]3110from two input vectors, returning a vector with the same element type as
3111the input and length that is the same as the shuffle mask.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3112</p>
3113
3114<h5>Arguments:</h5>
3115
3116<p>
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]3117The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
3118with types that match each other. The third argument is a shuffle mask whose
3119element type is always 'i32'. The result of the instruction is a vector whose
3120length is the same as the shuffle mask and whose element type is the same as
3121the element type of the first two operands.
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3122</p>
3123
3124<p>
3125The shuffle mask operand is required to be a constant vector with either
3126constant integer or undef values.
3127</p>
3128
3129<h5>Semantics:</h5>
3130
3131<p>
3132The elements of the two input vectors are numbered from left to right across
3133both of the vectors. The shuffle mask operand specifies, for each element of
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]3134the result vector, which element of the two input vectors the result element
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3135gets. The element selector may be undef (meaning "don't care") and the second
3136operand may be undef if performing a shuffle from only one vector.
3137</p>
3138
3139<h5>Example:</h5>
3140
3141<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3142 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3143 &lt;4 x i32&gt; &lt;i32 0, i32 4, i32 1, i32 5&gt; <i>; yields &lt;4 x i32&gt;</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3144 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; undef,
3145 &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.
Mon P Wangaeb06d22008-11-10 04:46:22 +0000[diff] [blame]3146 %result = shufflevector &lt;8 x i32&gt; %v1, &lt;8 x i32&gt; undef,
3147 &lt;4 x i32&gt; &lt;i32 0, i32 1, i32 2, i32 3&gt; <i>; yields &lt;4 x i32&gt;</i>
3148 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
3149 &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]3150</pre>
3151</div>
3152
Tanya Lattner09474292006-04-14 19:24:33 +0000[diff] [blame]3153
Chris Lattner3df241e2006-04-08 23:07:04 +0000[diff] [blame]3154<!-- ======================================================================= -->
3155<div class="doc_subsection">
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3156 <a name="aggregateops">Aggregate Operations</a>
3157</div>
3158
3159<div class="doc_text">
3160
3161<p>LLVM supports several instructions for working with aggregate values.
3162</p>
3163
3164</div>
3165
3166<!-- _______________________________________________________________________ -->
3167<div class="doc_subsubsection">
3168 <a name="i_extractvalue">'<tt>extractvalue</tt>' Instruction</a>
3169</div>
3170
3171<div class="doc_text">
3172
3173<h5>Syntax:</h5>
3174
3175<pre>
3176 &lt;result&gt; = extractvalue &lt;aggregate type&gt; &lt;val&gt;, &lt;idx&gt;{, &lt;idx&gt;}*
3177</pre>
3178
3179<h5>Overview:</h5>
3180
3181<p>
Dan Gohmanc3dac5c2008-05-13 18:16:06 +0000[diff] [blame]3182The '<tt>extractvalue</tt>' instruction extracts the value of a struct field
3183or array element from an aggregate value.
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3184</p>
3185
3186
3187<h5>Arguments:</h5>
3188
3189<p>
3190The first operand of an '<tt>extractvalue</tt>' instruction is a
3191value of <a href="#t_struct">struct</a> or <a href="#t_array">array</a>
Dan Gohmanc3dac5c2008-05-13 18:16:06 +0000[diff] [blame]3192type. The operands are constant indices to specify which value to extract
Dan Gohman81a0c0b2008-05-31 00:58:22 +0000[diff] [blame]3193in a similar manner as indices in a
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3194'<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.
3195</p>
3196
3197<h5>Semantics:</h5>
3198
3199<p>
3200The result is the value at the position in the aggregate specified by
3201the index operands.
3202</p>
3203
3204<h5>Example:</h5>
3205
3206<pre>
Dan Gohman81a0c0b2008-05-31 00:58:22 +0000[diff] [blame]3207 %result = extractvalue {i32, float} %agg, 0 <i>; yields i32</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3208</pre>
3209</div>
3210
3211
3212<!-- _______________________________________________________________________ -->
3213<div class="doc_subsubsection">
3214 <a name="i_insertvalue">'<tt>insertvalue</tt>' Instruction</a>
3215</div>
3216
3217<div class="doc_text">
3218
3219<h5>Syntax:</h5>
3220
3221<pre>
Dan Gohman81a0c0b2008-05-31 00:58:22 +0000[diff] [blame]3222 &lt;result&gt; = insertvalue &lt;aggregate type&gt; &lt;val&gt;, &lt;ty&gt; &lt;val&gt;, &lt;idx&gt; <i>; yields &lt;n x &lt;ty&gt;&gt;</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3223</pre>
3224
3225<h5>Overview:</h5>
3226
3227<p>
3228The '<tt>insertvalue</tt>' instruction inserts a value
Dan Gohmanc3dac5c2008-05-13 18:16:06 +0000[diff] [blame]3229into a struct field or array element in an aggregate.
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3230</p>
3231
3232
3233<h5>Arguments:</h5>
3234
3235<p>
3236The first operand of an '<tt>insertvalue</tt>' instruction is a
3237value of <a href="#t_struct">struct</a> or <a href="#t_array">array</a> type.
3238The second operand is a first-class value to insert.
Dan Gohmanc4b49eb2008-05-23 21:53:15 +0000[diff] [blame]3239The following operands are constant indices
Dan Gohman81a0c0b2008-05-31 00:58:22 +0000[diff] [blame]3240indicating the position at which to insert the value in a similar manner as
Dan Gohmanc3dac5c2008-05-13 18:16:06 +0000[diff] [blame]3241indices in a
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3242'<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.
3243The value to insert must have the same type as the value identified
Dan Gohmanc3dac5c2008-05-13 18:16:06 +0000[diff] [blame]3244by the indices.
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]3245</p>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3246
3247<h5>Semantics:</h5>
3248
3249<p>
3250The result is an aggregate of the same type as <tt>val</tt>. Its
3251value is that of <tt>val</tt> except that the value at the position
Dan Gohmanc3dac5c2008-05-13 18:16:06 +0000[diff] [blame]3252specified by the indices is that of <tt>elt</tt>.
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3253</p>
3254
3255<h5>Example:</h5>
3256
3257<pre>
Dan Gohman52bb2db2008-06-23 15:26:37 +0000[diff] [blame]3258 %result = insertvalue {i32, float} %agg, i32 1, 0 <i>; yields {i32, float}</i>
Dan Gohmana334d5f2008-05-12 23:51:09 +0000[diff] [blame]3259</pre>
3260</div>
3261
3262
3263<!-- ======================================================================= -->
3264<div class="doc_subsection">
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]3265 <a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3266</div>
3267
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3268<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3269
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3270<p>A key design point of an SSA-based representation is how it
3271represents memory. In LLVM, no memory locations are in SSA form, which
3272makes things very simple. This section describes how to read, write,
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]3273allocate, and free memory in LLVM.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3274
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3275</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3276
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3277<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3278<div class="doc_subsubsection">
3279 <a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
3280</div>
3281
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3282<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3283
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3284<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3285
3286<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3287 &lt;result&gt; = malloc &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3288</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3289
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3290<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3291
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3292<p>The '<tt>malloc</tt>' instruction allocates memory from the system
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]3293heap and returns a pointer to it. The object is always allocated in the generic
3294address space (address space zero).</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3295
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3296<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3297
3298<p>The '<tt>malloc</tt>' instruction allocates
3299<tt>sizeof(&lt;type&gt;)*NumElements</tt>
John Criswell6e4ca612004-02-24 16:13:56 +0000[diff] [blame]3300bytes of memory from the operating system and returns a pointer of the
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3301appropriate type to the program. If "NumElements" is specified, it is the
Gabor Greif1acd2ee2008-02-09 22:24:34 +0000[diff] [blame]3302number of elements allocated, otherwise "NumElements" is defaulted to be one.
Chris Lattner4316dec2008-04-02 00:38:26 +0000[diff] [blame]3303If a constant alignment is specified, the value result of the allocation is guaranteed to
Gabor Greif1acd2ee2008-02-09 22:24:34 +0000[diff] [blame]3304be aligned to at least that boundary. If not specified, or if zero, the target can
3305choose to align the allocation on any convenient boundary.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3306
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3307<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3308
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3309<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3310
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3311<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
Nick Lewycky02ff3082008-11-24 03:41:24 +0000[diff] [blame]3312a pointer is returned. The result of a zero byte allocation is undefined. The
Chris Lattner72ed2002008-04-19 21:01:16 +0000[diff] [blame]3313result is null if there is insufficient memory available.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3314
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3315<h5>Example:</h5>
3316
3317<pre>
Dan Gohman81e21672009-01-04 23:49:44 +0000[diff] [blame]3318 %array = malloc [4 x i8] <i>; yields {[%4 x i8]*}:array</i>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3319
Bill Wendlingaac388b2007-05-29 09:42:13 +0000[diff] [blame]3320 %size = <a href="#i_add">add</a> i32 2, 2 <i>; yields {i32}:size = i32 4</i>
3321 %array1 = malloc i8, i32 4 <i>; yields {i8*}:array1</i>
3322 %array2 = malloc [12 x i8], i32 %size <i>; yields {[12 x i8]*}:array2</i>
3323 %array3 = malloc i32, i32 4, align 1024 <i>; yields {i32*}:array3</i>
3324 %array4 = malloc i32, align 1024 <i>; yields {i32*}:array4</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3325</pre>
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]3326
3327<p>Note that the code generator does not yet respect the
3328 alignment value.</p>
3329
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3330</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3331
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3332<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3333<div class="doc_subsubsection">
3334 <a name="i_free">'<tt>free</tt>' Instruction</a>
3335</div>
3336
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3337<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3338
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3339<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3340
3341<pre>
Dan Gohman81e21672009-01-04 23:49:44 +0000[diff] [blame]3342 free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3343</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3344
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3345<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3346
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3347<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]3348memory heap to be reallocated in the future.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3349
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3350<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3351
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3352<p>'<tt>value</tt>' shall be a pointer value that points to a value
3353that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
3354instruction.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3355
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3356<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3357
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]3358<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattnere0db56d2008-04-19 22:41:32 +0000[diff] [blame]3359after this instruction executes. If the pointer is null, the operation
3360is a noop.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3361
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3362<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3363
3364<pre>
Dan Gohman81e21672009-01-04 23:49:44 +0000[diff] [blame]3365 %array = <a href="#i_malloc">malloc</a> [4 x i8] <i>; yields {[4 x i8]*}:array</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3366 free [4 x i8]* %array
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3367</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3368</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3369
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3370<!-- _______________________________________________________________________ -->
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3371<div class="doc_subsubsection">
3372 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
3373</div>
3374
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3375<div class="doc_text">
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3376
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3377<h5>Syntax:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3378
3379<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3380 &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]3381</pre>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3382
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3383<h5>Overview:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3384
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3385<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
3386currently executing function, to be automatically released when this function
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]3387returns to its caller. The object is always allocated in the generic address
3388space (address space zero).</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3389
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3390<h5>Arguments:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3391
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]3392<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3393bytes of memory on the runtime stack, returning a pointer of the
Gabor Greif1acd2ee2008-02-09 22:24:34 +0000[diff] [blame]3394appropriate type to the program. If "NumElements" is specified, it is the
3395number of elements allocated, otherwise "NumElements" is defaulted to be one.
Chris Lattner4316dec2008-04-02 00:38:26 +0000[diff] [blame]3396If a constant alignment is specified, the value result of the allocation is guaranteed
Gabor Greif1acd2ee2008-02-09 22:24:34 +0000[diff] [blame]3397to be aligned to at least that boundary. If not specified, or if zero, the target
3398can choose to align the allocation on any convenient boundary.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3399
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3400<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3401
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3402<h5>Semantics:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3403
Chris Lattner72ed2002008-04-19 21:01:16 +0000[diff] [blame]3404<p>Memory is allocated; a pointer is returned. The operation is undefiend if
3405there is insufficient stack space for the allocation. '<tt>alloca</tt>'d
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3406memory is automatically released when the function returns. The '<tt>alloca</tt>'
3407instruction is commonly used to represent automatic variables that must
3408have an address available. When the function returns (either with the <tt><a
John Criswelldae2e932005-05-12 16:55:34 +0000[diff] [blame]3409 href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
Chris Lattner4316dec2008-04-02 00:38:26 +0000[diff] [blame]3410instructions), the memory is reclaimed. Allocating zero bytes
3411is legal, but the result is undefined.</p>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3412
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3413<h5>Example:</h5>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3414
3415<pre>
Dan Gohman81e21672009-01-04 23:49:44 +0000[diff] [blame]3416 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
3417 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
3418 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
3419 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3420</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3421</div>
Chris Lattner2cbdc452005-11-06 08:02:57 +0000[diff] [blame]3422
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3423<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3424<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
3425Instruction</a> </div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3426<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3427<h5>Syntax:</h5>
Christopher Lamb2330e4d2007-04-21 08:16:25 +0000[diff] [blame]3428<pre> &lt;result&gt; = load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;]<br> &lt;result&gt; = volatile load &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;]<br></pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3429<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3430<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3431<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3432<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]3433address from which to load. The pointer must point to a <a
Chris Lattnere53e5082004-06-03 22:57:15 +0000[diff] [blame]3434 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell0ec250c2005-10-24 16:17:18 +0000[diff] [blame]3435marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3436the number or order of execution of this <tt>load</tt> with other
3437volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
3438instructions. </p>
Chris Lattnera31d1d72008-01-06 21:04:43 +0000[diff] [blame]3439<p>
Chris Lattner4316dec2008-04-02 00:38:26 +0000[diff] [blame]3440The optional constant "align" argument specifies the alignment of the operation
Chris Lattnera31d1d72008-01-06 21:04:43 +0000[diff] [blame]3441(that is, the alignment of the memory address). A value of 0 or an
3442omitted "align" argument means that the operation has the preferential
3443alignment for the target. It is the responsibility of the code emitter
3444to ensure that the alignment information is correct. Overestimating
3445the alignment results in an undefined behavior. Underestimating the
3446alignment may produce less efficient code. An alignment of 1 is always
3447safe.
3448</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3449<h5>Semantics:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3450<p>The location of memory pointed to is loaded.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3451<h5>Examples:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3452<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3453 <a
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3454 href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
3455 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3456</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3457</div>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3458<!-- _______________________________________________________________________ -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3459<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
3460Instruction</a> </div>
Reid Spencer035ab572006-11-09 21:18:01 +0000[diff] [blame]3461<div class="doc_text">
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3462<h5>Syntax:</h5>
Christopher Lamb2330e4d2007-04-21 08:16:25 +0000[diff] [blame]3463<pre> store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
3464 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]3465</pre>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3466<h5>Overview:</h5>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3467<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3468<h5>Arguments:</h5>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3469<p>There are two arguments to the '<tt>store</tt>' instruction: a value
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3470to store and an address at which to store it. The type of the '<tt>&lt;pointer&gt;</tt>'
Chris Lattner4316dec2008-04-02 00:38:26 +0000[diff] [blame]3471operand must be a pointer to the <a href="#t_firstclass">first class</a> type
3472of the '<tt>&lt;value&gt;</tt>'
John Criswellc1f786c2005-05-13 22:25:59 +0000[diff] [blame]3473operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3474optimizer is not allowed to modify the number or order of execution of
3475this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
3476 href="#i_store">store</a></tt> instructions.</p>
Chris Lattnera31d1d72008-01-06 21:04:43 +0000[diff] [blame]3477<p>
Chris Lattner4316dec2008-04-02 00:38:26 +0000[diff] [blame]3478The optional constant "align" argument specifies the alignment of the operation
Chris Lattnera31d1d72008-01-06 21:04:43 +0000[diff] [blame]3479(that is, the alignment of the memory address). A value of 0 or an
3480omitted "align" argument means that the operation has the preferential
3481alignment for the target. It is the responsibility of the code emitter
3482to ensure that the alignment information is correct. Overestimating
3483the alignment results in an undefined behavior. Underestimating the
3484alignment may produce less efficient code. An alignment of 1 is always
3485safe.
