Gitiles
Code Review Sign In
gerrit-public.fairphone.software / toolchain / llvm-project / 308121c7576cda01a3fa8ec615a7c3eccf295947 / . / llvm / docs / LangRef.html
blob: fd40c92f9aab93c4f05032f4e2db8968ebc189ff [file] [log] [blame]
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3<html>
4<head>
5 <title>LLVM Assembly Language Reference Manual</title>
Reid Spencercb84e432004-08-26 20:44:00 +0000[diff] [blame]6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
7 <meta name="author" content="Chris Lattner">
8 <meta name="description"
9 content="LLVM Assembly Language Reference Manual.">
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]10 <link rel="stylesheet" href="llvm.css" type="text/css">
11</head>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]12
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]13<body>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]14
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]15<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]16<ol>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]17 <li><a href="#abstract">Abstract</a></li>
18 <li><a href="#introduction">Introduction</a></li>
19 <li><a href="#identifiers">Identifiers</a></li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]20 <li><a href="#highlevel">High Level Structure</a>
21 <ol>
22 <li><a href="#modulestructure">Module Structure</a></li>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]23 <li><a href="#linkage">Linkage Types</a></li>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]24 <li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]25 <li><a href="#globalvars">Global Variables</a></li>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]26 <li><a href="#functionstructure">Functions</a></li>
Anton Korobeynikov546ea7e2007-04-29 18:02:48 +0000[diff] [blame]27 <li><a href="#aliasstructure">Aliases</a>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]28 <li><a href="#paramattrs">Parameter Attributes</a></li>
Gordon Henriksen71183b62007-12-10 03:18:06 +0000[diff] [blame]29 <li><a href="#gc">Garbage Collector Names</a></li>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]30 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]31 <li><a href="#datalayout">Data Layout</a></li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]32 </ol>
33 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]34 <li><a href="#typesystem">Type System</a>
35 <ol>
Robert Bocchino820bc75b2006-02-17 21:18:08 +0000[diff] [blame]36 <li><a href="#t_primitive">Primitive Types</a>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]37 <ol>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]38 <li><a href="#t_classifications">Type Classifications</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]39 </ol>
40 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]41 <li><a href="#t_derived">Derived Types</a>
42 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]43 <li><a href="#t_array">Array Type</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]44 <li><a href="#t_function">Function Type</a></li>
45 <li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]46 <li><a href="#t_struct">Structure Type</a></li>
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]47 <li><a href="#t_pstruct">Packed Structure Type</a></li>
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]48 <li><a href="#t_vector">Vector Type</a></li>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]49 <li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]50 </ol>
51 </li>
52 </ol>
53 </li>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]54 <li><a href="#constants">Constants</a>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]55 <ol>
56 <li><a href="#simpleconstants">Simple Constants</a>
57 <li><a href="#aggregateconstants">Aggregate Constants</a>
58 <li><a href="#globalconstants">Global Variable and Function Addresses</a>
59 <li><a href="#undefvalues">Undefined Values</a>
60 <li><a href="#constantexprs">Constant Expressions</a>
61 </ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]62 </li>
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]63 <li><a href="#othervalues">Other Values</a>
64 <ol>
65 <li><a href="#inlineasm">Inline Assembler Expressions</a>
66 </ol>
67 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]68 <li><a href="#instref">Instruction Reference</a>
69 <ol>
70 <li><a href="#terminators">Terminator Instructions</a>
71 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]72 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
73 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]74 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
75 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]76 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]77 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]78 </ol>
79 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]80 <li><a href="#binaryops">Binary Operations</a>
81 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]82 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
83 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
84 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]85 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
86 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
87 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]88 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
89 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
90 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]91 </ol>
92 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]93 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
94 <ol>
Reid Spencer2ab01932007-02-02 13:57:07 +0000[diff] [blame]95 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
96 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
97 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]98 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]99 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]100 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]101 </ol>
102 </li>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]103 <li><a href="#vectorops">Vector Operations</a>
104 <ol>
105 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
106 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
107 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]108 </ol>
109 </li>
Chris Lattner6ab66722006-08-15 00:45:58 +0000[diff] [blame]110 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]111 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]112 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
113 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
114 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino820bc75b2006-02-17 21:18:08 +0000[diff] [blame]115 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
116 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
117 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]118 </ol>
119 </li>
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]120 <li><a href="#convertops">Conversion Operations</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]121 <ol>
122 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
123 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
124 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
125 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
126 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]127 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
128 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
129 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
130 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]131 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
132 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]133 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]134 </ol>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]135 <li><a href="#otherops">Other Operations</a>
136 <ol>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]137 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
138 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]139 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]140 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]141 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattner33337472006-01-13 23:26:01 +0000[diff] [blame]142 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]143 </ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]144 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]145 </ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]146 </li>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]147 <li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]148 <ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]149 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
150 <ol>
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]151 <li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
152 <li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
153 <li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]154 </ol>
155 </li>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]156 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
157 <ol>
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]158 <li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
159 <li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
160 <li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]161 </ol>
162 </li>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]163 <li><a href="#int_codegen">Code Generator Intrinsics</a>
164 <ol>
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]165 <li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
166 <li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
167 <li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
168 <li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
169 <li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
170 <li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
171 <li><a href="#int_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswellaa1c3c12004-04-09 16:43:20 +0000[diff] [blame]172 </ol>
173 </li>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]174 <li><a href="#int_libc">Standard C Library Intrinsics</a>
175 <ol>
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]176 <li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
177 <li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
178 <li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
179 <li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
180 <li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]181 <li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
182 <li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
183 <li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]184 </ol>
185 </li>
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]186 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]187 <ol>
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]188 <li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattnerb748c672006-01-16 22:34:14 +0000[diff] [blame]189 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
190 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
191 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Reid Spencer5bf54c82007-04-11 23:23:49 +0000[diff] [blame]192 <li><a href="#int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic </a></li>
193 <li><a href="#int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic </a></li>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]194 </ol>
195 </li>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]196 <li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskey2211f492007-03-14 19:31:19 +0000[diff] [blame]197 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]198 <li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]199 <ol>
200 <li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]201 </ol>
202 </li>
Reid Spencer5b2cb0f2007-07-20 19:59:11 +0000[diff] [blame]203 <li><a href="#int_general">General intrinsics</a>
Tanya Lattnercb1b9602007-06-15 20:50:54 +0000[diff] [blame]204 <ol>
Reid Spencer5b2cb0f2007-07-20 19:59:11 +0000[diff] [blame]205 <li><a href="#int_var_annotation">
Tanya Lattner08abc812007-09-22 00:01:26 +0000[diff] [blame]206 <tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Reid Spencer5b2cb0f2007-07-20 19:59:11 +0000[diff] [blame]207 </ol>
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]208 <ol>
209 <li><a href="#int_annotation">
Tanya Lattner08abc812007-09-22 00:01:26 +0000[diff] [blame]210 <tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]211 </ol>
Tanya Lattnercb1b9602007-06-15 20:50:54 +0000[diff] [blame]212 </li>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]213 </ol>
214 </li>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]215</ol>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]216
217<div class="doc_author">
218 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
219 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]220</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]221
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]222<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]223<div class="doc_section"> <a name="abstract">Abstract </a></div>
224<!-- *********************************************************************** -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]225
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]226<div class="doc_text">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]227<p>This document is a reference manual for the LLVM assembly language.
228LLVM is an SSA based representation that provides type safety,
229low-level operations, flexibility, and the capability of representing
230'all' high-level languages cleanly. It is the common code
231representation used throughout all phases of the LLVM compilation
232strategy.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]233</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]234
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]235<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]236<div class="doc_section"> <a name="introduction">Introduction</a> </div>
237<!-- *********************************************************************** -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]238
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]239<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]240
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]241<p>The LLVM code representation is designed to be used in three
Gabor Greifa54634a2007-07-06 22:07:22 +0000[diff] [blame]242different forms: as an in-memory compiler IR, as an on-disk bitcode
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]243representation (suitable for fast loading by a Just-In-Time compiler),
244and as a human readable assembly language representation. This allows
245LLVM to provide a powerful intermediate representation for efficient
246compiler transformations and analysis, while providing a natural means
247to debug and visualize the transformations. The three different forms
248of LLVM are all equivalent. This document describes the human readable
249representation and notation.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]250
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]251<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]252while being expressive, typed, and extensible at the same time. It
253aims to be a "universal IR" of sorts, by being at a low enough level
254that high-level ideas may be cleanly mapped to it (similar to how
255microprocessors are "universal IR's", allowing many source languages to
256be mapped to them). By providing type information, LLVM can be used as
257the target of optimizations: for example, through pointer analysis, it
258can be proven that a C automatic variable is never accessed outside of
259the current function... allowing it to be promoted to a simple SSA
260value instead of a memory location.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]261
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]262</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]263
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]264<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]265<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]266
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]267<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]268
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]269<p>It is important to note that this document describes 'well formed'
270LLVM assembly language. There is a difference between what the parser
271accepts and what is considered 'well formed'. For example, the
272following instruction is syntactically okay, but not well formed:</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]273
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]274<div class="doc_code">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]275<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]276%x = <a href="#i_add">add</a> i32 1, %x
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]277</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]278</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]279
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]280<p>...because the definition of <tt>%x</tt> does not dominate all of
281its uses. The LLVM infrastructure provides a verification pass that may
282be used to verify that an LLVM module is well formed. This pass is
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]283automatically run by the parser after parsing input assembly and by
Gabor Greifa54634a2007-07-06 22:07:22 +0000[diff] [blame]284the optimizer before it outputs bitcode. The violations pointed out
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]285by the verifier pass indicate bugs in transformation passes or input to
286the parser.</p>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]287</div>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]288
Chris Lattner87a3dbe2007-10-03 17:34:29 +0000[diff] [blame]289<!-- Describe the typesetting conventions here. -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]290
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]291<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]292<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]293<!-- *********************************************************************** -->
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]294
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]295<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]296
Reid Spencerb23b65f2007-08-07 14:34:28 +0000[diff] [blame]297 <p>LLVM identifiers come in two basic types: global and local. Global
298 identifiers (functions, global variables) begin with the @ character. Local
299 identifiers (register names, types) begin with the % character. Additionally,
300 there are three different formats for identifiers, for different purposes:
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]301
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]302<ol>
Reid Spencerb23b65f2007-08-07 14:34:28 +0000[diff] [blame]303 <li>Named values are represented as a string of characters with their prefix.
304 For example, %foo, @DivisionByZero, %a.really.long.identifier. The actual
305 regular expression used is '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]306 Identifiers which require other characters in their names can be surrounded
Reid Spencerb23b65f2007-08-07 14:34:28 +0000[diff] [blame]307 with quotes. In this way, anything except a <tt>&quot;</tt> character can
308 be used in a named value.</li>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]309
Reid Spencerb23b65f2007-08-07 14:34:28 +0000[diff] [blame]310 <li>Unnamed values are represented as an unsigned numeric value with their
311 prefix. For example, %12, @2, %44.</li>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]312
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]313 <li>Constants, which are described in a <a href="#constants">section about
314 constants</a>, below.</li>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]315</ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]316
Reid Spencerb23b65f2007-08-07 14:34:28 +0000[diff] [blame]317<p>LLVM requires that values start with a prefix for two reasons: Compilers
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]318don't need to worry about name clashes with reserved words, and the set of
319reserved words may be expanded in the future without penalty. Additionally,
320unnamed identifiers allow a compiler to quickly come up with a temporary
321variable without having to avoid symbol table conflicts.</p>
322
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]323<p>Reserved words in LLVM are very similar to reserved words in other
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]324languages. There are keywords for different opcodes
325('<tt><a href="#i_add">add</a></tt>',
326 '<tt><a href="#i_bitcast">bitcast</a></tt>',
327 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]328href="#t_void">void</a></tt>', '<tt><a href="#t_primitive">i32</a></tt>', etc...),
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]329and others. These reserved words cannot conflict with variable names, because
Reid Spencerb23b65f2007-08-07 14:34:28 +0000[diff] [blame]330none of them start with a prefix character ('%' or '@').</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]331
332<p>Here is an example of LLVM code to multiply the integer variable
333'<tt>%X</tt>' by 8:</p>
334
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]335<p>The easy way:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]336
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]337<div class="doc_code">
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]338<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]339%result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]340</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]341</div>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]342
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]343<p>After strength reduction:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]344
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]345<div class="doc_code">
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]346<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]347%result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]348</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]349</div>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]350
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]351<p>And the hard way:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]352
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]353<div class="doc_code">
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]354<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]355<a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
356<a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
357%result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]358</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]359</div>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]360
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]361<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
362important lexical features of LLVM:</p>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]363
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]364<ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]365
366 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
367 line.</li>
368
369 <li>Unnamed temporaries are created when the result of a computation is not
370 assigned to a named value.</li>
371
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]372 <li>Unnamed temporaries are numbered sequentially</li>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]373
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]374</ol>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]375
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]376<p>...and it also shows a convention that we follow in this document. When
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]377demonstrating instructions, we will follow an instruction with a comment that
378defines the type and name of value produced. Comments are shown in italic
379text.</p>
380
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]381</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]382
383<!-- *********************************************************************** -->
384<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
385<!-- *********************************************************************** -->
386
387<!-- ======================================================================= -->
388<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
389</div>
390
391<div class="doc_text">
392
393<p>LLVM programs are composed of "Module"s, each of which is a
394translation unit of the input programs. Each module consists of
395functions, global variables, and symbol table entries. Modules may be
396combined together with the LLVM linker, which merges function (and
397global variable) definitions, resolves forward declarations, and merges
398symbol table entries. Here is an example of the "hello world" module:</p>
399
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]400<div class="doc_code">
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]401<pre><i>; Declare the string constant as a global constant...</i>
Chris Lattner2b0bf4f2007-06-12 17:00:26 +0000[diff] [blame]402<a href="#identifiers">@.LC0</a> = <a href="#linkage_internal">internal</a> <a
403 href="#globalvars">constant</a> <a href="#t_array">[13 x i8]</a> c"hello world\0A\00" <i>; [13 x i8]*</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]404
405<i>; External declaration of the puts function</i>
Chris Lattner2b0bf4f2007-06-12 17:00:26 +0000[diff] [blame]406<a href="#functionstructure">declare</a> i32 @puts(i8 *) <i>; i32(i8 *)* </i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]407
408<i>; Definition of main function</i>
Chris Lattner2b0bf4f2007-06-12 17:00:26 +0000[diff] [blame]409define i32 @main() { <i>; i32()* </i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]410 <i>; Convert [13x i8 ]* to i8 *...</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]411 %cast210 = <a
Chris Lattner2150cde2007-06-12 17:01:15 +0000[diff] [blame]412 href="#i_getelementptr">getelementptr</a> [13 x i8 ]* @.LC0, i64 0, i64 0 <i>; i8 *</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]413
414 <i>; Call puts function to write out the string to stdout...</i>
415 <a
Chris Lattner2b0bf4f2007-06-12 17:00:26 +0000[diff] [blame]416 href="#i_call">call</a> i32 @puts(i8 * %cast210) <i>; i32</i>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]417 <a
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]418 href="#i_ret">ret</a> i32 0<br>}<br>
419</pre>
420</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]421
422<p>This example is made up of a <a href="#globalvars">global variable</a>
423named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
424function, and a <a href="#functionstructure">function definition</a>
425for "<tt>main</tt>".</p>
426
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]427<p>In general, a module is made up of a list of global values,
428where both functions and global variables are global values. Global values are
429represented by a pointer to a memory location (in this case, a pointer to an
430array of char, and a pointer to a function), and have one of the following <a
431href="#linkage">linkage types</a>.</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]432
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]433</div>
434
435<!-- ======================================================================= -->
436<div class="doc_subsection">
437 <a name="linkage">Linkage Types</a>
438</div>
439
440<div class="doc_text">
441
442<p>
443All Global Variables and Functions have one of the following types of linkage:
444</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]445
446<dl>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]447
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]448 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]449
450 <dd>Global values with internal linkage are only directly accessible by
451 objects in the current module. In particular, linking code into a module with
452 an internal global value may cause the internal to be renamed as necessary to
453 avoid collisions. Because the symbol is internal to the module, all
454 references can be updated. This corresponds to the notion of the
Chris Lattnere20b4702007-01-14 06:51:48 +0000[diff] [blame]455 '<tt>static</tt>' keyword in C.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]456 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]457
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]458 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]459
Chris Lattnere20b4702007-01-14 06:51:48 +0000[diff] [blame]460 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
461 the same name when linkage occurs. This is typically used to implement
462 inline functions, templates, or other code which must be generated in each
463 translation unit that uses it. Unreferenced <tt>linkonce</tt> globals are
464 allowed to be discarded.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]465 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]466
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]467 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]468
469 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
470 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
Chris Lattnere20b4702007-01-14 06:51:48 +0000[diff] [blame]471 used for globals that may be emitted in multiple translation units, but that
472 are not guaranteed to be emitted into every translation unit that uses them.