3486</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3487<h5>Semantics:</h5>
3488<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
3489at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3490<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3491<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling8c6c72d2007-10-22 05:10:05 +0000[diff] [blame]3492 store i32 3, i32* %ptr <i>; yields {void}</i>
3493 %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]3494</pre>
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]3495</div>
3496
Chris Lattner2b7d3202002-05-06 03:03:22 +0000[diff] [blame]3497<!-- _______________________________________________________________________ -->
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3498<div class="doc_subsubsection">
3499 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
3500</div>
3501
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3502<div class="doc_text">
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3503<h5>Syntax:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3504<pre>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]3505 &lt;result&gt; = getelementptr &lt;pty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3506</pre>
3507
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3508<h5>Overview:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3509
3510<p>
3511The '<tt>getelementptr</tt>' instruction is used to get the address of a
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]3512subelement of an aggregate data structure. It performs address calculation only
3513and does not access memory.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3514
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3515<h5>Arguments:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3516
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]3517<p>The first argument is always a pointer, and forms the basis of the
3518calculation. The remaining arguments are indices, that indicate which of the
3519elements of the aggregate object are indexed. The interpretation of each index
3520is dependent on the type being indexed into. The first index always indexes the
3521pointer value given as the first argument, the second index indexes a value of
3522the type pointed to (not necessarily the value directly pointed to, since the
3523first index can be non-zero), etc. The first type indexed into must be a pointer
3524value, subsequent types can be arrays, vectors and structs. Note that subsequent
3525types being indexed into can never be pointers, since that would require loading
3526the pointer before continuing calculation.</p>
3527
3528<p>The type of each index argument depends on the type it is indexing into.
3529When indexing into a (packed) structure, only <tt>i32</tt> integer
3530<b>constants</b> are allowed. When indexing into an array, pointer or vector,
3531only integers of 32 or 64 bits are allowed (also non-constants). 32-bit values
3532will be sign extended to 64-bits if required.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3533
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3534<p>For example, let's consider a C code fragment and how it gets
3535compiled to LLVM:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3536
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3537<div class="doc_code">
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3538<pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3539struct RT {
3540 char A;
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]3541 int B[10][20];
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3542 char C;
3543};
3544struct ST {
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]3545 int X;
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3546 double Y;
3547 struct RT Z;
3548};
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3549
Chris Lattnercabc8462007-05-29 15:43:56 +0000[diff] [blame]3550int *foo(struct ST *s) {
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3551 return &amp;s[1].Z.B[5][13];
3552}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3553</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3554</div>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3555
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3556<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3557
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3558<div class="doc_code">
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3559<pre>
Chris Lattnere7886e42009-01-11 20:53:49 +0000[diff] [blame]3560%RT = <a href="#namedtypes">type</a> { i8 , [10 x [20 x i32]], i8 }
3561%ST = <a href="#namedtypes">type</a> { i32, double, %RT }
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3562
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3563define i32* %foo(%ST* %s) {
3564entry:
3565 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
3566 ret i32* %reg
3567}
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3568</pre>
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]3569</div>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3570
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3571<h5>Semantics:</h5>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3572
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3573<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3574type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3575}</tt>' type, a structure. The second index indexes into the third element of
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3576the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
3577i8 }</tt>' type, another structure. The third index indexes into the second
3578element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3579array. The two dimensions of the array are subscripted into, yielding an
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3580'<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
3581to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3582
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]3583<p>Note that it is perfectly legal to index partially through a
3584structure, returning a pointer to an inner element. Because of this,
3585the LLVM code for the given testcase is equivalent to:</p>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3586
3587<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3588 define i32* %foo(%ST* %s) {
3589 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
Jeff Cohen6f1cc772007-04-22 01:17:39 +0000[diff] [blame]3590 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
3591 %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]3592 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
3593 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
3594 ret i32* %t5
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3595 }
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3596</pre>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]3597
3598<p>Note that it is undefined to access an array out of bounds: array and
3599pointer indexes must always be within the defined bounds of the array type.
Chris Lattner05d67092008-04-24 05:59:56 +0000[diff] [blame]3600The one exception for this rule is zero length arrays. These arrays are
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]3601defined to be accessible as variable length arrays, which requires access
3602beyond the zero'th element.</p>
3603
Chris Lattner884a9702006-08-15 00:45:58 +0000[diff] [blame]3604<p>The getelementptr instruction is often confusing. For some more insight
3605into how it works, see <a href="GetElementPtr.html">the getelementptr
3606FAQ</a>.</p>
3607
Chris Lattner7faa8832002-04-14 06:13:44 +0000[diff] [blame]3608<h5>Example:</h5>
Chris Lattnere67a9512005-06-24 17:22:57 +0000[diff] [blame]3609
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3610<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3611 <i>; yields [12 x i8]*:aptr</i>
Matthijs Kooijmane49d0bc2008-10-13 13:44:15 +0000[diff] [blame]3612 %aptr = getelementptr {i32, [12 x i8]}* %saptr, i64 0, i32 1
3613 <i>; yields i8*:vptr</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]3614 %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]3615 <i>; yields i8*:eptr</i>
3616 %eptr = getelementptr [12 x i8]* %aptr, i64 0, i32 1
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3617</pre>
Chris Lattnerf74d5c72004-04-05 01:30:49 +0000[diff] [blame]3618</div>
Reid Spencer47ce1792006-11-09 21:15:49 +0000[diff] [blame]3619
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]3620<!-- ======================================================================= -->
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3621<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3622</div>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3623<div class="doc_text">
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]3624<p>The instructions in this category are the conversion instructions (casting)
3625which all take a single operand and a type. They perform various bit conversions
3626on the operand.</p>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]3627</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3628
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3629<!-- _______________________________________________________________________ -->
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3630<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3631 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
3632</div>
3633<div class="doc_text">
3634
3635<h5>Syntax:</h5>
3636<pre>
3637 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3638</pre>
3639
3640<h5>Overview:</h5>
3641<p>
3642The '<tt>trunc</tt>' instruction truncates its operand to the type <tt>ty2</tt>.
3643</p>
3644
3645<h5>Arguments:</h5>
3646<p>
3647The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
3648be an <a href="#t_integer">integer</a> type, and a type that specifies the size
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3649and type of the result, which must be an <a href="#t_integer">integer</a>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3650type. The bit size of <tt>value</tt> must be larger than the bit size of
3651<tt>ty2</tt>. Equal sized types are not allowed.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3652
3653<h5>Semantics:</h5>
3654<p>
3655The '<tt>trunc</tt>' instruction truncates the high order bits in <tt>value</tt>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3656and converts the remaining bits to <tt>ty2</tt>. Since the source size must be
3657larger than the destination size, <tt>trunc</tt> cannot be a <i>no-op cast</i>.
3658It will always truncate bits.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3659
3660<h5>Example:</h5>
3661<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3662 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3663 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
3664 %Y = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3665</pre>
3666</div>
3667
3668<!-- _______________________________________________________________________ -->
3669<div class="doc_subsubsection">
3670 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
3671</div>
3672<div class="doc_text">
3673
3674<h5>Syntax:</h5>
3675<pre>
3676 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3677</pre>
3678
3679<h5>Overview:</h5>
3680<p>The '<tt>zext</tt>' instruction zero extends its operand to type
3681<tt>ty2</tt>.</p>
3682
3683
3684<h5>Arguments:</h5>
3685<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3686<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3687also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3688<tt>value</tt> must be smaller than the bit size of the destination type,
3689<tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3690
3691<h5>Semantics:</h5>
3692<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
Chris Lattnerd1d25172007-05-24 19:13:27 +0000[diff] [blame]3693bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3694
Reid Spencerb5929522007-01-12 15:46:11 +0000[diff] [blame]3695<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]3696
3697<h5>Example:</h5>
3698<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3699 %X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3700 %Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3701</pre>
3702</div>
3703
3704<!-- _______________________________________________________________________ -->
3705<div class="doc_subsubsection">
3706 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
3707</div>
3708<div class="doc_text">
3709
3710<h5>Syntax:</h5>
3711<pre>
3712 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3713</pre>
3714
3715<h5>Overview:</h5>
3716<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
3717
3718<h5>Arguments:</h5>
3719<p>
3720The '<tt>sext</tt>' instruction takes a value to cast, which must be of
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]3721<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3722also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3723<tt>value</tt> must be smaller than the bit size of the destination type,
3724<tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3725
3726<h5>Semantics:</h5>
3727<p>
3728The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
3729bit (highest order bit) of the <tt>value</tt> until it reaches the bit size of
Chris Lattnerd1d25172007-05-24 19:13:27 +0000[diff] [blame]3730the type <tt>ty2</tt>.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3731
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3732<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]3733
3734<h5>Example:</h5>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3735<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3736 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3737 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3738</pre>
3739</div>
3740
3741<!-- _______________________________________________________________________ -->
3742<div class="doc_subsubsection">
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]3743 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
3744</div>
3745
3746<div class="doc_text">
3747
3748<h5>Syntax:</h5>
3749
3750<pre>
3751 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3752</pre>
3753
3754<h5>Overview:</h5>
3755<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
3756<tt>ty2</tt>.</p>
3757
3758
3759<h5>Arguments:</h5>
3760<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
3761 point</a> value to cast and a <a href="#t_floating">floating point</a> type to
3762cast it to. The size of <tt>value</tt> must be larger than the size of
3763<tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
3764<i>no-op cast</i>.</p>
3765
3766<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3767<p> The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
3768<a href="#t_floating">floating point</a> type to a smaller
3769<a href="#t_floating">floating point</a> type. If the value cannot fit within
3770the destination type, <tt>ty2</tt>, then the results are undefined.</p>
Reid Spencer3fa91b02006-11-09 21:48:10 +0000[diff] [blame]3771
3772<h5>Example:</h5>
3773<pre>
3774 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
3775 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
3776</pre>
3777</div>
3778
3779<!-- _______________________________________________________________________ -->
3780<div class="doc_subsubsection">
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3781 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
3782</div>
3783<div class="doc_text">
3784
3785<h5>Syntax:</h5>
3786<pre>
3787 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3788</pre>
3789
3790<h5>Overview:</h5>
3791<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
3792floating point value.</p>
3793
3794<h5>Arguments:</h5>
3795<p>The '<tt>fpext</tt>' instruction takes a
3796<a href="#t_floating">floating point</a> <tt>value</tt> to cast,
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3797and a <a href="#t_floating">floating point</a> type to cast it to. The source
3798type must be smaller than the destination type.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3799
3800<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3801<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]3802<a href="#t_floating">floating point</a> type to a larger
3803<a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3804used to make a <i>no-op cast</i> because it always changes bits. Use
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]3805<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]3806
3807<h5>Example:</h5>
3808<pre>
3809 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
3810 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
3811</pre>
3812</div>
3813
3814<!-- _______________________________________________________________________ -->
3815<div class="doc_subsubsection">
Reid Spencer24d6da52007-01-21 00:29:26 +0000[diff] [blame]3816 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3817</div>
3818<div class="doc_text">
3819
3820<h5>Syntax:</h5>
3821<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3822 &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]3823</pre>
3824
3825<h5>Overview:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3826<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3827unsigned integer equivalent of type <tt>ty2</tt>.
3828</p>
3829
3830<h5>Arguments:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3831<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3832scalar or vector <a href="#t_floating">floating point</a> value, and a type
3833to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3834type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3835vector integer type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3836
3837<h5>Semantics:</h5>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3838<p> The '<tt>fptoui</tt>' instruction converts its
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3839<a href="#t_floating">floating point</a> operand into the nearest (rounding
3840towards zero) unsigned integer value. If the value cannot fit in <tt>ty2</tt>,
3841the results are undefined.</p>
3842
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3843<h5>Example:</h5>
3844<pre>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3845 %X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]3846 %Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Reid Spencer1539a1c2007-07-31 14:40:14 +0000[diff] [blame]3847 %X = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3848</pre>
3849</div>
3850
3851<!-- _______________________________________________________________________ -->
3852<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3853 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3854</div>
3855<div class="doc_text">
3856
3857<h5>Syntax:</h5>
3858<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3859 &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]3860</pre>
3861
3862<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3863<p>The '<tt>fptosi</tt>' instruction converts
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3864<a href="#t_floating">floating point</a> <tt>value</tt> to type <tt>ty2</tt>.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3865</p>
3866
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3867<h5>Arguments:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3868<p> The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3869scalar or vector <a href="#t_floating">floating point</a> value, and a type
3870to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3871type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3872vector integer type with the same number of elements as <tt>ty</tt></p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3873
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]3874<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3875<p>The '<tt>fptosi</tt>' instruction converts its
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3876<a href="#t_floating">floating point</a> operand into the nearest (rounding
3877towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
3878the results are undefined.</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3879
Chris Lattner33ba0d92001-07-09 00:26:23 +0000[diff] [blame]3880<h5>Example:</h5>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]3881<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]3882 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner88519042007-09-22 03:17:52 +0000[diff] [blame]3883 %Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3884 %X = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3885</pre>
3886</div>
3887
3888<!-- _______________________________________________________________________ -->
3889<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3890 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3891</div>
3892<div class="doc_text">
3893
3894<h5>Syntax:</h5>
3895<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3896 &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]3897</pre>
3898
3899<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3900<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3901integer and converts that value to the <tt>ty2</tt> type.</p>
3902
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3903<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3904<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be a
3905scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3906to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3907type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3908floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3909
3910<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3911<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3912integer quantity and converts it to the corresponding floating point value. If
Jeff Cohencb757312007-04-22 14:56:37 +0000[diff] [blame]3913the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3914
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3915<h5>Example:</h5>
3916<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3917 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]3918 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3919</pre>
3920</div>
3921
3922<!-- _______________________________________________________________________ -->
3923<div class="doc_subsubsection">
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3924 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3925</div>
3926<div class="doc_text">
3927
3928<h5>Syntax:</h5>
3929<pre>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3930 &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]3931</pre>
3932
3933<h5>Overview:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3934<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3935integer and converts that value to the <tt>ty2</tt> type.</p>
3936
3937<h5>Arguments:</h5>
Nate Begemanb348d182007-11-17 03:58:34 +0000[diff] [blame]3938<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be a
3939scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3940to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3941type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3942floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3943
3944<h5>Semantics:</h5>
Reid Spencerd4448792006-11-09 23:03:26 +0000[diff] [blame]3945<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3946integer quantity and converts it to the corresponding floating point value. If
Jeff Cohencb757312007-04-22 14:56:37 +0000[diff] [blame]3947the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3948
3949<h5>Example:</h5>
3950<pre>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]3951 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]3952 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]3953</pre>
3954</div>
3955
3956<!-- _______________________________________________________________________ -->
3957<div class="doc_subsubsection">
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3958 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
3959</div>
3960<div class="doc_text">
3961
3962<h5>Syntax:</h5>
3963<pre>
3964 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3965</pre>
3966
3967<h5>Overview:</h5>
3968<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
3969the integer type <tt>ty2</tt>.</p>
3970
3971<h5>Arguments:</h5>
3972<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]3973must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]3974<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]3975
3976<h5>Semantics:</h5>
3977<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
3978<tt>ty2</tt> by interpreting the pointer value as an integer and either
3979truncating or zero extending that value to the size of the integer type. If
3980<tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
3981<tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3982are the same size, then nothing is done (<i>no-op cast</i>) other than a type
3983change.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]3984
3985<h5>Example:</h5>
3986<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]3987 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
3988 %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]3989</pre>
3990</div>
3991
3992<!-- _______________________________________________________________________ -->
3993<div class="doc_subsubsection">
3994 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
3995</div>
3996<div class="doc_text">
3997
3998<h5>Syntax:</h5>
3999<pre>
4000 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
4001</pre>
4002
4003<h5>Overview:</h5>
4004<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to
4005a pointer type, <tt>ty2</tt>.</p>
4006
4007<h5>Arguments:</h5>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]4008<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4009value to cast, and a type to cast it to, which must be a
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4010<a href="#t_pointer">pointer</a> type.</p>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4011
4012<h5>Semantics:</h5>
4013<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
4014<tt>ty2</tt> by applying either a zero extension or a truncation depending on
4015the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
4016size of a pointer then a truncation is done. If <tt>value</tt> is smaller than
4017the size of a pointer then a zero extension is done. If they are the same size,
4018nothing is done (<i>no-op cast</i>).</p>
4019
4020<h5>Example:</h5>
4021<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4022 %X = inttoptr i32 255 to i32* <i>; yields zero extension on 64-bit architecture</i>
4023 %X = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
4024 %Y = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4025</pre>
4026</div>
4027
4028<!-- _______________________________________________________________________ -->
4029<div class="doc_subsubsection">
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4030 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4031</div>
4032<div class="doc_text">
4033
4034<h5>Syntax:</h5>
4035<pre>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4036 &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]4037</pre>
4038
4039<h5>Overview:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4040
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4041<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4042<tt>ty2</tt> without changing any bits.</p>
4043
4044<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4045
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4046<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be
Dan Gohman500233a2008-09-08 16:45:59 +0000[diff] [blame]4047a non-aggregate first class value, and a type to cast it to, which must also be
4048a non-aggregate <a href="#t_firstclass">first class</a> type. The bit sizes of
4049<tt>value</tt>
Reid Spencer19b569f2007-01-09 20:08:58 +0000[diff] [blame]4050and the destination type, <tt>ty2</tt>, must be identical. If the source
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4051type is a pointer, the destination type must also be a pointer. This
4052instruction supports bitwise conversion of vectors to integers and to vectors
4053of other types (as long as they have the same size).</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4054
4055<h5>Semantics:</h5>
Reid Spencer5c0ef472006-11-11 23:08:07 +0000[diff] [blame]4056<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer72679252006-11-11 21:00:47 +0000[diff] [blame]4057<tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
4058this conversion. The conversion is done as if the <tt>value</tt> had been
4059stored to memory and read back as type <tt>ty2</tt>. Pointer types may only be
4060converted to other pointer types with this instruction. To convert pointers to
4061other types, use the <a href="#i_inttoptr">inttoptr</a> or
4062<a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer9dee3ac2006-11-08 01:11:31 +0000[diff] [blame]4063
4064<h5>Example:</h5>
4065<pre>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4066 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4067 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4068 %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]4069</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4070</div>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4071
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4072<!-- ======================================================================= -->
4073<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
4074<div class="doc_text">
4075<p>The instructions in this category are the "miscellaneous"
4076instructions, which defy better classification.</p>
4077</div>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4078
4079<!-- _______________________________________________________________________ -->
4080<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
4081</div>
4082<div class="doc_text">
4083<h5>Syntax:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4084<pre> &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]4085</pre>
4086<h5>Overview:</h5>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4087<p>The '<tt>icmp</tt>' instruction returns a boolean value or
4088a vector of boolean values based on comparison
4089of its two integer, integer vector, or pointer operands.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4090<h5>Arguments:</h5>
4091<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4092the condition code indicating the kind of comparison to perform. It is not
4093a value, just a keyword. The possible condition code are:
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4094</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4095<ol>
4096 <li><tt>eq</tt>: equal</li>
4097 <li><tt>ne</tt>: not equal </li>
4098 <li><tt>ugt</tt>: unsigned greater than</li>
4099 <li><tt>uge</tt>: unsigned greater or equal</li>
4100 <li><tt>ult</tt>: unsigned less than</li>
4101 <li><tt>ule</tt>: unsigned less or equal</li>
4102 <li><tt>sgt</tt>: signed greater than</li>
4103 <li><tt>sge</tt>: signed greater or equal</li>
4104 <li><tt>slt</tt>: signed less than</li>
4105 <li><tt>sle</tt>: signed less or equal</li>
4106</ol>
Chris Lattner3b19d652007-01-15 01:54:13 +0000[diff] [blame]4107<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4108<a href="#t_pointer">pointer</a>
4109or integer <a href="#t_vector">vector</a> typed.
4110They must also be identical types.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4111<h5>Semantics:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4112<p>The '<tt>icmp</tt>' compares <tt>op1</tt> and <tt>op2</tt> according to
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4113the condition code given as <tt>cond</tt>. The comparison performed always
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4114yields either an <a href="#t_primitive"><tt>i1</tt></a> or vector of <tt>i1</tt> result, as follows:
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4115</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4116<ol>
4117 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
4118 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
4119 </li>
4120 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4121 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4122 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4123 <tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4124 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4125 <tt>true</tt> if <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4126 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4127 <tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4128 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4129 <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]4130 <li><tt>sgt</tt>: interprets the operands as signed values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4131 <tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4132 <li><tt>sge</tt>: interprets the operands as signed values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4133 <tt>true</tt> if <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4134 <li><tt>slt</tt>: interprets the operands as signed values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4135 <tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4136 <li><tt>sle</tt>: interprets the operands as signed values and yields
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4137 <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]4138</ol>
4139<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4140values are compared as if they were integers.</p>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4141<p>If the operands are integer vectors, then they are compared
4142element by element. The result is an <tt>i1</tt> vector with
4143the same number of elements as the values being compared.
4144Otherwise, the result is an <tt>i1</tt>.
4145</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4146
4147<h5>Example:</h5>
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4148<pre> &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
4149 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
4150 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
4151 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
4152 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
4153 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4154</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]4155
4156<p>Note that the code generator does not yet support vector types with
4157 the <tt>icmp</tt> instruction.</p>
4158
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4159</div>
4160
4161<!-- _______________________________________________________________________ -->
4162<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
4163</div>
4164<div class="doc_text">
4165<h5>Syntax:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4166<pre> &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]4167</pre>
4168<h5>Overview:</h5>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4169<p>The '<tt>fcmp</tt>' instruction returns a boolean value
4170or vector of boolean values based on comparison
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4171of its operands.</p>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4172<p>
4173If the operands are floating point scalars, then the result
4174type is a boolean (<a href="#t_primitive"><tt>i1</tt></a>).
4175</p>
4176<p>If the operands are floating point vectors, then the result type
4177is a vector of boolean with the same number of elements as the
4178operands being compared.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4179<h5>Arguments:</h5>
4180<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4181the condition code indicating the kind of comparison to perform. It is not
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4182a value, just a keyword. The possible condition code are:</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4183<ol>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4184 <li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4185 <li><tt>oeq</tt>: ordered and equal</li>
4186 <li><tt>ogt</tt>: ordered and greater than </li>
4187 <li><tt>oge</tt>: ordered and greater than or equal</li>
4188 <li><tt>olt</tt>: ordered and less than </li>
4189 <li><tt>ole</tt>: ordered and less than or equal</li>
4190 <li><tt>one</tt>: ordered and not equal</li>
4191 <li><tt>ord</tt>: ordered (no nans)</li>
4192 <li><tt>ueq</tt>: unordered or equal</li>
4193 <li><tt>ugt</tt>: unordered or greater than </li>
4194 <li><tt>uge</tt>: unordered or greater than or equal</li>
4195 <li><tt>ult</tt>: unordered or less than </li>
4196 <li><tt>ule</tt>: unordered or less than or equal</li>
4197 <li><tt>une</tt>: unordered or not equal</li>
4198 <li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4199 <li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4200</ol>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4201<p><i>Ordered</i> means that neither operand is a QNAN while
Reid Spencer93a49852006-12-06 07:08:07 +0000[diff] [blame]4202<i>unordered</i> means that either operand may be a QNAN.</p>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4203<p>Each of <tt>val1</tt> and <tt>val2</tt> arguments must be
4204either a <a href="#t_floating">floating point</a> type
4205or a <a href="#t_vector">vector</a> of floating point type.
4206They must have identical types.</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4207<h5>Semantics:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4208<p>The '<tt>fcmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4209according to the condition code given as <tt>cond</tt>.
4210If the operands are vectors, then the vectors are compared
4211element by element.
4212Each comparison performed
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4213always yields an <a href="#t_primitive">i1</a> result, as follows:</p>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4214<ol>
4215 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4216 <li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4217 <tt>op1</tt> is equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4218 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4219 <tt>op1</tt> is greather than <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4220 <li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4221 <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4222 <li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4223 <tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4224 <li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4225 <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4226 <li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4227 <tt>op1</tt> is not equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4228 <li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
4229 <li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4230 <tt>op1</tt> is equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4231 <li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4232 <tt>op1</tt> is greater than <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4233 <li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4234 <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4235 <li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4236 <tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4237 <li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4238 <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4239 <li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4240 <tt>op1</tt> is not equal to <tt>op2</tt>.</li>
Reid Spencerb7f26282006-11-19 03:00:14 +0000[diff] [blame]4241 <li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4242 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
4243</ol>
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4244
4245<h5>Example:</h5>
4246<pre> &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]4247 &lt;result&gt; = fcmp one float 4.0, 5.0 <i>; yields: result=true</i>
4248 &lt;result&gt; = fcmp olt float 4.0, 5.0 <i>; yields: result=true</i>
4249 &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]4250</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]4251
4252<p>Note that the code generator does not yet support vector types with
4253 the <tt>fcmp</tt> instruction.</p>
4254
Reid Spencerf3a70a62006-11-18 21:50:54 +0000[diff] [blame]4255</div>
4256
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4257<!-- _______________________________________________________________________ -->
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4258<div class="doc_subsubsection">
4259 <a name="i_vicmp">'<tt>vicmp</tt>' Instruction</a>
4260</div>
4261<div class="doc_text">
4262<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4263<pre> &lt;result&gt; = vicmp &lt;cond&gt; &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt; <i>; yields {ty}:result</i>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4264</pre>
4265<h5>Overview:</h5>
4266<p>The '<tt>vicmp</tt>' instruction returns an integer vector value based on
4267element-wise comparison of its two integer vector operands.</p>
4268<h5>Arguments:</h5>
4269<p>The '<tt>vicmp</tt>' instruction takes three operands. The first operand is
4270the condition code indicating the kind of comparison to perform. It is not
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4271a value, just a keyword. The possible condition code are:</p>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4272<ol>
4273 <li><tt>eq</tt>: equal</li>
4274 <li><tt>ne</tt>: not equal </li>
4275 <li><tt>ugt</tt>: unsigned greater than</li>
4276 <li><tt>uge</tt>: unsigned greater or equal</li>
4277 <li><tt>ult</tt>: unsigned less than</li>
4278 <li><tt>ule</tt>: unsigned less or equal</li>
4279 <li><tt>sgt</tt>: signed greater than</li>
4280 <li><tt>sge</tt>: signed greater or equal</li>
4281 <li><tt>slt</tt>: signed less than</li>
4282 <li><tt>sle</tt>: signed less or equal</li>
4283</ol>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4284<p>The remaining two arguments must be <a href="#t_vector">vector</a> or
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4285<a href="#t_integer">integer</a> typed. They must also be identical types.</p>
4286<h5>Semantics:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4287<p>The '<tt>vicmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4288according to the condition code given as <tt>cond</tt>. The comparison yields a
4289<a href="#t_vector">vector</a> of <a href="#t_integer">integer</a> result, of
4290identical type as the values being compared. The most significant bit in each
4291element is 1 if the element-wise comparison evaluates to true, and is 0
4292otherwise. All other bits of the result are undefined. The condition codes
4293are evaluated identically to the <a href="#i_icmp">'<tt>icmp</tt>'
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4294instruction</a>.</p>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4295
4296<h5>Example:</h5>
4297<pre>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4298 &lt;result&gt; = vicmp eq &lt;2 x i32&gt; &lt; i32 4, i32 0&gt;, &lt; i32 5, i32 0&gt; <i>; yields: result=&lt;2 x i32&gt; &lt; i32 0, i32 -1 &gt;</i>
4299 &lt;result&gt; = vicmp ult &lt;2 x i8 &gt; &lt; i8 1, i8 2&gt;, &lt; i8 2, i8 2 &gt; <i>; yields: result=&lt;2 x i8&gt; &lt; i8 -1, i8 0 &gt;</i>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4300</pre>
4301</div>
4302
4303<!-- _______________________________________________________________________ -->
4304<div class="doc_subsubsection">
4305 <a name="i_vfcmp">'<tt>vfcmp</tt>' Instruction</a>
4306</div>
4307<div class="doc_text">
4308<h5>Syntax:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4309<pre> &lt;result&gt; = vfcmp &lt;cond&gt; &lt;ty&gt; &lt;op1&gt;, &lt;op2&gt;</pre>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4310<h5>Overview:</h5>
4311<p>The '<tt>vfcmp</tt>' instruction returns an integer vector value based on
4312element-wise comparison of its two floating point vector operands. The output
4313elements have the same width as the input elements.</p>
4314<h5>Arguments:</h5>
4315<p>The '<tt>vfcmp</tt>' instruction takes three operands. The first operand is
4316the condition code indicating the kind of comparison to perform. It is not
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4317a value, just a keyword. The possible condition code are:</p>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4318<ol>
4319 <li><tt>false</tt>: no comparison, always returns false</li>
4320 <li><tt>oeq</tt>: ordered and equal</li>
4321 <li><tt>ogt</tt>: ordered and greater than </li>
4322 <li><tt>oge</tt>: ordered and greater than or equal</li>
4323 <li><tt>olt</tt>: ordered and less than </li>
4324 <li><tt>ole</tt>: ordered and less than or equal</li>
4325 <li><tt>one</tt>: ordered and not equal</li>
4326 <li><tt>ord</tt>: ordered (no nans)</li>
4327 <li><tt>ueq</tt>: unordered or equal</li>
4328 <li><tt>ugt</tt>: unordered or greater than </li>
4329 <li><tt>uge</tt>: unordered or greater than or equal</li>
4330 <li><tt>ult</tt>: unordered or less than </li>
4331 <li><tt>ule</tt>: unordered or less than or equal</li>
4332 <li><tt>une</tt>: unordered or not equal</li>
4333 <li><tt>uno</tt>: unordered (either nans)</li>
4334 <li><tt>true</tt>: no comparison, always returns true</li>
4335</ol>
4336<p>The remaining two arguments must be <a href="#t_vector">vector</a> of
4337<a href="#t_floating">floating point</a> typed. They must also be identical
4338types.</p>
4339<h5>Semantics:</h5>
Gabor Greiffb224a22008-08-07 21:46:00 +0000[diff] [blame]4340<p>The '<tt>vfcmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4341according to the condition code given as <tt>cond</tt>. The comparison yields a
4342<a href="#t_vector">vector</a> of <a href="#t_integer">integer</a> result, with
4343an identical number of elements as the values being compared, and each element
4344having identical with to the width of the floating point elements. The most
4345significant bit in each element is 1 if the element-wise comparison evaluates to
4346true, and is 0 otherwise. All other bits of the result are undefined. The
4347condition codes are evaluated identically to the
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4348<a href="#i_fcmp">'<tt>fcmp</tt>' instruction</a>.</p>
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4349
4350<h5>Example:</h5>
4351<pre>
Chris Lattner50ad45c2008-10-13 16:55:18 +0000[diff] [blame]4352 <i>; yields: result=&lt;2 x i32&gt; &lt; i32 0, i32 -1 &gt;</i>
4353 &lt;result&gt; = vfcmp oeq &lt;2 x float&gt; &lt; float 4, float 0 &gt;, &lt; float 5, float 0 &gt;
4354
4355 <i>; yields: result=&lt;2 x i64&gt; &lt; i64 -1, i64 0 &gt;</i>
4356 &lt;result&gt; = vfcmp ult &lt;2 x double&gt; &lt; double 1, double 2 &gt;, &lt; double 2, double 2&gt;
Nate Begemanac80ade2008-05-12 19:01:56 +0000[diff] [blame]4357</pre>
4358</div>
4359
4360<!-- _______________________________________________________________________ -->
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4361<div class="doc_subsubsection">
4362 <a name="i_phi">'<tt>phi</tt>' Instruction</a>
4363</div>
4364
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4365<div class="doc_text">
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4366
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4367<h5>Syntax:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4368
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4369<pre> &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...<br></pre>
4370<h5>Overview:</h5>
4371<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
4372the SSA graph representing the function.</p>
4373<h5>Arguments:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4374
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4375<p>The type of the incoming values is specified with the first type
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4376field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
4377as arguments, with one pair for each predecessor basic block of the
4378current block. Only values of <a href="#t_firstclass">first class</a>
4379type may be used as the value arguments to the PHI node. Only labels
4380may be used as the label arguments.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4381
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4382<p>There must be no non-phi instructions between the start of a basic
4383block and the PHI instructions: i.e. PHI instructions must be first in
4384a basic block.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4385
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4386<h5>Semantics:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4387
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4388<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
4389specified by the pair corresponding to the predecessor basic block that executed
4390just prior to the current block.</p>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4391
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4392<h5>Example:</h5>
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4393<pre>
4394Loop: ; Infinite loop that counts from 0 on up...
4395 %indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]
4396 %nextindvar = add i32 %indvar, 1
4397 br label %Loop
4398</pre>
Reid Spencer2fd21e62006-11-08 01:18:52 +0000[diff] [blame]4399</div>
4400
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4401<!-- _______________________________________________________________________ -->
4402<div class="doc_subsubsection">
4403 <a name="i_select">'<tt>select</tt>' Instruction</a>
4404</div>
4405
4406<div class="doc_text">
4407
4408<h5>Syntax:</h5>
4409
4410<pre>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4411 &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>
4412
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4413 <i>selty</i> is either i1 or {&lt;N x i1&gt;}
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4414</pre>
4415
4416<h5>Overview:</h5>
4417
4418<p>
4419The '<tt>select</tt>' instruction is used to choose one value based on a
4420condition, without branching.
4421</p>
4422
4423
4424<h5>Arguments:</h5>
4425
4426<p>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4427The '<tt>select</tt>' instruction requires an 'i1' value or
4428a vector of 'i1' values indicating the
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4429condition, and two values of the same <a href="#t_firstclass">first class</a>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4430type. If the val1/val2 are vectors and
4431the condition is a scalar, then entire vectors are selected, not
Chris Lattner5568e942008-05-20 20:48:21 +0000[diff] [blame]4432individual elements.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4433</p>
4434
4435<h5>Semantics:</h5>
4436
4437<p>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4438If the condition is an i1 and it evaluates to 1, the instruction returns the first
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4439value argument; otherwise, it returns the second value argument.