473 One example of this are common globals in C, such as "<tt>int X;</tt>" at
474 global scope.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]475 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]476
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]477 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]478
479 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
480 pointer to array type. When two global variables with appending linkage are
481 linked together, the two global arrays are appended together. This is the
482 LLVM, typesafe, equivalent of having the system linker append together
483 "sections" with identical names when .o files are linked.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]484 </dd>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]485
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]486 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt>: </dt>
487 <dd>The semantics of this linkage follow the ELF model: the symbol is weak
488 until linked, if not linked, the symbol becomes null instead of being an
489 undefined reference.
490 </dd>
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]491
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]492 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattnerd79749a2004-12-09 16:36:40 +0000[diff] [blame]493
494 <dd>If none of the above identifiers are used, the global is externally
495 visible, meaning that it participates in linkage and can be used to resolve
496 external symbol references.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]497 </dd>
Reid Spencer7972c472007-04-11 23:49:50 +0000[diff] [blame]498</dl>
Anton Korobeynikovd61d39e2006-09-14 18:23:27 +0000[diff] [blame]499
Anton Korobeynikovd61d39e2006-09-14 18:23:27 +0000[diff] [blame]500 <p>
501 The next two types of linkage are targeted for Microsoft Windows platform
502 only. They are designed to support importing (exporting) symbols from (to)
503 DLLs.
504 </p>
505
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]506 <dl>
Anton Korobeynikovd61d39e2006-09-14 18:23:27 +0000[diff] [blame]507 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt>: </dt>
508
509 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
510 or variable via a global pointer to a pointer that is set up by the DLL
511 exporting the symbol. On Microsoft Windows targets, the pointer name is
512 formed by combining <code>_imp__</code> and the function or variable name.
513 </dd>
514
515 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt>: </dt>
516
517 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
518 pointer to a pointer in a DLL, so that it can be referenced with the
519 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
520 name is formed by combining <code>_imp__</code> and the function or variable
521 name.
522 </dd>
523
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]524</dl>
525
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]526<p><a name="linkage_external"></a>For example, since the "<tt>.LC0</tt>"
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]527variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
528variable and was linked with this one, one of the two would be renamed,
529preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
530external (i.e., lacking any linkage declarations), they are accessible
Reid Spencer92c671e2007-01-05 00:59:10 +0000[diff] [blame]531outside of the current module.</p>
532<p>It is illegal for a function <i>declaration</i>
533to have any linkage type other than "externally visible", <tt>dllimport</tt>,
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]534or <tt>extern_weak</tt>.</p>
Anton Korobeynikova97b6942007-04-25 14:27:10 +0000[diff] [blame]535<p>Aliases can have only <tt>external</tt>, <tt>internal</tt> and <tt>weak</tt>
536linkages.
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]537</div>
538
539<!-- ======================================================================= -->
540<div class="doc_subsection">
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]541 <a name="callingconv">Calling Conventions</a>
542</div>
543
544<div class="doc_text">
545
546<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
547and <a href="#i_invoke">invokes</a> can all have an optional calling convention
548specified for the call. The calling convention of any pair of dynamic
549caller/callee must match, or the behavior of the program is undefined. The
550following calling conventions are supported by LLVM, and more may be added in
551the future:</p>
552
553<dl>
554 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
555
556 <dd>This calling convention (the default if no other calling convention is
557 specified) matches the target C calling conventions. This calling convention
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]558 supports varargs function calls and tolerates some mismatch in the declared
Reid Spencer72ba4992006-12-31 21:30:18 +0000[diff] [blame]559 prototype and implemented declaration of the function (as does normal C).
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]560 </dd>
561
562 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
563
564 <dd>This calling convention attempts to make calls as fast as possible
565 (e.g. by passing things in registers). This calling convention allows the
566 target to use whatever tricks it wants to produce fast code for the target,
Chris Lattnerc792eb32005-05-06 23:08:23 +0000[diff] [blame]567 without having to conform to an externally specified ABI. Implementations of
568 this convention should allow arbitrary tail call optimization to be supported.
569 This calling convention does not support varargs and requires the prototype of
570 all callees to exactly match the prototype of the function definition.
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]571 </dd>
572
573 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
574
575 <dd>This calling convention attempts to make code in the caller as efficient
576 as possible under the assumption that the call is not commonly executed. As
577 such, these calls often preserve all registers so that the call does not break
578 any live ranges in the caller side. This calling convention does not support
579 varargs and requires the prototype of all callees to exactly match the
580 prototype of the function definition.
581 </dd>
582
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]583 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]584
585 <dd>Any calling convention may be specified by number, allowing
586 target-specific calling conventions to be used. Target specific calling
587 conventions start at 64.
588 </dd>
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]589</dl>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]590
591<p>More calling conventions can be added/defined on an as-needed basis, to
592support pascal conventions or any other well-known target-independent
593convention.</p>
594
595</div>
596
597<!-- ======================================================================= -->
598<div class="doc_subsection">
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]599 <a name="visibility">Visibility Styles</a>
600</div>
601
602<div class="doc_text">
603
604<p>
605All Global Variables and Functions have one of the following visibility styles:
606</p>
607
608<dl>
609 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
610
611 <dd>On ELF, default visibility means that the declaration is visible to other
612 modules and, in shared libraries, means that the declared entity may be
613 overridden. On Darwin, default visibility means that the declaration is
614 visible to other modules. Default visibility corresponds to "external
615 linkage" in the language.
616 </dd>
617
618 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
619
620 <dd>Two declarations of an object with hidden visibility refer to the same
621 object if they are in the same shared object. Usually, hidden visibility
622 indicates that the symbol will not be placed into the dynamic symbol table,
623 so no other module (executable or shared library) can reference it
624 directly.
625 </dd>
626
Anton Korobeynikov39f3cff2007-04-29 18:35:00 +0000[diff] [blame]627 <dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
628
629 <dd>On ELF, protected visibility indicates that the symbol will be placed in
630 the dynamic symbol table, but that references within the defining module will
631 bind to the local symbol. That is, the symbol cannot be overridden by another
632 module.
633 </dd>
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]634</dl>
635
636</div>
637
638<!-- ======================================================================= -->
639<div class="doc_subsection">
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]640 <a name="globalvars">Global Variables</a>
641</div>
642
643<div class="doc_text">
644
Chris Lattner5d5aede2005-02-12 19:30:21 +0000[diff] [blame]645<p>Global variables define regions of memory allocated at compilation time
Chris Lattner662c8722005-11-12 00:45:07 +0000[diff] [blame]646instead of run-time. Global variables may optionally be initialized, may have
Anton Korobeynikova97b6942007-04-25 14:27:10 +0000[diff] [blame]647an explicit section to be placed in, and may have an optional explicit alignment
648specified. A variable may be defined as "thread_local", which means that it
649will not be shared by threads (each thread will have a separated copy of the
650variable). A variable may be defined as a global "constant," which indicates
651that the contents of the variable will <b>never</b> be modified (enabling better
Chris Lattner5d5aede2005-02-12 19:30:21 +0000[diff] [blame]652optimization, allowing the global data to be placed in the read-only section of
653an executable, etc). Note that variables that need runtime initialization
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]654cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner5d5aede2005-02-12 19:30:21 +0000[diff] [blame]655
656<p>
657LLVM explicitly allows <em>declarations</em> of global variables to be marked
658constant, even if the final definition of the global is not. This capability
659can be used to enable slightly better optimization of the program, but requires
660the language definition to guarantee that optimizations based on the
661'constantness' are valid for the translation units that do not include the
662definition.