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4440</p>
Dan Gohmanf72fb672008-09-09 01:02:47 +0000[diff] [blame]4441<p>
4442If the condition is a vector of i1, then the value arguments must
4443be vectors of the same size, and the selection is done element
4444by element.
4445</p>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4446
4447<h5>Example:</h5>
4448
4449<pre>
Reid Spencerc78f3372007-01-12 03:35:51 +0000[diff] [blame]4450 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4451</pre>
Dan Gohmanc766f722009-01-22 01:39:38 +0000[diff] [blame]4452
4453<p>Note that the code generator does not yet support conditions
4454 with vector type.</p>
4455
Chris Lattnercc37aae2004-03-12 05:50:16 +0000[diff] [blame]4456</div>
4457
Robert Bocchino05ccd702006-01-15 20:48:27 +0000[diff] [blame]4458
4459<!-- _______________________________________________________________________ -->
4460<div class="doc_subsubsection">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4461 <a name="i_call">'<tt>call</tt>' Instruction</a>
4462</div>
4463
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4464<div class="doc_text">
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4465
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4466<h5>Syntax:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4467<pre>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]4468 &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]4469</pre>
4470
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4471<h5>Overview:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4472
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4473<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4474
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4475<h5>Arguments:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4476
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4477<p>This instruction requires several arguments:</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4478
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4479<ol>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4480 <li>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]4481 <p>The optional "tail" marker indicates whether the callee function accesses
4482 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4483 function call is eligible for tail call optimization. Note that calls may
4484 be marked "tail" even if they do not occur before a <a
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4485 href="#i_ret"><tt>ret</tt></a> instruction.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4486 </li>
4487 <li>
Duncan Sands8036ca42007-03-30 12:22:09 +0000[diff] [blame]4488 <p>The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]4489 convention</a> the call should use. If none is specified, the call defaults
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4490 to using C calling conventions.</p>
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]4491 </li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]4492
4493 <li>
4494 <p>The optional <a href="#paramattrs">Parameter Attributes</a> list for
4495 return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>',
4496 and '<tt>inreg</tt>' attributes are valid here.</p>
4497 </li>
4498
Chris Lattnerbad10ee2005-05-06 22:57:40 +0000[diff] [blame]4499 <li>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]4500 <p>'<tt>ty</tt>': the type of the call instruction itself which is also
4501 the type of the return value. Functions that return no value are marked
4502 <tt><a href="#t_void">void</a></tt>.</p>
4503 </li>
4504 <li>
4505 <p>'<tt>fnty</tt>': shall be the signature of the pointer to function
4506 value being invoked. The argument types must match the types implied by
4507 this signature. This type can be omitted if the function is not varargs
4508 and if the function type does not return a pointer to a function.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4509 </li>
4510 <li>
4511 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
4512 be invoked. In most cases, this is a direct function invocation, but
4513 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4514 to function value.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4515 </li>
4516 <li>
4517 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencera7e302a2005-05-01 22:22:57 +0000[diff] [blame]4518 function signature argument types. All arguments must be of
4519 <a href="#t_firstclass">first class</a> type. If the function signature
4520 indicates the function accepts a variable number of arguments, the extra
4521 arguments can be specified.</p>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4522 </li>
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]4523 <li>
Devang Patel307e8ab2008-10-07 17:48:33 +0000[diff] [blame]4524 <p>The optional <a href="#fnattrs">function attributes</a> list. Only
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]4525 '<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
4526 '<tt>readnone</tt>' attributes are valid here.</p>
4527 </li>
Chris Lattner6536cfe2002-05-06 22:08:29 +0000[diff] [blame]4528</ol>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4529
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4530<h5>Semantics:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4531
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4532<p>The '<tt>call</tt>' instruction is used to cause control flow to
4533transfer to a specified function, with its incoming arguments bound to
4534the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
4535instruction in the called function, control flow continues with the
4536instruction after the function call, and the return value of the
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4537function is bound to the result argument.</p>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4538
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]4539<h5>Example:</h5>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4540
4541<pre>
Nick Lewyckydb7e3c92007-09-08 13:57:50 +0000[diff] [blame]4542 %retval = call i32 @test(i32 %argc)
Chris Lattner772fccf2008-03-21 17:24:17 +0000[diff] [blame]4543 call i32 (i8 *, ...)* @printf(i8 * %msg, i32 12, i8 42) <i>; yields i32</i>
4544 %X = tail call i32 @foo() <i>; yields i32</i>
4545 %Y = tail call <a href="#callingconv">fastcc</a> i32 @foo() <i>; yields i32</i>
4546 call void %foo(i8 97 signext)
Devang Patelc3fc6df2008-03-10 20:49:15 +0000[diff] [blame]4547
4548 %struct.A = type { i32, i8 }
Devang Patelf642f472008-10-06 18:50:38 +0000[diff] [blame]4549 %r = call %struct.A @foo() <i>; yields { 32, i8 }</i>
Dan Gohmanb1e6b962008-10-04 19:00:07 +0000[diff] [blame]4550 %gr = extractvalue %struct.A %r, 0 <i>; yields i32</i>
4551 %gr1 = extractvalue %struct.A %r, 1 <i>; yields i8</i>
Chris Lattner85a350f2008-10-08 06:26:11 +0000[diff] [blame]4552 %Z = call void @foo() noreturn <i>; indicates that %foo never returns normally</i>
Matthijs Kooijmancb73d192008-10-07 10:03:45 +0000[diff] [blame]4553 %ZZ = call zeroext i32 @bar() <i>; Return value is %zero extended</i>
Chris Lattner2bff5242005-05-06 05:47:36 +0000[diff] [blame]4554</pre>
4555
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4556</div>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4557
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4558<!-- _______________________________________________________________________ -->
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4559<div class="doc_subsubsection">
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]4560 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4561</div>
4562
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4563<div class="doc_text">
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4564
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]4565<h5>Syntax:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4566
4567<pre>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4568 &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]4569</pre>
4570
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]4571<h5>Overview:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4572
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4573<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4574the "variable argument" area of a function call. It is used to implement the
4575<tt>va_arg</tt> macro in C.</p>
4576
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]4577<h5>Arguments:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4578
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4579<p>This instruction takes a <tt>va_list*</tt> value and the type of
4580the argument. It returns a value of the specified argument type and
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4581increments the <tt>va_list</tt> to point to the next argument. The
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4582actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4583
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]4584<h5>Semantics:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4585
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4586<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
4587type from the specified <tt>va_list</tt> and causes the
4588<tt>va_list</tt> to point to the next argument. For more information,
4589see the variable argument handling <a href="#int_varargs">Intrinsic
4590Functions</a>.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4591
4592<p>It is legal for this instruction to be called in a function which does not
4593take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4594function.</p>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4595
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4596<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4597href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4598argument.</p>
4599
Chris Lattner8d1a81d2003-10-18 05:51:36 +0000[diff] [blame]4600<h5>Example:</h5>
Chris Lattnere19d7a72004-09-27 21:51:25 +0000[diff] [blame]4601
4602<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
4603
Dan Gohmanf3e60bd2009-01-12 23:12:39 +0000[diff] [blame]4604<p>Note that the code generator does not yet fully support va_arg
4605 on many targets. Also, it does not currently support va_arg with
4606 aggregate types on any target.</p>
4607
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4608</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4609
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4610<!-- *********************************************************************** -->
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4611<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
4612<!-- *********************************************************************** -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4613
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4614<div class="doc_text">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4615
4616<p>LLVM supports the notion of an "intrinsic function". These functions have
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4617well known names and semantics and are required to follow certain restrictions.
4618Overall, these intrinsics represent an extension mechanism for the LLVM
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4619language that does not require changing all of the transformations in LLVM when
Gabor Greif04367bf2007-07-06 22:07:22 +0000[diff] [blame]4620adding to the language (or the bitcode reader/writer, the parser, etc...).</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4621
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]4622<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4623prefix is reserved in LLVM for intrinsic names; thus, function names may not
4624begin with this prefix. Intrinsic functions must always be external functions:
4625you cannot define the body of intrinsic functions. Intrinsic functions may
4626only be used in call or invoke instructions: it is illegal to take the address
4627of an intrinsic function. Additionally, because intrinsic functions are part
4628of the LLVM language, it is required if any are added that they be documented
4629here.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4630
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4631<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents
4632a family of functions that perform the same operation but on different data
4633types. Because LLVM can represent over 8 million different integer types,
4634overloading is used commonly to allow an intrinsic function to operate on any
4635integer type. One or more of the argument types or the result type can be
4636overloaded to accept any integer type. Argument types may also be defined as
4637exactly matching a previous argument's type or the result type. This allows an
4638intrinsic function which accepts multiple arguments, but needs all of them to
4639be of the same type, to only be overloaded with respect to a single argument or
4640the result.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4641
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]4642<p>Overloaded intrinsics will have the names of its overloaded argument types
4643encoded into its function name, each preceded by a period. Only those types
4644which are overloaded result in a name suffix. Arguments whose type is matched
4645against another type do not. For example, the <tt>llvm.ctpop</tt> function can
4646take an integer of any width and returns an integer of exactly the same integer
4647width. This leads to a family of functions such as
4648<tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29 %val)</tt>.
4649Only one type, the return type, is overloaded, and only one type suffix is
4650required. Because the argument's type is matched against the return type, it
4651does not require its own name suffix.</p>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]4652
4653<p>To learn how to add an intrinsic function, please see the
4654<a href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4655</p>
4656
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4657</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4658
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4659<!-- ======================================================================= -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4660<div class="doc_subsection">
4661 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
4662</div>
4663
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4664<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4665
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4666<p>Variable argument support is defined in LLVM with the <a
Chris Lattnerfb6977d2006-01-13 23:26:01 +0000[diff] [blame]4667 href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4668intrinsic functions. These functions are related to the similarly
4669named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4670
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4671<p>All of these functions operate on arguments that use a
4672target-specific value type "<tt>va_list</tt>". The LLVM assembly
4673language reference manual does not define what this type is, so all
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4674transformations should be prepared to handle these functions regardless of
4675the type used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4676
Chris Lattner374ab302006-05-15 17:26:46 +0000[diff] [blame]4677<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4678instruction and the variable argument handling intrinsic functions are
4679used.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4680
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4681<div class="doc_code">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4682<pre>
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]4683define i32 @test(i32 %X, ...) {
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4684 ; Initialize variable argument processing
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4685 %ap = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4686 %ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]4687 call void @llvm.va_start(i8* %ap2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4688
4689 ; Read a single integer argument
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4690 %tmp = va_arg i8** %ap, i32
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4691
4692 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4693 %aq = alloca i8*
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4694 %aq2 = bitcast i8** %aq to i8*
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4695 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]4696 call void @llvm.va_end(i8* %aq2)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4697
4698 ; Stop processing of arguments.
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]4699 call void @llvm.va_end(i8* %ap2)
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]4700 ret i32 %tmp
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4701}
Anton Korobeynikov5d522f32007-03-21 23:58:04 +0000[diff] [blame]4702
4703declare void @llvm.va_start(i8*)
4704declare void @llvm.va_copy(i8*, i8*)
4705declare void @llvm.va_end(i8*)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4706</pre>
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4707</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4708
Bill Wendling2f7a8b02007-05-29 09:04:49 +0000[diff] [blame]4709</div>
4710
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4711<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4712<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4713 <a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4714</div>
4715
4716
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4717<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4718<h5>Syntax:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4719<pre> declare void %llvm.va_start(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4720<h5>Overview:</h5>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4721<p>The '<tt>llvm.va_start</tt>' intrinsic initializes
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4722<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
4723href="#i_va_arg">va_arg</a></tt>.</p>
4724
4725<h5>Arguments:</h5>
4726
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4727<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]4728
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4729<h5>Semantics:</h5>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4730
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]4731<p>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4732macro available in C. In a target-dependent way, it initializes the
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4733<tt>va_list</tt> element to which the argument points, so that the next call to
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4734<tt>va_arg</tt> will produce the first variable argument passed to the function.
4735Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4736last argument of the function as the compiler can figure that out.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4737
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4738</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4739
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4740<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4741<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4742 <a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4743</div>
4744
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4745<div class="doc_text">
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4746<h5>Syntax:</h5>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4747<pre> declare void @llvm.va_end(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4748<h5>Overview:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4749
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4750<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*&lt;arglist&gt;</tt>,
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4751which has been initialized previously with <tt><a href="#int_va_start">llvm.va_start</a></tt>
Chris Lattner261efe92003-11-25 01:02:51 +0000[diff] [blame]4752or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4753
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4754<h5>Arguments:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4755
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4756<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]4757
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4758<h5>Semantics:</h5>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4759
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4760<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4761macro available in C. In a target-dependent way, it destroys the
4762<tt>va_list</tt> element to which the argument points. Calls to <a
4763href="#int_va_start"><tt>llvm.va_start</tt></a> and <a href="#int_va_copy">
4764<tt>llvm.va_copy</tt></a> must be matched exactly with calls to
4765<tt>llvm.va_end</tt>.</p>
Chris Lattnerb75137d2007-01-08 07:55:15 +0000[diff] [blame]4766
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4767</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4768
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4769<!-- _______________________________________________________________________ -->
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4770<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4771 <a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4772</div>
4773
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4774<div class="doc_text">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4775
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4776<h5>Syntax:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4777
4778<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4779 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4780</pre>
4781
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4782<h5>Overview:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4783
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4784<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
4785from the source argument list to the destination argument list.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4786
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4787<h5>Arguments:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4788
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4789<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Andrew Lenharthd0a4c622005-06-22 20:38:11 +0000[diff] [blame]4790The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
Andrew Lenharth8bf607a2005-06-18 18:28:17 +0000[diff] [blame]4791
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4792
Chris Lattnerd9ad5b32003-05-08 04:57:36 +0000[diff] [blame]4793<h5>Semantics:</h5>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4794
Jeff Cohenb627eab2007-04-29 01:07:00 +0000[diff] [blame]4795<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
4796macro available in C. In a target-dependent way, it copies the source
4797<tt>va_list</tt> element into the destination <tt>va_list</tt> element. This
4798intrinsic is necessary because the <tt><a href="#int_va_start">
4799llvm.va_start</a></tt> intrinsic may be arbitrarily complex and require, for
4800example, memory allocation.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4801
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]4802</div>
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]4803
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]4804<!-- ======================================================================= -->
4805<div class="doc_subsection">
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4806 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
4807</div>
4808
4809<div class="doc_text">
4810
4811<p>
4812LLVM support for <a href="GarbageCollection.html">Accurate Garbage
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]4813Collection</a> (GC) requires the implementation and generation of these
4814intrinsics.
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4815These intrinsics allow identification of <a href="#int_gcroot">GC roots on the
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4816stack</a>, as well as garbage collector implementations that require <a
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4817href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a> barriers.
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4818Front-ends for type-safe garbage collected languages should generate these
4819intrinsics to make use of the LLVM garbage collectors. For more details, see <a
4820href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
4821</p>
Christopher Lamb303dae92007-12-17 01:00:21 +0000[diff] [blame]4822
4823<p>The garbage collection intrinsics only operate on objects in the generic
4824 address space (address space zero).</p>
4825
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4826</div>
4827
4828<!-- _______________________________________________________________________ -->
4829<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4830 <a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4831</div>
4832
4833<div class="doc_text">
4834
4835<h5>Syntax:</h5>
4836
4837<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4838 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4839</pre>
4840
4841<h5>Overview:</h5>
4842
John Criswell9e2485c2004-12-10 15:51:16 +0000[diff] [blame]4843<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4844the code generator, and allows some metadata to be associated with it.</p>
4845
4846<h5>Arguments:</h5>
4847
4848<p>The first argument specifies the address of a stack object that contains the
4849root pointer. The second pointer (which must be either a constant or a global
4850value address) contains the meta-data to be associated with the root.</p>
4851
4852<h5>Semantics:</h5>
4853
Chris Lattner05d67092008-04-24 05:59:56 +0000[diff] [blame]4854<p>At runtime, a call to this intrinsic stores a null pointer into the "ptrloc"
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4855location. At compile-time, the code generator generates information to allow
Gordon Henriksene1433f22007-12-25 02:31:26 +0000[diff] [blame]4856the runtime to find the pointer at GC safe points. The '<tt>llvm.gcroot</tt>'
4857intrinsic may only be used in a function which <a href="#gc">specifies a GC
4858algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4859
4860</div>
4861
4862
4863<!-- _______________________________________________________________________ -->
4864<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4865 <a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4866</div>
4867
4868<div class="doc_text">
4869
4870<h5>Syntax:</h5>
4871
4872<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4873 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4874</pre>
4875
4876<h5>Overview:</h5>
4877
4878<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
4879locations, allowing garbage collector implementations that require read
4880barriers.</p>
4881
4882<h5>Arguments:</h5>
4883
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]4884<p>The second argument is the address to read from, which should be an address
4885allocated from the garbage collector. The first object is a pointer to the
4886start of the referenced object, if needed by the language runtime (otherwise
4887null).</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4888
4889<h5>Semantics:</h5>
4890
4891<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
4892instruction, but may be replaced with substantially more complex code by the
Gordon Henriksene1433f22007-12-25 02:31:26 +0000[diff] [blame]4893garbage collector runtime, as needed. The '<tt>llvm.gcread</tt>' intrinsic
4894may only be used in a function which <a href="#gc">specifies a GC
4895algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4896
4897</div>
4898
4899
4900<!-- _______________________________________________________________________ -->
4901<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4902 <a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4903</div>
4904
4905<div class="doc_text">
4906
4907<h5>Syntax:</h5>
4908
4909<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]4910 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4911</pre>
4912
4913<h5>Overview:</h5>
4914
4915<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
4916locations, allowing garbage collector implementations that require write
4917barriers (such as generational or reference counting collectors).</p>
4918
4919<h5>Arguments:</h5>
4920
Chris Lattner80626e92006-03-14 20:02:51 +0000[diff] [blame]4921<p>The first argument is the reference to store, the second is the start of the
4922object to store it to, and the third is the address of the field of Obj to
4923store to. If the runtime does not require a pointer to the object, Obj may be
4924null.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4925
4926<h5>Semantics:</h5>
4927
4928<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
4929instruction, but may be replaced with substantially more complex code by the
Gordon Henriksene1433f22007-12-25 02:31:26 +0000[diff] [blame]4930garbage collector runtime, as needed. The '<tt>llvm.gcwrite</tt>' intrinsic
4931may only be used in a function which <a href="#gc">specifies a GC
4932algorithm</a>.</p>
Chris Lattnerd7923912004-05-23 21:06:01 +0000[diff] [blame]4933
4934</div>
4935
4936
4937
4938<!-- ======================================================================= -->
4939<div class="doc_subsection">
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4940 <a name="int_codegen">Code Generator Intrinsics</a>
4941</div>
4942
4943<div class="doc_text">
4944<p>
4945These intrinsics are provided by LLVM to expose special features that may only
4946be implemented with code generator support.
4947</p>
4948
4949</div>
4950
4951<!-- _______________________________________________________________________ -->
4952<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4953 <a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4954</div>
4955
4956<div class="doc_text">
4957
4958<h5>Syntax:</h5>
4959<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]4960 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4961</pre>
4962
4963<h5>Overview:</h5>
4964
4965<p>
Chris Lattner32b5d712006-10-15 20:05:59 +0000[diff] [blame]4966The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
4967target-specific value indicating the return address of the current function
4968or one of its callers.
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4969</p>
4970
4971<h5>Arguments:</h5>
4972
4973<p>
4974The argument to this intrinsic indicates which function to return the address
4975for. Zero indicates the calling function, one indicates its caller, etc. The
4976argument is <b>required</b> to be a constant integer value.
4977</p>
4978
4979<h5>Semantics:</h5>
4980
4981<p>
4982The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
4983the return address of the specified call frame, or zero if it cannot be
4984identified. The value returned by this intrinsic is likely to be incorrect or 0
4985for arguments other than zero, so it should only be used for debugging purposes.
4986</p>
4987
4988<p>
4989Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]4990aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4991source-language caller.
4992</p>
4993</div>
4994
4995
4996<!-- _______________________________________________________________________ -->
4997<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]4998 <a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]4999</div>
5000
5001<div class="doc_text">
5002
5003<h5>Syntax:</h5>
5004<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5005 declare i8 *@llvm.frameaddress(i32 &lt;level&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5006</pre>
5007
5008<h5>Overview:</h5>
5009
5010<p>
Chris Lattner32b5d712006-10-15 20:05:59 +0000[diff] [blame]5011The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
5012target-specific frame pointer value for the specified stack frame.
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5013</p>
5014
5015<h5>Arguments:</h5>
5016
5017<p>
5018The argument to this intrinsic indicates which function to return the frame
5019pointer for. Zero indicates the calling function, one indicates its caller,
5020etc. The argument is <b>required</b> to be a constant integer value.
5021</p>
5022
5023<h5>Semantics:</h5>
5024
5025<p>
5026The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
5027the frame address of the specified call frame, or zero if it cannot be
5028identified. The value returned by this intrinsic is likely to be incorrect or 0
5029for arguments other than zero, so it should only be used for debugging purposes.
5030</p>
5031
5032<p>
5033Note that calling this intrinsic does not prevent function inlining or other
Chris Lattnerb40bb382005-03-07 20:30:51 +0000[diff] [blame]5034aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5035source-language caller.
5036</p>
5037</div>
5038
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5039<!-- _______________________________________________________________________ -->
5040<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5041 <a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5042</div>
5043
5044<div class="doc_text">
5045
5046<h5>Syntax:</h5>
5047<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5048 declare i8 *@llvm.stacksave()
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5049</pre>
5050
5051<h5>Overview:</h5>
5052
5053<p>
5054The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5055the function stack, for use with <a href="#int_stackrestore">
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5056<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
5057features like scoped automatic variable sized arrays in C99.
5058</p>
5059
5060<h5>Semantics:</h5>
5061
5062<p>
5063This intrinsic returns a opaque pointer value that can be passed to <a
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5064href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5065<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
5066<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
5067state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
5068practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
5069that were allocated after the <tt>llvm.stacksave</tt> was executed.
5070</p>
5071
5072</div>
5073
5074<!-- _______________________________________________________________________ -->
5075<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5076 <a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5077</div>
5078
5079<div class="doc_text">
5080
5081<h5>Syntax:</h5>
5082<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5083 declare void @llvm.stackrestore(i8 * %ptr)
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5084</pre>
5085
5086<h5>Overview:</h5>
5087
5088<p>
5089The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
5090the function stack to the state it was in when the corresponding <a
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5091href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5092useful for implementing language features like scoped automatic variable sized
5093arrays in C99.
5094</p>
5095
5096<h5>Semantics:</h5>
5097
5098<p>
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5099See the description for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.
Chris Lattner57e1f392006-01-13 02:03:13 +0000[diff] [blame]5100</p>
5101
5102</div>
5103
5104
5105<!-- _______________________________________________________________________ -->
5106<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5107 <a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5108</div>
5109
5110<div class="doc_text">
5111
5112<h5>Syntax:</h5>
5113<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5114 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]5115</pre>
5116
5117<h5>Overview:</h5>
5118
5119
5120<p>
5121The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]5122a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
5123no
5124effect on the behavior of the program but can change its performance
Chris Lattner2a615362005-02-28 19:47:14 +0000[diff] [blame]5125characteristics.
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5126</p>
5127
5128<h5>Arguments:</h5>
5129
5130<p>
5131<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
5132determining if the fetch should be for a read (0) or write (1), and
5133<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattneraeffb4a2005-03-07 20:31:38 +0000[diff] [blame]5134locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattner9a9d7ac2005-02-28 19:24:19 +0000[diff] [blame]5135<tt>locality</tt> arguments must be constant integers.
5136</p>
5137
5138<h5>Semantics:</h5>
5139
5140<p>
5141This intrinsic does not modify the behavior of the program. In particular,
5142prefetches cannot trap and do not produce a value. On targets that support this
5143intrinsic, the prefetch can provide hints to the processor cache for better
5144performance.
5145</p>
5146
5147</div>
5148
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5149<!-- _______________________________________________________________________ -->
5150<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5151 <a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5152</div>
5153
5154<div class="doc_text">
5155
5156<h5>Syntax:</h5>
5157<pre>
Chris Lattner1df4f752007-09-21 17:30:40 +0000[diff] [blame]5158 declare void @llvm.pcmarker(i32 &lt;id&gt;)
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5159</pre>
5160
5161<h5>Overview:</h5>
5162
5163
5164<p>
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]5165The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]5166(PC) in a region of
5167code to simulators and other tools. The method is target specific, but it is
5168expected that the marker will use exported symbols to transmit the PC of the
5169marker.
5170The marker makes no guarantees that it will remain with any specific instruction
5171after optimizations. It is possible that the presence of a marker will inhibit
Chris Lattnerb3e7afd2006-03-24 07:16:10 +0000[diff] [blame]5172optimizations. The intended use is to be inserted after optimizations to allow
John Criswellfc6b8952005-05-16 16:17:45 +0000[diff] [blame]5173correlations of simulation runs.
Andrew Lenharth7f4ec3b2005-03-28 20:05:49 +0000[diff] [blame]5174</p>
5175
5176<h5>Arguments:</h5>
5177
5178<p>
5179<tt>id</tt> is a numerical id identifying the marker.
5180</p>
5181
5182<h5>Semantics:</h5>
5183
5184<p>
5185This intrinsic does not modify the behavior of the program. Backends that do not
5186support this intrinisic may ignore it.
5187</p>
5188
5189</div>
5190
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5191<!-- _______________________________________________________________________ -->
5192<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5193 <a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5194</div>
5195
5196<div class="doc_text">
5197
5198<h5>Syntax:</h5>
5199<pre>
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5200 declare i64 @llvm.readcyclecounter( )
Andrew Lenharth51b8d542005-11-11 16:47:30 +0000[diff] [blame]5201</pre>
5202
5203<h5>Overview:</h5>
5204
5205
5206<p>
5207The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
5208counter register (or similar low latency, high accuracy clocks) on those targets
5209that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
5210As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
5211should only be used for small timings.
5212</p>
5213
5214<h5>Semantics:</h5>
5215
5216<p>
5217When directly supported, reading the cycle counter should not modify any memory.
5218Implementations are allowed to either return a application specific value or a
5219system wide value. On backends without support, this is lowered to a constant 0.
5220</p>
5221
5222</div>
5223
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5224<!-- ======================================================================= -->
5225<div class="doc_subsection">
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5226 <a name="int_libc">Standard C Library Intrinsics</a>
5227</div>
5228
5229<div class="doc_text">
5230<p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5231LLVM provides intrinsics for a few important standard C library functions.
5232These intrinsics allow source-language front-ends to pass information about the
5233alignment of the pointer arguments to the code generator, providing opportunity
5234for more efficient code generation.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5235</p>
5236
5237</div>
5238
5239<!-- _______________________________________________________________________ -->
5240<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5241 <a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5242</div>
5243
5244<div class="doc_text">
5245
5246<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5247<p>This is an overloaded intrinsic. You can use llvm.memcpy on any integer bit
5248width. Not all targets support all bit widths however.</p>
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5249<pre>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5250 declare void @llvm.memcpy.i8(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
5251 i8 &lt;len&gt;, i32 &lt;align&gt;)
5252 declare void @llvm.memcpy.i16(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
5253 i16 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5254 declare void @llvm.memcpy.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5255 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5256 declare void @llvm.memcpy.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5257 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5258</pre>
5259
5260<h5>Overview:</h5>
5261
5262<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5263The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5264location to the destination location.
5265</p>
5266
5267<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5268Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
5269intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5270</p>
5271
5272<h5>Arguments:</h5>
5273
5274<p>
5275The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5276the source. The third argument is an integer argument
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5277specifying the number of bytes to copy, and the fourth argument is the alignment
5278of the source and destination locations.
5279</p>
5280
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]5281<p>
5282If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattnerf0afc2c2006-03-04 00:02:10 +0000[diff] [blame]5283the caller guarantees that both the source and destination pointers are aligned
5284to that boundary.
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]5285</p>
5286
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5287<h5>Semantics:</h5>
5288
5289<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5290The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner33aec9e2004-02-12 17:01:32 +0000[diff] [blame]5291location to the destination location, which are not allowed to overlap. It
5292copies "len" bytes of memory over. If the argument is known to be aligned to
5293some boundary, this can be specified as the fourth argument, otherwise it should
5294be set to 0 or 1.
5295</p>
5296</div>
5297
5298
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5299<!-- _______________________________________________________________________ -->
5300<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5301 <a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5302</div>
5303
5304<div class="doc_text">
5305
5306<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5307<p>This is an overloaded intrinsic. You can use llvm.memmove on any integer bit
5308width. Not all targets support all bit widths however.</p>
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5309<pre>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5310 declare void @llvm.memmove.i8(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
5311 i8 &lt;len&gt;, i32 &lt;align&gt;)
5312 declare void @llvm.memmove.i16(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
5313 i16 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5314 declare void @llvm.memmove.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5315 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5316 declare void @llvm.memmove.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5317 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5318</pre>
5319
5320<h5>Overview:</h5>
5321
5322<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5323The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the source
5324location to the destination location. It is similar to the
Chris Lattner4b2cbcf2008-01-06 19:51:52 +0000[diff] [blame]5325'<tt>llvm.memcpy</tt>' intrinsic but allows the two memory locations to overlap.
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5326</p>
5327
5328<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5329Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
5330intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5331</p>
5332
5333<h5>Arguments:</h5>
5334
5335<p>
5336The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5337the source. The third argument is an integer argument
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5338specifying the number of bytes to copy, and the fourth argument is the alignment
5339of the source and destination locations.
5340</p>
5341
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]5342<p>
5343If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattnerf0afc2c2006-03-04 00:02:10 +0000[diff] [blame]5344the caller guarantees that the source and destination pointers are aligned to
5345that boundary.
Chris Lattner3301ced2004-02-12 21:18:15 +0000[diff] [blame]5346</p>
5347
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5348<h5>Semantics:</h5>
5349
5350<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5351The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner0eb51b42004-02-12 18:10:10 +0000[diff] [blame]5352location to the destination location, which may overlap. It
5353copies "len" bytes of memory over. If the argument is known to be aligned to
5354some boundary, this can be specified as the fourth argument, otherwise it should
5355be set to 0 or 1.
5356</p>
5357</div>
5358
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5359
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5360<!-- _______________________________________________________________________ -->
5361<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5362 <a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5363</div>
5364
5365<div class="doc_text">
5366
5367<h5>Syntax:</h5>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5368<p>This is an overloaded intrinsic. You can use llvm.memset on any integer bit
5369width. Not all targets support all bit widths however.</p>
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5370<pre>
Chris Lattner824b9582008-11-21 16:42:48 +0000[diff] [blame]5371 declare void @llvm.memset.i8(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
5372 i8 &lt;len&gt;, i32 &lt;align&gt;)
5373 declare void @llvm.memset.i16(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
5374 i16 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5375 declare void @llvm.memset.i32(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5376 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5377 declare void @llvm.memset.i64(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5378 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5379</pre>
5380
5381<h5>Overview:</h5>
5382
5383<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5384The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a particular
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5385byte value.
5386</p>
5387
5388<p>
5389Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
5390does not return a value, and takes an extra alignment argument.
5391</p>
5392
5393<h5>Arguments:</h5>
5394
5395<p>
5396The first argument is a pointer to the destination to fill, the second is the
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5397byte value to fill it with, the third argument is an integer
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5398argument specifying the number of bytes to fill, and the fourth argument is the
5399known alignment of destination location.
5400</p>
5401
5402<p>
5403If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattnerf0afc2c2006-03-04 00:02:10 +0000[diff] [blame]5404the caller guarantees that the destination pointer is aligned to that boundary.
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5405</p>
5406
5407<h5>Semantics:</h5>
5408
5409<p>
Chris Lattner5b310c32006-03-03 00:07:20 +0000[diff] [blame]5410The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting at
5411the
Chris Lattner10610642004-02-14 04:08:35 +0000[diff] [blame]5412destination location. If the argument is known to be aligned to some boundary,
5413this can be specified as the fourth argument, otherwise it should be set to 0 or
54141.
5415</p>
5416</div>
5417
5418
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]5419<!-- _______________________________________________________________________ -->
5420<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5421 <a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5422</div>
5423
5424<div class="doc_text">
5425
5426<h5>Syntax:</h5>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]5427<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5428floating point or vector of floating point type. Not all targets support all
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5429types however.</p>
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5430<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]5431 declare float @llvm.sqrt.f32(float %Val)
5432 declare double @llvm.sqrt.f64(double %Val)
5433 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
5434 declare fp128 @llvm.sqrt.f128(fp128 %Val)
5435 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5436</pre>
5437
5438<h5>Overview:</h5>
5439
5440<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]5441The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5442returning the same value as the libm '<tt>sqrt</tt>' functions would. Unlike
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5443<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
Chris Lattner103e2d72008-01-29 07:00:44 +0000[diff] [blame]5444negative numbers other than -0.0 (which allows for better optimization, because
5445there is no need to worry about errno being set). <tt>llvm.sqrt(-0.0)</tt> is
5446defined to return -0.0 like IEEE sqrt.
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5447</p>
5448
5449<h5>Arguments:</h5>
5450
5451<p>
5452The argument and return value are floating point numbers of the same type.