663</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]664
665<p>As SSA values, global variables define pointer values that are in
666scope (i.e. they dominate) all basic blocks in the program. Global
667variables always define a pointer to their "content" type because they
668describe a region of memory, and all memory objects in LLVM are
669accessed through pointers.</p>
670
Christopher Lamb308121c2007-12-11 09:31:00 +0000[diff] [blame^]671<p>A global variable may be declared to reside in a target-specifc numbered
672address space. For targets that support them, address spaces may affect how
673optimizations are performed and/or what target instructions are used to access
674the variable. The default address space is zero.</p>
675
Chris Lattner662c8722005-11-12 00:45:07 +0000[diff] [blame]676<p>LLVM allows an explicit section to be specified for globals. If the target
677supports it, it will emit globals to the section specified.</p>
678
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]679<p>An explicit alignment may be specified for a global. If not present, or if
680the alignment is set to zero, the alignment of the global is set by the target
681to whatever it feels convenient. If an explicit alignment is specified, the
682global is forced to have at least that much alignment. All alignments must be
683a power of 2.</p>
684
Christopher Lamb308121c2007-12-11 09:31:00 +0000[diff] [blame^]685<p>For example, the following defines a global in a numbered address space with
686an initializer, section, and alignment:</p>
Chris Lattner5760c502007-01-14 00:27:09 +0000[diff] [blame]687
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]688<div class="doc_code">
Chris Lattner5760c502007-01-14 00:27:09 +0000[diff] [blame]689<pre>
Christopher Lamb308121c2007-12-11 09:31:00 +0000[diff] [blame^]690@G = constant float 1.0 addrspace(5), section "foo", align 4
Chris Lattner5760c502007-01-14 00:27:09 +0000[diff] [blame]691</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]692</div>
Chris Lattner5760c502007-01-14 00:27:09 +0000[diff] [blame]693
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]694</div>
695
696
697<!-- ======================================================================= -->
698<div class="doc_subsection">
699 <a name="functionstructure">Functions</a>
700</div>
701
702<div class="doc_text">
703
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]704<p>LLVM function definitions consist of the "<tt>define</tt>" keyord,
705an optional <a href="#linkage">linkage type</a>, an optional
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]706<a href="#visibility">visibility style</a>, an optional
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]707<a href="#callingconv">calling convention</a>, a return type, an optional
708<a href="#paramattrs">parameter attribute</a> for the return type, a function
709name, a (possibly empty) argument list (each with optional
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]710<a href="#paramattrs">parameter attributes</a>), an optional section, an
Gordon Henriksendc5cafb2007-12-10 03:30:21 +0000[diff] [blame]711optional alignment, an optional <a href="#gc">garbage collector name</a>, an
Gordon Henriksen71183b62007-12-10 03:18:06 +0000[diff] [blame]712opening curly brace, a list of basic blocks, and a closing curly brace.
Anton Korobeynikovc7f9f3d2007-01-23 12:35:46 +0000[diff] [blame]713
714LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
715optional <a href="#linkage">linkage type</a>, an optional
716<a href="#visibility">visibility style</a>, an optional
717<a href="#callingconv">calling convention</a>, a return type, an optional
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]718<a href="#paramattrs">parameter attribute</a> for the return type, a function
Gordon Henriksen71183b62007-12-10 03:18:06 +0000[diff] [blame]719name, a possibly empty list of arguments, an optional alignment, and an optional
Gordon Henriksendc5cafb2007-12-10 03:30:21 +0000[diff] [blame]720<a href="#gc">garbage collector name</a>.</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]721
722<p>A function definition contains a list of basic blocks, forming the CFG for
723the function. Each basic block may optionally start with a label (giving the
724basic block a symbol table entry), contains a list of instructions, and ends
725with a <a href="#terminators">terminator</a> instruction (such as a branch or
726function return).</p>
727
Chris Lattnera59fb102007-06-08 16:52:14 +0000[diff] [blame]728<p>The first basic block in a function is special in two ways: it is immediately
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]729executed on entrance to the function, and it is not allowed to have predecessor
730basic blocks (i.e. there can not be any branches to the entry block of a
731function). Because the block can have no predecessors, it also cannot have any
732<a href="#i_phi">PHI nodes</a>.</p>
733
Chris Lattner662c8722005-11-12 00:45:07 +0000[diff] [blame]734<p>LLVM allows an explicit section to be specified for functions. If the target
735supports it, it will emit functions to the section specified.</p>
736
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]737<p>An explicit alignment may be specified for a function. If not present, or if
738the alignment is set to zero, the alignment of the function is set by the target
739to whatever it feels convenient. If an explicit alignment is specified, the
740function is forced to have at least that much alignment. All alignments must be
741a power of 2.</p>
742
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]743</div>
744
Anton Korobeynikova97b6942007-04-25 14:27:10 +0000[diff] [blame]745
746<!-- ======================================================================= -->
747<div class="doc_subsection">
748 <a name="aliasstructure">Aliases</a>
749</div>
750<div class="doc_text">
751 <p>Aliases act as "second name" for the aliasee value (which can be either
Anton Korobeynikovb18f8f82007-04-28 13:45:00 +0000[diff] [blame]752 function or global variable or bitcast of global value). Aliases may have an
Anton Korobeynikova97b6942007-04-25 14:27:10 +0000[diff] [blame]753 optional <a href="#linkage">linkage type</a>, and an
754 optional <a href="#visibility">visibility style</a>.</p>
755
756 <h5>Syntax:</h5>
757
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]758<div class="doc_code">
Bill Wendling2d8b9a82007-05-29 09:42:13 +0000[diff] [blame]759<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]760@&lt;Name&gt; = [Linkage] [Visibility] alias &lt;AliaseeTy&gt; @&lt;Aliasee&gt;
Bill Wendling2d8b9a82007-05-29 09:42:13 +0000[diff] [blame]761</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]762</div>
Anton Korobeynikova97b6942007-04-25 14:27:10 +0000[diff] [blame]763
764</div>
765
766
767
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]768<!-- ======================================================================= -->
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]769<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
770<div class="doc_text">
771 <p>The return type and each parameter of a function type may have a set of
772 <i>parameter attributes</i> associated with them. Parameter attributes are
773 used to communicate additional information about the result or parameters of
Duncan Sandsad0ea2d2007-11-27 13:23:08 +0000[diff] [blame]774 a function. Parameter attributes are considered to be part of the function,
775 not of the function type, so functions with different parameter attributes
776 can have the same function type.</p>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]777
Reid Spencercf7ebf52007-01-15 18:27:39 +0000[diff] [blame]778 <p>Parameter attributes are simple keywords that follow the type specified. If
779 multiple parameter attributes are needed, they are space separated. For
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]780 example:</p>
781
782<div class="doc_code">
783<pre>
Duncan Sandsad0ea2d2007-11-27 13:23:08 +0000[diff] [blame]784declare i32 @printf(i8* noalias , ...) nounwind
785declare i32 @atoi(i8*) nounwind readonly
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]786</pre>
787</div>
788
Duncan Sandsad0ea2d2007-11-27 13:23:08 +0000[diff] [blame]789 <p>Note that any attributes for the function result (<tt>nounwind</tt>,
790 <tt>readonly</tt>) come immediately after the argument list.</p>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]791
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]792 <p>Currently, only the following parameter attributes are defined:</p>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]793 <dl>
Reid Spencer314e1cb2007-07-19 23:13:04 +0000[diff] [blame]794 <dt><tt>zeroext</tt></dt>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]795 <dd>This indicates that the parameter should be zero extended just before
796 a call to this function.</dd>
Reid Spencer314e1cb2007-07-19 23:13:04 +0000[diff] [blame]797 <dt><tt>signext</tt></dt>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]798 <dd>This indicates that the parameter should be sign extended just before
799 a call to this function.</dd>
Anton Korobeynikove8166852007-01-28 14:30:45 +0000[diff] [blame]800 <dt><tt>inreg</tt></dt>
801 <dd>This indicates that the parameter should be placed in register (if
Anton Korobeynikove93c6e82007-01-28 15:27:21 +0000[diff] [blame]802 possible) during assembling function call. Support for this attribute is
803 target-specific</dd>
Anton Korobeynikove8166852007-01-28 14:30:45 +0000[diff] [blame]804 <dt><tt>sret</tt></dt>
Anton Korobeynikove93c6e82007-01-28 15:27:21 +0000[diff] [blame]805 <dd>This indicates that the parameter specifies the address of a structure
Reid Spencer05dbb9d2007-03-22 02:02:11 +0000[diff] [blame]806 that is the return value of the function in the source program.</dd>
Zhou Sheng2444a9a2007-06-05 05:28:26 +0000[diff] [blame]807 <dt><tt>noalias</tt></dt>
808 <dd>This indicates that the parameter not alias any other object or any
809 other "noalias" objects during the function call.