5453</p>
5454
5455<h5>Semantics:</h5>
5456
5457<p>
Dan Gohmand6257fe2007-07-16 14:37:41 +0000[diff] [blame]5458This function returns the sqrt of the specified operand if it is a nonnegative
Chris Lattnera4d74142005-07-21 01:29:16 +0000[diff] [blame]5459floating point number.
5460</p>
5461</div>
5462
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5463<!-- _______________________________________________________________________ -->
5464<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5465 <a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5466</div>
5467
5468<div class="doc_text">
5469
5470<h5>Syntax:</h5>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]5471<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5472floating point or vector of floating point type. Not all targets support all
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5473types however.</p>
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5474<pre>
Dale Johannesen408f9c12007-10-02 17:47:38 +0000[diff] [blame]5475 declare float @llvm.powi.f32(float %Val, i32 %power)
5476 declare double @llvm.powi.f64(double %Val, i32 %power)
5477 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
5478 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
5479 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5480</pre>
5481
5482<h5>Overview:</h5>
5483
5484<p>
5485The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
5486specified (positive or negative) power. The order of evaluation of
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5487multiplications is not defined. When a vector of floating point type is
5488used, the second argument remains a scalar integer value.
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5489</p>
5490
5491<h5>Arguments:</h5>
5492
5493<p>
5494The second argument is an integer power, and the first is a value to raise to
5495that power.
5496</p>
5497
5498<h5>Semantics:</h5>
5499
5500<p>
5501This function returns the first value raised to the second power with an
5502unspecified sequence of rounding operations.</p>
5503</div>
5504
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5505<!-- _______________________________________________________________________ -->
5506<div class="doc_subsubsection">
5507 <a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
5508</div>
5509
5510<div class="doc_text">
5511
5512<h5>Syntax:</h5>
5513<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
5514floating point or vector of floating point type. Not all targets support all
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5515types however.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5516<pre>
5517 declare float @llvm.sin.f32(float %Val)
5518 declare double @llvm.sin.f64(double %Val)
5519 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
5520 declare fp128 @llvm.sin.f128(fp128 %Val)
5521 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
5522</pre>
5523
5524<h5>Overview:</h5>
5525
5526<p>
5527The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.
5528</p>
5529
5530<h5>Arguments:</h5>
5531
5532<p>
5533The argument and return value are floating point numbers of the same type.
5534</p>
5535
5536<h5>Semantics:</h5>
5537
5538<p>
5539This function returns the sine of the specified operand, returning the
5540same values as the libm <tt>sin</tt> functions would, and handles error
Dan Gohmanba83b7e2007-10-17 18:05:13 +0000[diff] [blame]5541conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5542</div>
5543
5544<!-- _______________________________________________________________________ -->
5545<div class="doc_subsubsection">
5546 <a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
5547</div>
5548
5549<div class="doc_text">
5550
5551<h5>Syntax:</h5>
5552<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
5553floating point or vector of floating point type. Not all targets support all
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5554types however.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5555<pre>
5556 declare float @llvm.cos.f32(float %Val)
5557 declare double @llvm.cos.f64(double %Val)
5558 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
5559 declare fp128 @llvm.cos.f128(fp128 %Val)
5560 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
5561</pre>
5562
5563<h5>Overview:</h5>
5564
5565<p>
5566The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.
5567</p>
5568
5569<h5>Arguments:</h5>
5570
5571<p>
5572The argument and return value are floating point numbers of the same type.
5573</p>
5574
5575<h5>Semantics:</h5>
5576
5577<p>
5578This function returns the cosine of the specified operand, returning the
5579same values as the libm <tt>cos</tt> functions would, and handles error
Dan Gohmanba83b7e2007-10-17 18:05:13 +0000[diff] [blame]5580conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5581</div>
5582
5583<!-- _______________________________________________________________________ -->
5584<div class="doc_subsubsection">
5585 <a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
5586</div>
5587
5588<div class="doc_text">
5589
5590<h5>Syntax:</h5>
5591<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
5592floating point or vector of floating point type. Not all targets support all
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5593types however.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5594<pre>
5595 declare float @llvm.pow.f32(float %Val, float %Power)
5596 declare double @llvm.pow.f64(double %Val, double %Power)
5597 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
5598 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
5599 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
5600</pre>
5601
5602<h5>Overview:</h5>
5603
5604<p>
5605The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
5606specified (positive or negative) power.
5607</p>
5608
5609<h5>Arguments:</h5>
5610
5611<p>
5612The second argument is a floating point power, and the first is a value to
5613raise to that power.
5614</p>
5615
5616<h5>Semantics:</h5>
5617
5618<p>
5619This function returns the first value raised to the second power,
5620returning the
5621same values as the libm <tt>pow</tt> functions would, and handles error
Dan Gohmanba83b7e2007-10-17 18:05:13 +0000[diff] [blame]5622conditions in the same way.</p>
Dan Gohman91c284c2007-10-15 20:30:11 +0000[diff] [blame]5623</div>
5624
Chris Lattnerf4d252d2006-09-08 06:34:02 +0000[diff] [blame]5625
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5626<!-- ======================================================================= -->
5627<div class="doc_subsection">
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5628 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5629</div>
5630
5631<div class="doc_text">
5632<p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5633LLVM provides intrinsics for a few important bit manipulation operations.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5634These allow efficient code generation for some algorithms.
5635</p>
5636
5637</div>
5638
5639<!-- _______________________________________________________________________ -->
5640<div class="doc_subsubsection">
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5641 <a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5642</div>
5643
5644<div class="doc_text">
5645
5646<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5647<p>This is an overloaded intrinsic function. You can use bswap on any integer
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5648type that is an even number of bytes (i.e. BitWidth % 16 == 0).</p>
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5649<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5650 declare i16 @llvm.bswap.i16(i16 &lt;id&gt;)
5651 declare i32 @llvm.bswap.i32(i32 &lt;id&gt;)
5652 declare i64 @llvm.bswap.i64(i64 &lt;id&gt;)
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5653</pre>
5654
5655<h5>Overview:</h5>
5656
5657<p>
Reid Spencer338ea092007-04-02 02:25:19 +0000[diff] [blame]5658The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5659values with an even number of bytes (positive multiple of 16 bits). These are
5660useful for performing operations on data that is not in the target's native
5661byte order.
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5662</p>
5663
5664<h5>Semantics:</h5>
5665
5666<p>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5667The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5668and low byte of the input i16 swapped. Similarly, the <tt>llvm.bswap.i32</tt>
5669intrinsic returns an i32 value that has the four bytes of the input i32
5670swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the returned
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5671i32 will have its bytes in 3, 2, 1, 0 order. The <tt>llvm.bswap.i48</tt>,
5672<tt>llvm.bswap.i64</tt> and other intrinsics extend this concept to
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5673additional even-byte lengths (6 bytes, 8 bytes and more, respectively).
Nate Begeman7e36c472006-01-13 23:26:38 +0000[diff] [blame]5674</p>
5675
5676</div>
5677
5678<!-- _______________________________________________________________________ -->
5679<div class="doc_subsubsection">
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]5680 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5681</div>
5682
5683<div class="doc_text">
5684
5685<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5686<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5687width. Not all targets support all bit widths however.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5688<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5689 declare i8 @llvm.ctpop.i8 (i8 &lt;src&gt;)
5690 declare i16 @llvm.ctpop.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5691 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5692 declare i64 @llvm.ctpop.i64(i64 &lt;src&gt;)
5693 declare i256 @llvm.ctpop.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5694</pre>
5695
5696<h5>Overview:</h5>
5697
5698<p>
Chris Lattnerec6cb612006-01-16 22:38:59 +0000[diff] [blame]5699The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set in a
5700value.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5701</p>
5702
5703<h5>Arguments:</h5>
5704
5705<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]5706The only argument is the value to be counted. The argument may be of any
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]5707integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5708</p>
5709
5710<h5>Semantics:</h5>
5711
5712<p>
5713The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
5714</p>
5715</div>
5716
5717<!-- _______________________________________________________________________ -->
5718<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]5719 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5720</div>
5721
5722<div class="doc_text">
5723
5724<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5725<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5726integer bit width. Not all targets support all bit widths however.</p>
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5727<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5728 declare i8 @llvm.ctlz.i8 (i8 &lt;src&gt;)
5729 declare i16 @llvm.ctlz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5730 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5731 declare i64 @llvm.ctlz.i64(i64 &lt;src&gt;)
5732 declare i256 @llvm.ctlz.i256(i256 &lt;src&gt;)
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5733</pre>
5734
5735<h5>Overview:</h5>
5736
5737<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]5738The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
5739leading zeros in a variable.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5740</p>
5741
5742<h5>Arguments:</h5>
5743
5744<p>
Chris Lattnercfe6b372005-05-07 01:46:40 +0000[diff] [blame]5745The only argument is the value to be counted. The argument may be of any
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]5746integer type. The return type must match the argument type.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5747</p>
5748
5749<h5>Semantics:</h5>
5750
5751<p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5752The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
5753in a variable. If the src == 0 then the result is the size in bits of the type
Reid Spencerca86e162006-12-31 07:07:53 +0000[diff] [blame]5754of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.
Andrew Lenharthec370fd2005-05-03 18:01:48 +0000[diff] [blame]5755</p>
5756</div>
Chris Lattner32006282004-06-11 02:28:03 +0000[diff] [blame]5757
5758
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5759
5760<!-- _______________________________________________________________________ -->
5761<div class="doc_subsubsection">
Chris Lattner8a886be2006-01-16 22:34:14 +0000[diff] [blame]5762 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5763</div>
5764
5765<div class="doc_text">
5766
5767<h5>Syntax:</h5>
Reid Spencer409e28f2007-04-01 08:04:23 +0000[diff] [blame]5768<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5769integer bit width. Not all targets support all bit widths however.</p>
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5770<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5771 declare i8 @llvm.cttz.i8 (i8 &lt;src&gt;)
5772 declare i16 @llvm.cttz.i16(i16 &lt;src&gt;)
Anton Korobeynikovec43a062007-03-22 00:02:17 +0000[diff] [blame]5773 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5774 declare i64 @llvm.cttz.i64(i64 &lt;src&gt;)
5775 declare i256 @llvm.cttz.i256(i256 &lt;src&gt;)
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5776</pre>
5777
5778<h5>Overview:</h5>
5779
5780<p>
Reid Spencer0b118202006-01-16 21:12:35 +0000[diff] [blame]5781The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
5782trailing zeros.
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5783</p>
5784
5785<h5>Arguments:</h5>
5786
5787<p>
5788The only argument is the value to be counted. The argument may be of any
Reid Spencera5173382007-01-04 16:43:23 +0000[diff] [blame]5789integer type. The return type must match the argument type.
Chris Lattnereff29ab2005-05-15 19:39:26 +0000[diff] [blame]5790</p>
5791
5792<h5>Semantics:</h5>
5793
5794<p>
5795The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
5796in a variable. If the src == 0 then the result is the size in bits of the type
5797of src. For example, <tt>llvm.cttz(2) = 1</tt>.
5798</p>
5799</div>
5800
Reid Spencer497d93e2007-04-01 08:27:01 +0000[diff] [blame]5801<!-- _______________________________________________________________________ -->
5802<div class="doc_subsubsection">
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5803 <a name="int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic</a>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5804</div>
5805
5806<div class="doc_text">
5807
5808<h5>Syntax:</h5>
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5809<p>This is an overloaded intrinsic. You can use <tt>llvm.part.select</tt>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5810on any integer bit width.</p>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5811<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5812 declare i17 @llvm.part.select.i17 (i17 %val, i32 %loBit, i32 %hiBit)
5813 declare i29 @llvm.part.select.i29 (i29 %val, i32 %loBit, i32 %hiBit)
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5814</pre>
5815
5816<h5>Overview:</h5>
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5817<p>The '<tt>llvm.part.select</tt>' family of intrinsic functions selects a
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5818range of bits from an integer value and returns them in the same bit width as
5819the original value.</p>
5820
5821<h5>Arguments:</h5>
5822<p>The first argument, <tt>%val</tt> and the result may be integer types of
5823any bit width but they must have the same bit width. The second and third
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5824arguments must be <tt>i32</tt> type since they specify only a bit index.</p>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5825
5826<h5>Semantics:</h5>
Reid Spencerbeacf662007-04-10 02:51:31 +0000[diff] [blame]5827<p>The operation of the '<tt>llvm.part.select</tt>' intrinsic has two modes
Reid Spencera3e435f2007-04-04 02:42:35 +0000[diff] [blame]5828of operation: forwards and reverse. If <tt>%loBit</tt> is greater than
5829<tt>%hiBits</tt> then the intrinsic operates in reverse mode. Otherwise it
5830operates in forward mode.</p>
5831<p>In forward mode, this intrinsic is the equivalent of shifting <tt>%val</tt>
5832right by <tt>%loBit</tt> bits and then ANDing it with a mask with
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5833only the <tt>%hiBit - %loBit</tt> bits set, as follows:</p>
5834<ol>
5835 <li>The <tt>%val</tt> is shifted right (LSHR) by the number of bits specified
5836 by <tt>%loBits</tt>. This normalizes the value to the low order bits.</li>
5837 <li>The <tt>%loBits</tt> value is subtracted from the <tt>%hiBits</tt> value
5838 to determine the number of bits to retain.</li>
5839 <li>A mask of the retained bits is created by shifting a -1 value.</li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5840 <li>The mask is ANDed with <tt>%val</tt> to produce the result.</li>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5841</ol>
Reid Spencerd6a85b52007-05-14 16:14:57 +0000[diff] [blame]5842<p>In reverse mode, a similar computation is made except that the bits are
5843returned in the reverse order. So, for example, if <tt>X</tt> has the value
5844<tt>i16 0x0ACF (101011001111)</tt> and we apply
5845<tt>part.select(i16 X, 8, 3)</tt> to it, we get back the value
5846<tt>i16 0x0026 (000000100110)</tt>.</p>
Reid Spencera13ba7d2007-04-01 19:00:37 +0000[diff] [blame]5847</div>
5848
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5849<div class="doc_subsubsection">
5850 <a name="int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic</a>
5851</div>
5852
5853<div class="doc_text">
5854
5855<h5>Syntax:</h5>
5856<p>This is an overloaded intrinsic. You can use <tt>llvm.part.set</tt>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5857on any integer bit width.</p>
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5858<pre>
Chandler Carruth69940402007-08-04 01:51:18 +0000[diff] [blame]5859 declare i17 @llvm.part.set.i17.i9 (i17 %val, i9 %repl, i32 %lo, i32 %hi)
5860 declare i29 @llvm.part.set.i29.i9 (i29 %val, i9 %repl, i32 %lo, i32 %hi)
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5861</pre>
5862
5863<h5>Overview:</h5>
5864<p>The '<tt>llvm.part.set</tt>' family of intrinsic functions replaces a range
5865of bits in an integer value with another integer value. It returns the integer
5866with the replaced bits.</p>
5867
5868<h5>Arguments:</h5>
5869<p>The first argument, <tt>%val</tt> and the result may be integer types of
5870any bit width but they must have the same bit width. <tt>%val</tt> is the value
5871whose bits will be replaced. The second argument, <tt>%repl</tt> may be an
5872integer of any bit width. The third and fourth arguments must be <tt>i32</tt>
5873type since they specify only a bit index.</p>
5874
5875<h5>Semantics:</h5>
5876<p>The operation of the '<tt>llvm.part.set</tt>' intrinsic has two modes
5877of operation: forwards and reverse. If <tt>%lo</tt> is greater than
5878<tt>%hi</tt> then the intrinsic operates in reverse mode. Otherwise it
5879operates in forward mode.</p>
5880<p>For both modes, the <tt>%repl</tt> value is prepared for use by either
5881truncating it down to the size of the replacement area or zero extending it
5882up to that size.</p>
5883<p>In forward mode, the bits between <tt>%lo</tt> and <tt>%hi</tt> (inclusive)
5884are replaced with corresponding bits from <tt>%repl</tt>. That is the 0th bit
5885in <tt>%repl</tt> replaces the <tt>%lo</tt>th bit in <tt>%val</tt> and etc. up
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5886to the <tt>%hi</tt>th bit.</p>
Reid Spencerc6749c42007-05-14 16:50:20 +0000[diff] [blame]5887<p>In reverse mode, a similar computation is made except that the bits are
5888reversed. That is, the <tt>0</tt>th bit in <tt>%repl</tt> replaces the
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]5889<tt>%hi</tt> bit in <tt>%val</tt> and etc. down to the <tt>%lo</tt>th bit.</p>
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5890<h5>Examples:</h5>
5891<pre>
Reid Spencerf0dbf642007-04-12 01:03:03 +0000[diff] [blame]5892 llvm.part.set(0xFFFF, 0, 4, 7) -&gt; 0xFF0F
Reid Spencerc6749c42007-05-14 16:50:20 +0000[diff] [blame]5893 llvm.part.set(0xFFFF, 0, 7, 4) -&gt; 0xFF0F
5894 llvm.part.set(0xFFFF, 1, 7, 4) -&gt; 0xFF8F
5895 llvm.part.set(0xFFFF, F, 8, 3) -&gt; 0xFFE7
Reid Spencerf0dbf642007-04-12 01:03:03 +0000[diff] [blame]5896 llvm.part.set(0xFFFF, 0, 3, 8) -&gt; 0xFE07
Reid Spencerc8910842007-04-11 23:49:50 +0000[diff] [blame]5897</pre>
Reid Spencerf86037f2007-04-11 23:23:49 +0000[diff] [blame]5898</div>
5899
Chris Lattner8ff75902004-01-06 05:31:32 +0000[diff] [blame]5900<!-- ======================================================================= -->
5901<div class="doc_subsection">
5902 <a name="int_debugger">Debugger Intrinsics</a>
5903</div>
5904
5905<div class="doc_text">
5906<p>
5907The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
5908are described in the <a
5909href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
5910Debugging</a> document.