Reid Spencer9d1700e2007-03-22 02:18:56 +0000[diff] [blame]810 <dt><tt>noreturn</tt></dt>
811 <dd>This function attribute indicates that the function never returns. This
812 indicates to LLVM that every call to this function should be treated as if
813 an <tt>unreachable</tt> instruction immediately followed the call.</dd>
Reid Spencer05dbb9d2007-03-22 02:02:11 +0000[diff] [blame]814 <dt><tt>nounwind</tt></dt>
815 <dd>This function attribute indicates that the function type does not use
816 the unwind instruction and does not allow stack unwinding to propagate
817 through it.</dd>
Duncan Sands27e91592007-07-27 19:57:41 +0000[diff] [blame]818 <dt><tt>nest</tt></dt>
819 <dd>This indicates that the parameter can be excised using the
820 <a href="#int_trampoline">trampoline intrinsics</a>.</dd>
Duncan Sandsa89a1132007-11-22 20:23:04 +0000[diff] [blame]821 <dt><tt>readonly</tt></dt>
Duncan Sands730a3262007-11-14 21:14:02 +0000[diff] [blame]822 <dd>This function attribute indicates that the function has no side-effects
Duncan Sandsa89a1132007-11-22 20:23:04 +0000[diff] [blame]823 except for producing a return value or throwing an exception. The value
824 returned must only depend on the function arguments and/or global variables.
825 It may use values obtained by dereferencing pointers.</dd>
826 <dt><tt>readnone</tt></dt>
827 <dd>A <tt>readnone</tt> function has the same restrictions as a <tt>readonly</tt>
Duncan Sands730a3262007-11-14 21:14:02 +0000[diff] [blame]828 function, but in addition it is not allowed to dereference any pointer arguments
829 or global variables.
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]830 </dl>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]831
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]832</div>
833
834<!-- ======================================================================= -->
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]835<div class="doc_subsection">
Gordon Henriksen71183b62007-12-10 03:18:06 +0000[diff] [blame]836 <a name="gc">Garbage Collector Names</a>
837</div>
838
839<div class="doc_text">
840<p>Each function may specify a garbage collector name, which is simply a
841string.</p>
842
843<div class="doc_code"><pre
844>define void @f() gc "name" { ...</pre></div>
845
846<p>The compiler declares the supported values of <i>name</i>. Specifying a
847collector which will cause the compiler to alter its output in order to support
848the named garbage collection algorithm.</p>
849</div>
850
851<!-- ======================================================================= -->
852<div class="doc_subsection">
Chris Lattner93564892006-04-08 04:40:53 +0000[diff] [blame]853 <a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]854</div>
855
856<div class="doc_text">
857<p>
858Modules may contain "module-level inline asm" blocks, which corresponds to the
859GCC "file scope inline asm" blocks. These blocks are internally concatenated by
860LLVM and treated as a single unit, but may be separated in the .ll file if
861desired. The syntax is very simple:
862</p>
863
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]864<div class="doc_code">
865<pre>
866module asm "inline asm code goes here"
867module asm "more can go here"
868</pre>
869</div>
Chris Lattner91c15c42006-01-23 23:23:47 +0000[diff] [blame]870
871<p>The strings can contain any character by escaping non-printable characters.
872 The escape sequence used is simply "\xx" where "xx" is the two digit hex code
873 for the number.
874</p>
875
876<p>
877 The inline asm code is simply printed to the machine code .s file when
878 assembly code is generated.
879</p>
880</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]881
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]882<!-- ======================================================================= -->
883<div class="doc_subsection">
884 <a name="datalayout">Data Layout</a>
885</div>
886
887<div class="doc_text">
888<p>A module may specify a target specific data layout string that specifies how
Reid Spencer7972c472007-04-11 23:49:50 +0000[diff] [blame]889data is to be laid out in memory. The syntax for the data layout is simply:</p>
890<pre> target datalayout = "<i>layout specification</i>"</pre>
891<p>The <i>layout specification</i> consists of a list of specifications
892separated by the minus sign character ('-'). Each specification starts with a
893letter and may include other information after the letter to define some
894aspect of the data layout. The specifications accepted are as follows: </p>
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]895<dl>
896 <dt><tt>E</tt></dt>
897 <dd>Specifies that the target lays out data in big-endian form. That is, the
898 bits with the most significance have the lowest address location.</dd>
899 <dt><tt>e</tt></dt>
900 <dd>Specifies that hte target lays out data in little-endian form. That is,
901 the bits with the least significance have the lowest address location.</dd>
902 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
903 <dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
904 <i>preferred</i> alignments. All sizes are in bits. Specifying the <i>pref</i>
905 alignment is optional. If omitted, the preceding <tt>:</tt> should be omitted
906 too.</dd>
907 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
908 <dd>This specifies the alignment for an integer type of a given bit
909 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
910 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
911 <dd>This specifies the alignment for a vector type of a given bit
912 <i>size</i>.</dd>
913 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
914 <dd>This specifies the alignment for a floating point type of a given bit
915 <i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
916 (double).</dd>
917 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
918 <dd>This specifies the alignment for an aggregate type of a given bit
919 <i>size</i>.</dd>
920</dl>
921<p>When constructing the data layout for a given target, LLVM starts with a
922default set of specifications which are then (possibly) overriden by the
923specifications in the <tt>datalayout</tt> keyword. The default specifications
924are given in this list:</p>
925<ul>
926 <li><tt>E</tt> - big endian</li>
927 <li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
928 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
929 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
930 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
931 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
932 <li><tt>i64:32:64</tt> - i64 has abi alignment of 32-bits but preferred
933 alignment of 64-bits</li>
934 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
935 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
936 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
937 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
938 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
939</ul>
940<p>When llvm is determining the alignment for a given type, it uses the
941following rules:
942<ol>
943 <li>If the type sought is an exact match for one of the specifications, that
944 specification is used.</li>
945 <li>If no match is found, and the type sought is an integer type, then the
946 smallest integer type that is larger than the bitwidth of the sought type is
947 used. If none of the specifications are larger than the bitwidth then the the
948 largest integer type is used. For example, given the default specifications
949 above, the i7 type will use the alignment of i8 (next largest) while both
950 i65 and i256 will use the alignment of i64 (largest specified).</li>
951 <li>If no match is found, and the type sought is a vector type, then the
952 largest vector type that is smaller than the sought vector type will be used
953 as a fall back. This happens because <128 x double> can be implemented in
954 terms of 64 <2 x double>, for example.</li>
955</ol>
956</div>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]957
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]958<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]959<div class="doc_section"> <a name="typesystem">Type System</a> </div>
960<!-- *********************************************************************** -->
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]961
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]962<div class="doc_text">
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]963
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]964<p>The LLVM type system is one of the most important features of the
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]965intermediate representation. Being typed enables a number of
966optimizations to be performed on the IR directly, without having to do
967extra analyses on the side before the transformation. A strong type
968system makes it easier to read the generated code and enables novel
969analyses and transformations that are not feasible to perform on normal
970three address code representations.</p>
Chris Lattner6af02f32004-12-09 16:11:40 +0000[diff] [blame]971
972</div>
973
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]974<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]975<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]976<div class="doc_text">
John Criswell417228d2004-04-09 16:48:45 +0000[diff] [blame]977<p>The primitive types are the fundamental building blocks of the LLVM
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]978system. The current set of primitive types is as follows:</p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]979
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]980<table class="layout">
981 <tr class="layout">
982 <td class="left">
983 <table>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]984 <tbody>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]985 <tr><th>Type</th><th>Description</th></tr>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]986 <tr><td><tt><a name="t_void">void</a></tt></td><td>No value</td></tr>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]987 <tr><td><tt>label</tt></td><td>Branch destination</td></tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]988 </tbody>
989 </table>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]990 </td>
991 <td class="right">
992 <table>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]993 <tbody>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]994 <tr><th>Type</th><th>Description</th></tr>
Reid Spencer138249b2007-05-16 18:44:01 +0000[diff] [blame]995 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]996 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]997 </tbody>
998 </table>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]999 </td>
1000 </tr>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1001</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1002</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1003
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1004<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1005<div class="doc_subsubsection"> <a name="t_classifications">Type
1006Classifications</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1007<div class="doc_text">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1008<p>These different primitive types fall into a few useful
1009classifications:</p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]1010
1011<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1012 <tbody>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1013 <tr><th>Classification</th><th>Types</th></tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1014 <tr>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1015 <td><a name="t_integer">integer</a></td>
Reid Spencer138249b2007-05-16 18:44:01 +0000[diff] [blame]1016 <td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1017 </tr>
1018 <tr>
1019 <td><a name="t_floating">floating point</a></td>
1020 <td><tt>float, double</tt></td>
1021 </tr>
1022 <tr>
1023 <td><a name="t_firstclass">first class</a></td>
Reid Spencer138249b2007-05-16 18:44:01 +0000[diff] [blame]1024 <td><tt>i1, ..., float, double, <br/>
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1025 <a href="#t_pointer">pointer</a>,<a href="#t_vector">vector</a></tt>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1026 </td>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1027 </tr>
1028 </tbody>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1029</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]1030
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1031<p>The <a href="#t_firstclass">first class</a> types are perhaps the
1032most important. Values of these types are the only ones which can be
1033produced by instructions, passed as arguments, or used as operands to
1034instructions. This means that all structures and arrays must be
1035manipulated either by pointer or by component.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1036</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1037
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1038<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1039<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1040
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1041<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1042
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1043<p>The real power in LLVM comes from the derived types in the system.
1044This is what allows a programmer to represent arrays, functions,
1045pointers, and other useful types. Note that these derived types may be
1046recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1047
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1048</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1049
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1050<!-- _______________________________________________________________________ -->
Reid Spencer138249b2007-05-16 18:44:01 +0000[diff] [blame]1051<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
1052
1053<div class="doc_text">
1054
1055<h5>Overview:</h5>
1056<p>The integer type is a very simple derived type that simply specifies an
1057arbitrary bit width for the integer type desired. Any bit width from 1 bit to
10582^23-1 (about 8 million) can be specified.</p>
1059
1060<h5>Syntax:</h5>
1061
1062<pre>
1063 iN
1064</pre>
1065
1066<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
1067value.</p>
1068
1069<h5>Examples:</h5>
1070<table class="layout">
1071 <tr class="layout">
1072 <td class="left">
1073 <tt>i1</tt><br/>
1074 <tt>i4</tt><br/>
1075 <tt>i8</tt><br/>
1076 <tt>i16</tt><br/>
1077 <tt>i32</tt><br/>
1078 <tt>i42</tt><br/>
1079 <tt>i64</tt><br/>
1080 <tt>i1942652</tt><br/>
1081 </td>
1082 <td class="left">
1083 A boolean integer of 1 bit<br/>
1084 A nibble sized integer of 4 bits.<br/>
1085 A byte sized integer of 8 bits.<br/>
1086 A half word sized integer of 16 bits.<br/>
1087 A word sized integer of 32 bits.<br/>
1088 An integer whose bit width is the answer. <br/>
1089 A double word sized integer of 64 bits.<br/>
1090 A really big integer of over 1 million bits.<br/>
1091 </td>
1092 </tr>
1093</table>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1094</div>
Reid Spencer138249b2007-05-16 18:44:01 +0000[diff] [blame]1095
1096<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1097<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1098
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1099<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1100
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1101<h5>Overview:</h5>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1102
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1103<p>The array type is a very simple derived type that arranges elements
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1104sequentially in memory. The array type requires a size (number of
1105elements) and an underlying data type.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1106
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1107<h5>Syntax:</h5>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1108
1109<pre>
1110 [&lt;# elements&gt; x &lt;elementtype&gt;]
1111</pre>
1112
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1113<p>The number of elements is a constant integer value; elementtype may
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1114be any type with a size.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1115
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1116<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1117<table class="layout">
1118 <tr class="layout">
1119 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1120 <tt>[40 x i32 ]</tt><br/>
1121 <tt>[41 x i32 ]</tt><br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1122 <tt>[40 x i8]</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1123 </td>
1124 <td class="left">
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1125 Array of 40 32-bit integer values.<br/>
1126 Array of 41 32-bit integer values.<br/>
1127 Array of 40 8-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1128 </td>
1129 </tr>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1130</table>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1131<p>Here are some examples of multidimensional arrays:</p>
1132<table class="layout">
1133 <tr class="layout">
1134 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1135 <tt>[3 x [4 x i32]]</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1136 <tt>[12 x [10 x float]]</tt><br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1137 <tt>[2 x [3 x [4 x i16]]]</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1138 </td>
1139 <td class="left">
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1140 3x4 array of 32-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1141 12x10 array of single precision floating point values.<br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1142 2x3x4 array of 16-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1143 </td>
1144 </tr>
1145</table>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]1146
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]1147<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
1148length array. Normally, accesses past the end of an array are undefined in
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]1149LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
1150As a special case, however, zero length arrays are recognized to be variable
1151length. This allows implementation of 'pascal style arrays' with the LLVM
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1152type "{ i32, [0 x float]}", for example.</p>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]1153
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1154</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1155
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1156<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1157<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1158<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1159<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1160<p>The function type can be thought of as a function signature. It
1161consists of a return type and a list of formal parameter types.
John Criswella0d50d22003-11-25 21:45:46 +0000[diff] [blame]1162Function types are usually used to build virtual function tables
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1163(which are structures of pointers to functions), for indirect function
1164calls, and when defining a function.</p>
John Criswella0d50d22003-11-25 21:45:46 +0000[diff] [blame]1165<p>
1166The return type of a function type cannot be an aggregate type.
1167</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1168<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1169<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]1170<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
Misha Brukman20f9a622004-08-12 20:16:08 +0000[diff] [blame]1171specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1172which indicates that the function takes a variable number of arguments.
1173Variable argument functions can access their arguments with the <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1174 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1175<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1176<table class="layout">
1177 <tr class="layout">
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1178 <td class="left"><tt>i32 (i32)</tt></td>
1179 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1180 </td>
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1181 </tr><tr class="layout">
Reid Spencer314e1cb2007-07-19 23:13:04 +0000[diff] [blame]1182 <td class="left"><tt>float&nbsp;(i16&nbsp;signext,&nbsp;i32&nbsp;*)&nbsp;*
Reid Spencer655dcc62006-12-31 07:20:23 +0000[diff] [blame]1183 </tt></td>
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1184 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
1185 an <tt>i16</tt> that should be sign extended and a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1186 <a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1187 <tt>float</tt>.
1188 </td>
1189 </tr><tr class="layout">
1190 <td class="left"><tt>i32 (i8*, ...)</tt></td>
1191 <td class="left">A vararg function that takes at least one
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1192 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
Reid Spencer58c08712006-12-31 07:18:34 +0000[diff] [blame]1193 which returns an integer. This is the signature for <tt>printf</tt> in
1194 LLVM.
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1195 </td>
1196 </tr>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1197</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]1198
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1199</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1200<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1201<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1202<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1203<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1204<p>The structure type is used to represent a collection of data members
1205together in memory. The packing of the field types is defined to match
1206the ABI of the underlying processor. The elements of a structure may
1207be any type that has a size.</p>
1208<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1209and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1210field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1211instruction.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1212<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1213<pre> { &lt;type list&gt; }<br></pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1214<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1215<table class="layout">
1216 <tr class="layout">
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]1217 <td class="left"><tt>{ i32, i32, i32 }</tt></td>
1218 <td class="left">A triple of three <tt>i32</tt> values</td>
1219 </tr><tr class="layout">
1220 <td class="left"><tt>{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}</tt></td>
1221 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1222 second element is a <a href="#t_pointer">pointer</a> to a
1223 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1224 an <tt>i32</tt>.</td>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1225 </tr>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1226</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1227</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1228
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1229<!-- _______________________________________________________________________ -->
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]1230<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1231</div>
1232<div class="doc_text">
1233<h5>Overview:</h5>
1234<p>The packed structure type is used to represent a collection of data members
1235together in memory. There is no padding between fields. Further, the alignment
1236of a packed structure is 1 byte. The elements of a packed structure may
1237be any type that has a size.</p>
1238<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1239and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1240field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1241instruction.</p>
1242<h5>Syntax:</h5>
1243<pre> &lt; { &lt;type list&gt; } &gt; <br></pre>
1244<h5>Examples:</h5>
1245<table class="layout">
1246 <tr class="layout">
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]1247 <td class="left"><tt>&lt; { i32, i32, i32 } &gt;</tt></td>
1248 <td class="left">A triple of three <tt>i32</tt> values</td>
1249 </tr><tr class="layout">
1250 <td class="left"><tt>&lt;&nbsp;{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}&nbsp;&gt;</tt></td>
1251 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1252 second element is a <a href="#t_pointer">pointer</a> to a
1253 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1254 an <tt>i32</tt>.</td>
Andrew Lenharth8df88e22006-12-08 17:13:00 +0000[diff] [blame]1255 </tr>
1256</table>
1257</div>
1258
1259<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1260<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1261<div class="doc_text">
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1262<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1263<p>As in many languages, the pointer type represents a pointer or
Christopher Lamb308121c2007-12-11 09:31:00 +0000[diff] [blame^]1264reference to another object, which must live in memory. Pointer types may have
1265an optional address space attribute defining the target-specific numbered
1266address space where the pointed-to object resides. The default address space is
1267zero.</p>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1268<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1269<pre> &lt;type&gt; *<br></pre>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1270<h5>Examples:</h5>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1271<table class="layout">
1272 <tr class="layout">
1273 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1274 <tt>[4x i32]*</tt><br/>
1275 <tt>i32 (i32 *) *</tt><br/>
Christopher Lamb308121c2007-12-11 09:31:00 +0000[diff] [blame^]1276 <tt>i32 addrspace(5)*</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1277 </td>
1278 <td class="left">
1279 A <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a> of
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1280 four <tt>i32</tt> values<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1281 A <a href="#t_pointer">pointer</a> to a <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1282 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
1283 <tt>i32</tt>.<br/>
Christopher Lamb308121c2007-12-11 09:31:00 +0000[diff] [blame^]1284 A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value that resides
1285 in address space 5.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1286 </td>
1287 </tr>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1288</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1289</div>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1290
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1291<!-- _______________________________________________________________________ -->
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1292<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1293<div class="doc_text">
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1294
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1295<h5>Overview:</h5>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1296
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1297<p>A vector type is a simple derived type that represents a vector
1298of elements. Vector types are used when multiple primitive data
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1299are operated in parallel using a single instruction (SIMD).