5911</p>
5912</div>
5913
5914
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]5915<!-- ======================================================================= -->
5916<div class="doc_subsection">
5917 <a name="int_eh">Exception Handling Intrinsics</a>
5918</div>
5919
5920<div class="doc_text">
5921<p> The LLVM exception handling intrinsics (which all start with
5922<tt>llvm.eh.</tt> prefix), are described in the <a
5923href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
5924Handling</a> document. </p>
5925</div>
5926
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]5927<!-- ======================================================================= -->
5928<div class="doc_subsection">
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5929 <a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5930</div>
5931
5932<div class="doc_text">
5933<p>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5934 This intrinsic makes it possible to excise one parameter, marked with
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5935 the <tt>nest</tt> attribute, from a function. The result is a callable
5936 function pointer lacking the nest parameter - the caller does not need
5937 to provide a value for it. Instead, the value to use is stored in
5938 advance in a "trampoline", a block of memory usually allocated
5939 on the stack, which also contains code to splice the nest value into the
5940 argument list. This is used to implement the GCC nested function address
5941 extension.
5942</p>
5943<p>
5944 For example, if the function is
5945 <tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
Bill Wendling03295ca2007-09-22 09:23:55 +0000[diff] [blame]5946 pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as follows:</p>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5947<pre>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5948 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
5949 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
5950 %p = call i8* @llvm.init.trampoline( i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8*), i8* %nval )
5951 %fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5952</pre>
Bill Wendling03295ca2007-09-22 09:23:55 +0000[diff] [blame]5953 <p>The call <tt>%val = call i32 %fp( i32 %x, i32 %y )</tt> is then equivalent
5954 to <tt>%val = call i32 %f( i8* %nval, i32 %x, i32 %y )</tt>.</p>
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5955</div>
5956
5957<!-- _______________________________________________________________________ -->
5958<div class="doc_subsubsection">
5959 <a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
5960</div>
5961<div class="doc_text">
5962<h5>Syntax:</h5>
5963<pre>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5964declare 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]5965</pre>
5966<h5>Overview:</h5>
5967<p>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5968 This fills the memory pointed to by <tt>tramp</tt> with code
5969 and returns a function pointer suitable for executing it.
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5970</p>
5971<h5>Arguments:</h5>
5972<p>
5973 The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
5974 pointers. The <tt>tramp</tt> argument must point to a sufficiently large
5975 and sufficiently aligned block of memory; this memory is written to by the
Duncan Sandsc00c2ba2007-08-22 23:39:54 +0000[diff] [blame]5976 intrinsic. Note that the size and the alignment are target-specific - LLVM
5977 currently provides no portable way of determining them, so a front-end that
5978 generates this intrinsic needs to have some target-specific knowledge.
5979 The <tt>func</tt> argument must hold a function bitcast to an <tt>i8*</tt>.
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5980</p>
5981<h5>Semantics:</h5>
5982<p>
5983 The block of memory pointed to by <tt>tramp</tt> is filled with target
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5984 dependent code, turning it into a function. A pointer to this function is
5985 returned, but needs to be bitcast to an
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5986 <a href="#int_trampoline">appropriate function pointer type</a>
Duncan Sandsf7331b32007-09-11 14:10:23 +0000[diff] [blame]5987 before being called. The new function's signature is the same as that of
5988 <tt>func</tt> with any arguments marked with the <tt>nest</tt> attribute
5989 removed. At most one such <tt>nest</tt> argument is allowed, and it must be
5990 of pointer type. Calling the new function is equivalent to calling
5991 <tt>func</tt> with the same argument list, but with <tt>nval</tt> used for the
5992 missing <tt>nest</tt> argument. If, after calling
5993 <tt>llvm.init.trampoline</tt>, the memory pointed to by <tt>tramp</tt> is
5994 modified, then the effect of any later call to the returned function pointer is
5995 undefined.
Duncan Sands36397f52007-07-27 12:58:54 +0000[diff] [blame]5996</p>
5997</div>
5998
5999<!-- ======================================================================= -->
6000<div class="doc_subsection">
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6001 <a name="int_atomics">Atomic Operations and Synchronization Intrinsics</a>
6002</div>
6003
6004<div class="doc_text">
6005<p>
6006 These intrinsic functions expand the "universal IR" of LLVM to represent
6007 hardware constructs for atomic operations and memory synchronization. This
6008 provides an interface to the hardware, not an interface to the programmer. It
Chris Lattnerd3eda892008-08-05 18:29:16 +0000[diff] [blame]6009 is aimed at a low enough level to allow any programming models or APIs
6010 (Application Programming Interfaces) which
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6011 need atomic behaviors to map cleanly onto it. It is also modeled primarily on
6012 hardware behavior. Just as hardware provides a "universal IR" for source
6013 languages, it also provides a starting point for developing a "universal"
6014 atomic operation and synchronization IR.
6015</p>
6016<p>
6017 These do <em>not</em> form an API such as high-level threading libraries,
6018 software transaction memory systems, atomic primitives, and intrinsic
6019 functions as found in BSD, GNU libc, atomic_ops, APR, and other system and
6020 application libraries. The hardware interface provided by LLVM should allow
6021 a clean implementation of all of these APIs and parallel programming models.
6022 No one model or paradigm should be selected above others unless the hardware
6023 itself ubiquitously does so.
6024
6025</p>
6026</div>
6027
6028<!-- _______________________________________________________________________ -->
6029<div class="doc_subsubsection">
6030 <a name="int_memory_barrier">'<tt>llvm.memory.barrier</tt>' Intrinsic</a>
6031</div>
6032<div class="doc_text">
6033<h5>Syntax:</h5>
6034<pre>
6035declare void @llvm.memory.barrier( i1 &lt;ll&gt;, i1 &lt;ls&gt;, i1 &lt;sl&gt;, i1 &lt;ss&gt;,
6036i1 &lt;device&gt; )
6037
6038</pre>
6039<h5>Overview:</h5>
6040<p>
6041 The <tt>llvm.memory.barrier</tt> intrinsic guarantees ordering between
6042 specific pairs of memory access types.
6043</p>
6044<h5>Arguments:</h5>
6045<p>
6046 The <tt>llvm.memory.barrier</tt> intrinsic requires five boolean arguments.
6047 The first four arguments enables a specific barrier as listed below. The fith
6048 argument specifies that the barrier applies to io or device or uncached memory.
6049
6050</p>
6051 <ul>
6052 <li><tt>ll</tt>: load-load barrier</li>
6053 <li><tt>ls</tt>: load-store barrier</li>
6054 <li><tt>sl</tt>: store-load barrier</li>
6055 <li><tt>ss</tt>: store-store barrier</li>
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]6056 <li><tt>device</tt>: barrier applies to device and uncached memory also.</li>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6057 </ul>
6058<h5>Semantics:</h5>
6059<p>
6060 This intrinsic causes the system to enforce some ordering constraints upon
6061 the loads and stores of the program. This barrier does not indicate
6062 <em>when</em> any events will occur, it only enforces an <em>order</em> in
6063 which they occur. For any of the specified pairs of load and store operations
6064 (f.ex. load-load, or store-load), all of the first operations preceding the
6065 barrier will complete before any of the second operations succeeding the
6066 barrier begin. Specifically the semantics for each pairing is as follows:
6067</p>
6068 <ul>
6069 <li><tt>ll</tt>: All loads before the barrier must complete before any load
6070 after the barrier begins.</li>
6071
6072 <li><tt>ls</tt>: All loads before the barrier must complete before any
6073 store after the barrier begins.</li>
6074 <li><tt>ss</tt>: All stores before the barrier must complete before any
6075 store after the barrier begins.</li>
6076 <li><tt>sl</tt>: All stores before the barrier must complete before any
6077 load after the barrier begins.</li>
6078 </ul>
6079<p>
6080 These semantics are applied with a logical "and" behavior when more than one
6081 is enabled in a single memory barrier intrinsic.
6082</p>
6083<p>
6084 Backends may implement stronger barriers than those requested when they do not
6085 support as fine grained a barrier as requested. Some architectures do not
6086 need all types of barriers and on such architectures, these become noops.
6087</p>
6088<h5>Example:</h5>
6089<pre>
6090%ptr = malloc i32
6091 store i32 4, %ptr
6092
6093%result1 = load i32* %ptr <i>; yields {i32}:result1 = 4</i>
6094 call void @llvm.memory.barrier( i1 false, i1 true, i1 false, i1 false )
6095 <i>; guarantee the above finishes</i>
6096 store i32 8, %ptr <i>; before this begins</i>
6097</pre>
6098</div>
6099
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6100<!-- _______________________________________________________________________ -->
6101<div class="doc_subsubsection">
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6102 <a name="int_atomic_cmp_swap">'<tt>llvm.atomic.cmp.swap.*</tt>' Intrinsic</a>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6103</div>
6104<div class="doc_text">
6105<h5>Syntax:</h5>
6106<p>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6107 This is an overloaded intrinsic. You can use <tt>llvm.atomic.cmp.swap</tt> on
6108 any integer bit width and for different address spaces. Not all targets
6109 support all bit widths however.</p>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6110
6111<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6112declare i8 @llvm.atomic.cmp.swap.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;cmp&gt;, i8 &lt;val&gt; )
6113declare i16 @llvm.atomic.cmp.swap.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;cmp&gt;, i16 &lt;val&gt; )
6114declare i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;cmp&gt;, i32 &lt;val&gt; )
6115declare 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]6116
6117</pre>
6118<h5>Overview:</h5>
6119<p>
6120 This loads a value in memory and compares it to a given value. If they are
6121 equal, it stores a new value into the memory.
6122</p>
6123<h5>Arguments:</h5>
6124<p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6125 The <tt>llvm.atomic.cmp.swap</tt> intrinsic takes three arguments. The result as
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6126 well as both <tt>cmp</tt> and <tt>val</tt> must be integer values with the
6127 same bit width. The <tt>ptr</tt> argument must be a pointer to a value of
6128 this integer type. While any bit width integer may be used, targets may only
6129 lower representations they support in hardware.
6130
6131</p>
6132<h5>Semantics:</h5>
6133<p>
6134 This entire intrinsic must be executed atomically. It first loads the value
6135 in memory pointed to by <tt>ptr</tt> and compares it with the value
6136 <tt>cmp</tt>. If they are equal, <tt>val</tt> is stored into the memory. The
6137 loaded value is yielded in all cases. This provides the equivalent of an
6138 atomic compare-and-swap operation within the SSA framework.
6139</p>
6140<h5>Examples:</h5>
6141
6142<pre>
6143%ptr = malloc i32
6144 store i32 4, %ptr
6145
6146%val1 = add i32 4, 4
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6147%result1 = call i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* %ptr, i32 4, %val1 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6148 <i>; yields {i32}:result1 = 4</i>
6149%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
6150%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
6151
6152%val2 = add i32 1, 1
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6153%result2 = call i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* %ptr, i32 5, %val2 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6154 <i>; yields {i32}:result2 = 8</i>
6155%stored2 = icmp eq i32 %result2, 5 <i>; yields {i1}:stored2 = false</i>
6156
6157%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 8</i>
6158</pre>
6159</div>
6160
6161<!-- _______________________________________________________________________ -->
6162<div class="doc_subsubsection">
6163 <a name="int_atomic_swap">'<tt>llvm.atomic.swap.*</tt>' Intrinsic</a>
6164</div>
6165<div class="doc_text">
6166<h5>Syntax:</h5>
6167
6168<p>
6169 This is an overloaded intrinsic. You can use <tt>llvm.atomic.swap</tt> on any
6170 integer bit width. Not all targets support all bit widths however.</p>
6171<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6172declare i8 @llvm.atomic.swap.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;val&gt; )
6173declare i16 @llvm.atomic.swap.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;val&gt; )
6174declare i32 @llvm.atomic.swap.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;val&gt; )
6175declare i64 @llvm.atomic.swap.i64.p0i64( i64* &lt;ptr&gt;, i64 &lt;val&gt; )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6176
6177</pre>
6178<h5>Overview:</h5>
6179<p>
6180 This intrinsic loads the value stored in memory at <tt>ptr</tt> and yields
6181 the value from memory. It then stores the value in <tt>val</tt> in the memory
6182 at <tt>ptr</tt>.
6183</p>
6184<h5>Arguments:</h5>
6185
6186<p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6187 The <tt>llvm.atomic.swap</tt> intrinsic takes two arguments. Both the
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6188 <tt>val</tt> argument and the result must be integers of the same bit width.
6189 The first argument, <tt>ptr</tt>, must be a pointer to a value of this
6190 integer type. The targets may only lower integer representations they
6191 support.
6192</p>
6193<h5>Semantics:</h5>
6194<p>
6195 This intrinsic loads the value pointed to by <tt>ptr</tt>, yields it, and
6196 stores <tt>val</tt> back into <tt>ptr</tt> atomically. This provides the
6197 equivalent of an atomic swap operation within the SSA framework.
6198
6199</p>
6200<h5>Examples:</h5>
6201<pre>
6202%ptr = malloc i32
6203 store i32 4, %ptr
6204
6205%val1 = add i32 4, 4
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6206%result1 = call i32 @llvm.atomic.swap.i32.p0i32( i32* %ptr, i32 %val1 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6207 <i>; yields {i32}:result1 = 4</i>
6208%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
6209%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
6210
6211%val2 = add i32 1, 1
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6212%result2 = call i32 @llvm.atomic.swap.i32.p0i32( i32* %ptr, i32 %val2 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6213 <i>; yields {i32}:result2 = 8</i>
6214
6215%stored2 = icmp eq i32 %result2, 8 <i>; yields {i1}:stored2 = true</i>
6216%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 2</i>
6217</pre>
6218</div>
6219
6220<!-- _______________________________________________________________________ -->
6221<div class="doc_subsubsection">
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6222 <a name="int_atomic_load_add">'<tt>llvm.atomic.load.add.*</tt>' Intrinsic</a>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6223
6224</div>
6225<div class="doc_text">
6226<h5>Syntax:</h5>
6227<p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6228 This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.add</tt> on any
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6229 integer bit width. Not all targets support all bit widths however.</p>
6230<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6231declare i8 @llvm.atomic.load.add.i8..p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6232declare i16 @llvm.atomic.load.add.i16..p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6233declare i32 @llvm.atomic.load.add.i32..p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6234declare 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]6235
6236</pre>
6237<h5>Overview:</h5>
6238<p>
6239 This intrinsic adds <tt>delta</tt> to the value stored in memory at
6240 <tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.
6241</p>
6242<h5>Arguments:</h5>
6243<p>
6244
6245 The intrinsic takes two arguments, the first a pointer to an integer value
6246 and the second an integer value. The result is also an integer value. These
6247 integer types can have any bit width, but they must all have the same bit
6248 width. The targets may only lower integer representations they support.
6249</p>
6250<h5>Semantics:</h5>
6251<p>
6252 This intrinsic does a series of operations atomically. It first loads the
6253 value stored at <tt>ptr</tt>. It then adds <tt>delta</tt>, stores the result
6254 to <tt>ptr</tt>. It yields the original value stored at <tt>ptr</tt>.
6255</p>
6256
6257<h5>Examples:</h5>
6258<pre>
6259%ptr = malloc i32
6260 store i32 4, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6261%result1 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 4 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6262 <i>; yields {i32}:result1 = 4</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6263%result2 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 2 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6264 <i>; yields {i32}:result2 = 8</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6265%result3 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 5 )
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6266 <i>; yields {i32}:result3 = 10</i>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6267%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 15</i>
Andrew Lenharthab0b9492008-02-21 06:45:13 +0000[diff] [blame]6268</pre>
6269</div>
6270
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6271<!-- _______________________________________________________________________ -->
6272<div class="doc_subsubsection">
6273 <a name="int_atomic_load_sub">'<tt>llvm.atomic.load.sub.*</tt>' Intrinsic</a>
6274
6275</div>
6276<div class="doc_text">
6277<h5>Syntax:</h5>
6278<p>
6279 This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.sub</tt> on
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6280 any integer bit width and for different address spaces. Not all targets
6281 support all bit widths however.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6282<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6283declare i8 @llvm.atomic.load.sub.i8.p0i32( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6284declare i16 @llvm.atomic.load.sub.i16.p0i32( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6285declare i32 @llvm.atomic.load.sub.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6286declare 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]6287
6288</pre>
6289<h5>Overview:</h5>
6290<p>
6291 This intrinsic subtracts <tt>delta</tt> to the value stored in memory at
6292 <tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.
6293</p>
6294<h5>Arguments:</h5>
6295<p>
6296
6297 The intrinsic takes two arguments, the first a pointer to an integer value
6298 and the second an integer value. The result is also an integer value. These
6299 integer types can have any bit width, but they must all have the same bit
6300 width. The targets may only lower integer representations they support.
6301</p>
6302<h5>Semantics:</h5>
6303<p>
6304 This intrinsic does a series of operations atomically. It first loads the
6305 value stored at <tt>ptr</tt>. It then subtracts <tt>delta</tt>, stores the
6306 result to <tt>ptr</tt>. It yields the original value stored at <tt>ptr</tt>.
6307</p>
6308
6309<h5>Examples:</h5>
6310<pre>
6311%ptr = malloc i32
6312 store i32 8, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6313%result1 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 4 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6314 <i>; yields {i32}:result1 = 8</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6315%result2 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 2 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6316 <i>; yields {i32}:result2 = 4</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6317%result3 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 5 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6318 <i>; yields {i32}:result3 = 2</i>
6319%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = -3</i>
6320</pre>
6321</div>
6322
6323<!-- _______________________________________________________________________ -->
6324<div class="doc_subsubsection">
6325 <a name="int_atomic_load_and">'<tt>llvm.atomic.load.and.*</tt>' Intrinsic</a><br>
6326 <a name="int_atomic_load_nand">'<tt>llvm.atomic.load.nand.*</tt>' Intrinsic</a><br>
6327 <a name="int_atomic_load_or">'<tt>llvm.atomic.load.or.*</tt>' Intrinsic</a><br>
6328 <a name="int_atomic_load_xor">'<tt>llvm.atomic.load.xor.*</tt>' Intrinsic</a><br>
6329
6330</div>
6331<div class="doc_text">
6332<h5>Syntax:</h5>
6333<p>
6334 These are overloaded intrinsics. You can use <tt>llvm.atomic.load_and</tt>,
6335 <tt>llvm.atomic.load_nand</tt>, <tt>llvm.atomic.load_or</tt>, and
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6336 <tt>llvm.atomic.load_xor</tt> on any integer bit width and for different
6337 address spaces. Not all targets support all bit widths however.</p>
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6338<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6339declare i8 @llvm.atomic.load.and.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6340declare i16 @llvm.atomic.load.and.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6341declare i32 @llvm.atomic.load.and.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6342declare 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]6343
6344</pre>
6345
6346<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6347declare i8 @llvm.atomic.load.or.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6348declare i16 @llvm.atomic.load.or.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6349declare i32 @llvm.atomic.load.or.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6350declare 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]6351
6352</pre>
6353
6354<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6355declare i8 @llvm.atomic.load.nand.i8.p0i32( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6356declare i16 @llvm.atomic.load.nand.i16.p0i32( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6357declare i32 @llvm.atomic.load.nand.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6358declare 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]6359
6360</pre>
6361
6362<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6363declare i8 @llvm.atomic.load.xor.i8.p0i32( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6364declare i16 @llvm.atomic.load.xor.i16.p0i32( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6365declare i32 @llvm.atomic.load.xor.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6366declare 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]6367
6368</pre>
6369<h5>Overview:</h5>
6370<p>
6371 These intrinsics bitwise the operation (and, nand, or, xor) <tt>delta</tt> to
6372 the value stored in memory at <tt>ptr</tt>. It yields the original value
6373 at <tt>ptr</tt>.
6374</p>
6375<h5>Arguments:</h5>
6376<p>
6377
6378 These intrinsics take two arguments, the first a pointer to an integer value
6379 and the second an integer value. The result is also an integer value. These
6380 integer types can have any bit width, but they must all have the same bit
6381 width. The targets may only lower integer representations they support.
6382</p>
6383<h5>Semantics:</h5>
6384<p>
6385 These intrinsics does a series of operations atomically. They first load the
6386 value stored at <tt>ptr</tt>. They then do the bitwise operation
6387 <tt>delta</tt>, store the result to <tt>ptr</tt>. They yield the original
6388 value stored at <tt>ptr</tt>.
6389</p>
6390
6391<h5>Examples:</h5>
6392<pre>
6393%ptr = malloc i32
6394 store i32 0x0F0F, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6395%result0 = call i32 @llvm.atomic.load.nand.i32.p0i32( i32* %ptr, i32 0xFF )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6396 <i>; yields {i32}:result0 = 0x0F0F</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6397%result1 = call i32 @llvm.atomic.load.and.i32.p0i32( i32* %ptr, i32 0xFF )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6398 <i>; yields {i32}:result1 = 0xFFFFFFF0</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6399%result2 = call i32 @llvm.atomic.load.or.i32.p0i32( i32* %ptr, i32 0F )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6400 <i>; yields {i32}:result2 = 0xF0</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6401%result3 = call i32 @llvm.atomic.load.xor.i32.p0i32( i32* %ptr, i32 0F )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6402 <i>; yields {i32}:result3 = FF</i>
6403%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = F0</i>
6404</pre>
6405</div>
6406
6407
6408<!-- _______________________________________________________________________ -->
6409<div class="doc_subsubsection">
6410 <a name="int_atomic_load_max">'<tt>llvm.atomic.load.max.*</tt>' Intrinsic</a><br>
6411 <a name="int_atomic_load_min">'<tt>llvm.atomic.load.min.*</tt>' Intrinsic</a><br>
6412 <a name="int_atomic_load_umax">'<tt>llvm.atomic.load.umax.*</tt>' Intrinsic</a><br>
6413 <a name="int_atomic_load_umin">'<tt>llvm.atomic.load.umin.*</tt>' Intrinsic</a><br>
6414
6415</div>
6416<div class="doc_text">
6417<h5>Syntax:</h5>
6418<p>
6419 These are overloaded intrinsics. You can use <tt>llvm.atomic.load_max</tt>,
6420 <tt>llvm.atomic.load_min</tt>, <tt>llvm.atomic.load_umax</tt>, and
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6421 <tt>llvm.atomic.load_umin</tt> on any integer bit width and for different
6422 address spaces. Not all targets
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6423 support all bit widths however.</p>
6424<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6425declare i8 @llvm.atomic.load.max.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6426declare i16 @llvm.atomic.load.max.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6427declare i32 @llvm.atomic.load.max.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6428declare 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]6429
6430</pre>
6431
6432<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6433declare i8 @llvm.atomic.load.min.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6434declare i16 @llvm.atomic.load.min.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6435declare i32 @llvm.atomic.load.min.i32..p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6436declare 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]6437
6438</pre>
6439
6440<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6441declare i8 @llvm.atomic.load.umax.i8.p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6442declare i16 @llvm.atomic.load.umax.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6443declare i32 @llvm.atomic.load.umax.i32.p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6444declare 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]6445
6446</pre>
6447
6448<pre>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6449declare i8 @llvm.atomic.load.umin.i8..p0i8( i8* &lt;ptr&gt;, i8 &lt;delta&gt; )
6450declare i16 @llvm.atomic.load.umin.i16.p0i16( i16* &lt;ptr&gt;, i16 &lt;delta&gt; )
6451declare i32 @llvm.atomic.load.umin.i32..p0i32( i32* &lt;ptr&gt;, i32 &lt;delta&gt; )
6452declare 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]6453
6454</pre>
6455<h5>Overview:</h5>
6456<p>
6457 These intrinsics takes the signed or unsigned minimum or maximum of
6458 <tt>delta</tt> and the value stored in memory at <tt>ptr</tt>. It yields the
6459 original value at <tt>ptr</tt>.
6460</p>
6461<h5>Arguments:</h5>
6462<p>
6463
6464 These intrinsics take two arguments, the first a pointer to an integer value
6465 and the second an integer value. The result is also an integer value. These
6466 integer types can have any bit width, but they must all have the same bit
6467 width. The targets may only lower integer representations they support.
6468</p>
6469<h5>Semantics:</h5>
6470<p>
6471 These intrinsics does a series of operations atomically. They first load the
6472 value stored at <tt>ptr</tt>. They then do the signed or unsigned min or max
6473 <tt>delta</tt> and the value, store the result to <tt>ptr</tt>. They yield
6474 the original value stored at <tt>ptr</tt>.
6475</p>
6476
6477<h5>Examples:</h5>
6478<pre>
6479%ptr = malloc i32
6480 store i32 7, %ptr
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6481%result0 = call i32 @llvm.atomic.load.min.i32.p0i32( i32* %ptr, i32 -2 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6482 <i>; yields {i32}:result0 = 7</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6483%result1 = call i32 @llvm.atomic.load.max.i32.p0i32( i32* %ptr, i32 8 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6484 <i>; yields {i32}:result1 = -2</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6485%result2 = call i32 @llvm.atomic.load.umin.i32.p0i32( i32* %ptr, i32 10 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6486 <i>; yields {i32}:result2 = 8</i>
Mon P Wange3b3a722008-07-30 04:36:53 +0000[diff] [blame]6487%result3 = call i32 @llvm.atomic.load.umax.i32.p0i32( i32* %ptr, i32 30 )
Mon P Wang28873102008-06-25 08:15:39 +0000[diff] [blame]6488 <i>; yields {i32}:result3 = 8</i>
6489%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 30</i>
6490</pre>
6491</div>
Andrew Lenharth22c5c1b2008-02-16 01:24:58 +0000[diff] [blame]6492
6493<!-- ======================================================================= -->
6494<div class="doc_subsection">
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]6495 <a name="int_general">General Intrinsics</a>
6496</div>
6497
6498<div class="doc_text">
6499<p> This class of intrinsics is designed to be generic and has
6500no specific purpose. </p>
6501</div>
6502
6503<!-- _______________________________________________________________________ -->
6504<div class="doc_subsubsection">
6505 <a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
6506</div>
6507
6508<div class="doc_text">
6509
6510<h5>Syntax:</h5>
6511<pre>
Tanya Lattnerd2e84422007-06-18 23:42:37 +0000[diff] [blame]6512 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]6513</pre>
6514
6515<h5>Overview:</h5>
6516
6517<p>
6518The '<tt>llvm.var.annotation</tt>' intrinsic
6519</p>
6520
6521<h5>Arguments:</h5>
6522
6523<p>
Tanya Lattnerd2e84422007-06-18 23:42:37 +0000[diff] [blame]6524The first argument is a pointer to a value, the second is a pointer to a
6525global string, the third is a pointer to a global string which is the source
6526file name, and the last argument is the line number.
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]6527</p>
6528
6529<h5>Semantics:</h5>
6530
6531<p>
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]6532This intrinsic allows annotation of local variables with arbitrary strings.
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]6533This can be useful for special purpose optimizations that want to look for these
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]6534annotations. These have no other defined use, they are ignored by code
6535generation and optimization.
6536</p>
Tanya Lattner6d806e92007-06-15 20:50:54 +0000[diff] [blame]6537</div>
6538
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]6539<!-- _______________________________________________________________________ -->
6540<div class="doc_subsubsection">
Tanya Lattnere1a8da02007-09-21 23:57:59 +0000[diff] [blame]6541 <a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]6542</div>
6543
6544<div class="doc_text">
6545
6546<h5>Syntax:</h5>
Tanya Lattner39cfba62007-09-21 23:56:27 +0000[diff] [blame]6547<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
6548any integer bit width.
6549</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]6550<pre>
Tanya Lattnerd3989a82007-09-22 00:03:01 +0000[diff] [blame]6551 declare i8 @llvm.annotation.i8(i8 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
6552 declare i16 @llvm.annotation.i16(i16 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
6553 declare i32 @llvm.annotation.i32(i32 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
6554 declare i64 @llvm.annotation.i64(i64 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
6555 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]6556</pre>
6557
6558<h5>Overview:</h5>
Tanya Lattner39cfba62007-09-21 23:56:27 +0000[diff] [blame]6559
6560<p>
6561The '<tt>llvm.annotation</tt>' intrinsic.
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]6562</p>
6563
6564<h5>Arguments:</h5>
6565
6566<p>
6567The first argument is an integer value (result of some expression),
6568the second is a pointer to a global string, the third is a pointer to a global
6569string which is the source file name, and the last argument is the line number.
Tanya Lattner39cfba62007-09-21 23:56:27 +0000[diff] [blame]6570It returns the value of the first argument.
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]6571</p>
6572
6573<h5>Semantics:</h5>
6574
6575<p>
6576This intrinsic allows annotations to be put on arbitrary expressions
6577with arbitrary strings. This can be useful for special purpose optimizations
6578that want to look for these annotations. These have no other defined use, they
6579are ignored by code generation and optimization.
Dan Gohman0e451ce2008-10-14 16:51:45 +0000[diff] [blame]6580</p>
Tanya Lattnerb6367882007-09-21 22:59:12 +0000[diff] [blame]6581</div>
Jim Laskeydd4ef1b2007-03-14 19:31:19 +0000[diff] [blame]6582
Anton Korobeynikov4cb86182008-01-15 22:31:34 +0000[diff] [blame]6583<!-- _______________________________________________________________________ -->
6584<div class="doc_subsubsection">
6585 <a name="int_trap">'<tt>llvm.trap</tt>' Intrinsic</a>
6586</div>
6587
6588<div class="doc_text">
6589
6590<h5>Syntax:</h5>
6591<pre>
6592 declare void @llvm.trap()
6593</pre>
6594
6595<h5>Overview:</h5>
6596
6597<p>
6598The '<tt>llvm.trap</tt>' intrinsic
6599</p>
6600
6601<h5>Arguments:</h5>
6602
6603<p>
6604None
6605</p>
6606
6607<h5>Semantics:</h5>
6608
6609<p>
6610This intrinsics is lowered to the target dependent trap instruction. If the
6611target does not have a trap instruction, this intrinsic will be lowered to the
6612call of the abort() function.
6613</p>
6614</div>
6615
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]6616<!-- _______________________________________________________________________ -->
6617<div class="doc_subsubsection">
Misha Brukmandccb0252008-11-22 23:55:29 +0000[diff] [blame]6618 <a name="int_stackprotector">'<tt>llvm.stackprotector</tt>' Intrinsic</a>
Bill Wendling69e4adb2008-11-19 05:56:17 +0000[diff] [blame]6619</div>
6620<div class="doc_text">
6621<h5>Syntax:</h5>
6622<pre>
6623declare void @llvm.stackprotector( i8* &lt;guard&gt;, i8** &lt;slot&gt; )
6624
6625</pre>
6626<h5>Overview:</h5>
6627<p>
6628 The <tt>llvm.stackprotector</tt> intrinsic takes the <tt>guard</tt> and stores
6629 it onto the stack at <tt>slot</tt>. The stack slot is adjusted to ensure that
6630 it is placed on the stack before local variables.
6631</p>
6632<h5>Arguments:</h5>
6633<p>
6634 The <tt>llvm.stackprotector</tt> intrinsic requires two pointer arguments. The
6635 first argument is the value loaded from the stack guard
6636 <tt>@__stack_chk_guard</tt>. The second variable is an <tt>alloca</tt> that
6637 has enough space to hold the value of the guard.
6638</p>
6639<h5>Semantics:</h5>
6640<p>
6641 This intrinsic causes the prologue/epilogue inserter to force the position of
6642 the <tt>AllocaInst</tt> stack slot to be before local variables on the
6643 stack. This is to ensure that if a local variable on the stack is overwritten,
6644 it will destroy the value of the guard. When the function exits, the guard on
6645 the stack is checked against the original guard. If they're different, then
6646 the program aborts by calling the <tt>__stack_chk_fail()</tt> function.
6647</p>
6648</div>
6649
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]6650<!-- *********************************************************************** -->
Chris Lattner00950542001-06-06 20:29:01 +0000[diff] [blame]6651<hr>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]6652<address>
6653 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]6654 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]6655 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]6656 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]6657
6658 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencer05fe4b02006-03-14 05:39:39 +0000[diff] [blame]6659 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukmandaa4cb02004-03-01 17:47:27 +0000[diff] [blame]6660 Last modified: $Date$
6661</address>
Chris Lattnerc7d3ab32008-01-04 04:33:49 +0000[diff] [blame]6662
Misha Brukman9d0919f2003-11-08 01:05:38 +0000[diff] [blame]6663</body>
6664</html>