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1300A vector type requires a size (number of
Chris Lattner330ce692005-11-10 01:44:22 +0000[diff] [blame]1301elements) and an underlying primitive data type. Vectors must have a power
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1302of two length (1, 2, 4, 8, 16 ...). Vector types are
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1303considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1304
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1305<h5>Syntax:</h5>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1306
1307<pre>
1308 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1309</pre>
1310
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1311<p>The number of elements is a constant integer value; elementtype may
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1312be any integer or floating point type.</p>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1313
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1314<h5>Examples:</h5>
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1315
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1316<table class="layout">
1317 <tr class="layout">
1318 <td class="left">
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1319 <tt>&lt;4 x i32&gt;</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1320 <tt>&lt;8 x float&gt;</tt><br/>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1321 <tt>&lt;2 x i64&gt;</tt><br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1322 </td>
1323 <td class="left">
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1324 Vector of 4 32-bit integer values.<br/>
1325 Vector of 8 floating-point values.<br/>
1326 Vector of 2 64-bit integer values.<br/>
Reid Spencerc3c4c4f2004-11-01 08:19:36 +0000[diff] [blame]1327 </td>
1328 </tr>
1329</table>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1330</div>
1331
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1332<!-- _______________________________________________________________________ -->
1333<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1334<div class="doc_text">
1335
1336<h5>Overview:</h5>
1337
1338<p>Opaque types are used to represent unknown types in the system. This
Gordon Henriksena699c4d2007-10-14 00:34:53 +0000[diff] [blame]1339corresponds (for example) to the C notion of a forward declared structure type.
Chris Lattner37b6b092005-04-25 17:34:15 +0000[diff] [blame]1340In LLVM, opaque types can eventually be resolved to any type (not just a
1341structure type).</p>
1342
1343<h5>Syntax:</h5>
1344
1345<pre>
1346 opaque
1347</pre>
1348
1349<h5>Examples:</h5>
1350
1351<table class="layout">
1352 <tr class="layout">
1353 <td class="left">
1354 <tt>opaque</tt>
1355 </td>
1356 <td class="left">
1357 An opaque type.<br/>
1358 </td>
1359 </tr>
1360</table>
1361</div>
1362
1363
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1364<!-- *********************************************************************** -->
1365<div class="doc_section"> <a name="constants">Constants</a> </div>
1366<!-- *********************************************************************** -->
1367
1368<div class="doc_text">
1369
1370<p>LLVM has several different basic types of constants. This section describes
1371them all and their syntax.</p>
1372
1373</div>
1374
1375<!-- ======================================================================= -->
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]1376<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1377
1378<div class="doc_text">
1379
1380<dl>
1381 <dt><b>Boolean constants</b></dt>
1382
1383 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1384 constants of the <tt><a href="#t_primitive">i1</a></tt> type.
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1385 </dd>
1386
1387 <dt><b>Integer constants</b></dt>
1388
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]1389 <dd>Standard integers (such as '4') are constants of the <a
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]1390 href="#t_integer">integer</a> type. Negative numbers may be used with
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1391 integer types.
1392 </dd>
1393
1394 <dt><b>Floating point constants</b></dt>
1395
1396 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
1397 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1398 notation (see below). Floating point constants must have a <a
1399 href="#t_floating">floating point</a> type. </dd>
1400
1401 <dt><b>Null pointer constants</b></dt>
1402
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1403 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1404 and must be of <a href="#t_pointer">pointer type</a>.</dd>
1405
1406</dl>
1407
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1408<p>The one non-intuitive notation for constants is the optional hexadecimal form
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1409of floating point constants. For example, the form '<tt>double
14100x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
14114.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencer8f08d802004-12-09 18:02:53 +0000[diff] [blame]1412(and the only time that they are generated by the disassembler) is when a
1413floating point constant must be emitted but it cannot be represented as a
1414decimal floating point number. For example, NaN's, infinities, and other
1415special values are represented in their IEEE hexadecimal format so that
1416assembly and disassembly do not cause any bits to change in the constants.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1417
1418</div>
1419
1420<!-- ======================================================================= -->
1421<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
1422</div>
1423
1424<div class="doc_text">
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]1425<p>Aggregate constants arise from aggregation of simple constants
1426and smaller aggregate constants.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1427
1428<dl>
1429 <dt><b>Structure constants</b></dt>
1430
1431 <dd>Structure constants are represented with notation similar to structure
1432 type definitions (a comma separated list of elements, surrounded by braces
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1433 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* %G }</tt>",
Chris Lattner0a2d0992007-07-13 20:01:46 +0000[diff] [blame]1434 where "<tt>%G</tt>" is declared as "<tt>@G = external global i32</tt>". Structure constants
Chris Lattner455fc8c2005-03-07 22:13:59 +0000[diff] [blame]1435 must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1436 types of elements must match those specified by the type.
1437 </dd>
1438
1439 <dt><b>Array constants</b></dt>
1440
1441 <dd>Array constants are represented with notation similar to array type
1442 definitions (a comma separated list of elements, surrounded by square brackets
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1443 (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74 ]</tt>". Array
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1444 constants must have <a href="#t_array">array type</a>, and the number and
1445 types of elements must match those specified by the type.
1446 </dd>
1447
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1448 <dt><b>Vector constants</b></dt>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1449
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1450 <dd>Vector constants are represented with notation similar to vector type
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1451 definitions (a comma separated list of elements, surrounded by
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1452 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32 42,
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]1453 i32 11, i32 74, i32 100 &gt;</tt>". Vector constants must have <a
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1454 href="#t_vector">vector type</a>, and the number and types of elements must
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1455 match those specified by the type.
1456 </dd>
1457
1458 <dt><b>Zero initialization</b></dt>
1459
1460 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
1461 value to zero of <em>any</em> type, including scalar and aggregate types.
1462 This is often used to avoid having to print large zero initializers (e.g. for
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]1463 large arrays) and is always exactly equivalent to using explicit zero
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1464 initializers.
1465 </dd>
1466</dl>
1467
1468</div>
1469
1470<!-- ======================================================================= -->
1471<div class="doc_subsection">
1472 <a name="globalconstants">Global Variable and Function Addresses</a>
1473</div>
1474
1475<div class="doc_text">
1476
1477<p>The addresses of <a href="#globalvars">global variables</a> and <a
1478href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1479constants. These constants are explicitly referenced when the <a
1480href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1481href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1482file:</p>
1483
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1484<div class="doc_code">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1485<pre>
Chris Lattner00538a12007-06-06 18:28:13 +0000[diff] [blame]1486@X = global i32 17
1487@Y = global i32 42
1488@Z = global [2 x i32*] [ i32* @X, i32* @Y ]
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1489</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1490</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1491
1492</div>
1493
1494<!-- ======================================================================= -->
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1495<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1496<div class="doc_text">
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1497 <p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1498 no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1499 a constant is permitted.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1500
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1501 <p>Undefined values indicate to the compiler that the program is well defined
1502 no matter what value is used, giving the compiler more freedom to optimize.
1503 </p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1504</div>
1505
1506<!-- ======================================================================= -->
1507<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1508</div>
1509
1510<div class="doc_text">
1511
1512<p>Constant expressions are used to allow expressions involving other constants
1513to be used as constants. Constant expressions may be of any <a
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1514href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1515that does not have side effects (e.g. load and call are not supported). The
1516following is the syntax for constant expressions:</p>
1517
1518<dl>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1519 <dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
1520 <dd>Truncate a constant to another type. The bit size of CST must be larger
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1521 than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1522
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1523 <dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
1524 <dd>Zero extend a constant to another type. The bit size of CST must be
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1525 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1526
1527 <dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
1528 <dd>Sign extend a constant to another type. The bit size of CST must be
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]1529 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1530
1531 <dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
1532 <dd>Truncate a floating point constant to another floating point type. The
1533 size of CST must be larger than the size of TYPE. Both types must be
1534 floating point.</dd>
1535
1536 <dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
1537 <dd>Floating point extend a constant to another type. The size of CST must be
1538 smaller or equal to the size of TYPE. Both types must be floating point.</dd>
1539
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]1540 <dt><b><tt>fptoui ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1541 <dd>Convert a floating point constant to the corresponding unsigned integer
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]1542 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1543 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1544 of the same number of elements. If the value won't fit in the integer type,
1545 the results are undefined.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1546
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]1547 <dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1548 <dd>Convert a floating point constant to the corresponding signed integer
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]1549 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1550 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1551 of the same number of elements. If the value won't fit in the integer type,
1552 the results are undefined.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1553
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]1554 <dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1555 <dd>Convert an unsigned integer constant to the corresponding floating point
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]1556 constant. TYPE must be a scalar or vector floating point type. CST must be of
1557 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1558 of the same number of elements. If the value won't fit in the floating point
1559 type, the results are undefined.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1560
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]1561 <dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1562 <dd>Convert a signed integer constant to the corresponding floating point
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]1563 constant. TYPE must be a scalar or vector floating point type. CST must be of
1564 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1565 of the same number of elements. If the value won't fit in the floating point
1566 type, the results are undefined.</dd>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1567
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]1568 <dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
1569 <dd>Convert a pointer typed constant to the corresponding integer constant
1570 TYPE must be an integer type. CST must be of pointer type. The CST value is
1571 zero extended, truncated, or unchanged to make it fit in TYPE.</dd>
1572
1573 <dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
1574 <dd>Convert a integer constant to a pointer constant. TYPE must be a
1575 pointer type. CST must be of integer type. The CST value is zero extended,
1576 truncated, or unchanged to make it fit in a pointer size. This one is
1577 <i>really</i> dangerous!</dd>
1578
1579 <dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1580 <dd>Convert a constant, CST, to another TYPE. The size of CST and TYPE must be
1581 identical (same number of bits). The conversion is done as if the CST value
1582 was stored to memory and read back as TYPE. In other words, no bits change
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]1583 with this operator, just the type. This can be used for conversion of
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1584 vector types to any other type, as long as they have the same bit width. For
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]1585 pointers it is only valid to cast to another pointer type.
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]1586 </dd>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1587
1588 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1589
1590 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1591 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1592 instruction, the index list may have zero or more indexes, which are required
1593 to make sense for the type of "CSTPTR".</dd>
1594
Robert Bocchino7e97a6d2006-01-10 19:31:34 +0000[diff] [blame]1595 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
1596
1597 <dd>Perform the <a href="#i_select">select operation</a> on
Reid Spencer9965ee72006-12-04 19:23:19 +0000[diff] [blame]1598 constants.</dd>
1599
1600 <dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
1601 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
1602
1603 <dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
1604 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino7e97a6d2006-01-10 19:31:34 +0000[diff] [blame]1605
1606 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
1607
1608 <dd>Perform the <a href="#i_extractelement">extractelement
1609 operation</a> on constants.
1610
Robert Bocchinof72fdfe2006-01-15 20:48:27 +0000[diff] [blame]1611 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
1612
1613 <dd>Perform the <a href="#i_insertelement">insertelement
Reid Spencer9965ee72006-12-04 19:23:19 +0000[diff] [blame]1614 operation</a> on constants.</dd>
Robert Bocchinof72fdfe2006-01-15 20:48:27 +0000[diff] [blame]1615
Chris Lattner016a0e52006-04-08 00:13:41 +0000[diff] [blame]1616
1617 <dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
1618
1619 <dd>Perform the <a href="#i_shufflevector">shufflevector
Reid Spencer9965ee72006-12-04 19:23:19 +0000[diff] [blame]1620 operation</a> on constants.</dd>
Chris Lattner016a0e52006-04-08 00:13:41 +0000[diff] [blame]1621
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1622 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1623
Reid Spencer641f5c92004-12-09 18:13:12 +0000[diff] [blame]1624 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1625 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1626 binary</a> operations. The constraints on operands are the same as those for
1627 the corresponding instruction (e.g. no bitwise operations on floating point
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1628 values are allowed).</dd>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1629</dl>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1630</div>
Chris Lattnerb1652612004-03-08 16:49:10 +0000[diff] [blame]1631
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1632<!-- *********************************************************************** -->
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1633<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
1634<!-- *********************************************************************** -->
1635
1636<!-- ======================================================================= -->
1637<div class="doc_subsection">
1638<a name="inlineasm">Inline Assembler Expressions</a>
1639</div>
1640
1641<div class="doc_text">
1642
1643<p>
1644LLVM supports inline assembler expressions (as opposed to <a href="#moduleasm">
1645Module-Level Inline Assembly</a>) through the use of a special value. This
1646value represents the inline assembler as a string (containing the instructions
1647to emit), a list of operand constraints (stored as a string), and a flag that
1648indicates whether or not the inline asm expression has side effects. An example
1649inline assembler expression is:
1650</p>
1651
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1652<div class="doc_code">
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1653<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1654i32 (i32) asm "bswap $0", "=r,r"
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1655</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1656</div>
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1657
1658<p>
1659Inline assembler expressions may <b>only</b> be used as the callee operand of
1660a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we have:
1661</p>
1662
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1663<div class="doc_code">
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1664<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1665%X = call i32 asm "<a href="#int_bswap">bswap</a> $0", "=r,r"(i32 %Y)
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1666</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1667</div>
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1668
1669<p>
1670Inline asms with side effects not visible in the constraint list must be marked
1671as having side effects. This is done through the use of the
1672'<tt>sideeffect</tt>' keyword, like so:
1673</p>
1674
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1675<div class="doc_code">
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1676<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1677call void asm sideeffect "eieio", ""()
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1678</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]1679</div>
Chris Lattner98f013c2006-01-25 23:47:57 +0000[diff] [blame]1680
1681<p>TODO: The format of the asm and constraints string still need to be
1682documented here. Constraints on what can be done (e.g. duplication, moving, etc
1683need to be documented).
1684</p>
1685
1686</div>
1687
1688<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1689<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1690<!-- *********************************************************************** -->
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1691
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1692<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1693
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1694<p>The LLVM instruction set consists of several different
1695classifications of instructions: <a href="#terminators">terminator
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1696instructions</a>, <a href="#binaryops">binary instructions</a>,
1697<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1698 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1699instructions</a>.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1700
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1701</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1702
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1703<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1704<div class="doc_subsection"> <a name="terminators">Terminator
1705Instructions</a> </div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1706
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1707<div class="doc_text">
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1708
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1709<p>As mentioned <a href="#functionstructure">previously</a>, every
1710basic block in a program ends with a "Terminator" instruction, which
1711indicates which block should be executed after the current block is
1712finished. These terminator instructions typically yield a '<tt>void</tt>'
1713value: they produce control flow, not values (the one exception being
1714the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1715<p>There are six different terminator instructions: the '<a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1716 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1717instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1718the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1719 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1720 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1721
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1722</div>
Chris Lattner74d3f822004-12-09 17:30:23 +0000[diff] [blame]1723
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1724<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1725<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1726Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1727<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1728<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1729<pre> ret &lt;type&gt; &lt;value&gt; <i>; Return a value from a non-void function</i>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1730 ret void <i>; Return from void function</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1731</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1732<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1733<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]1734value) from a function back to the caller.</p>
John Criswell417228d2004-04-09 16:48:45 +0000[diff] [blame]1735<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1736returns a value and then causes control flow, and one that just causes
1737control flow to occur.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1738<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1739<p>The '<tt>ret</tt>' instruction may return any '<a
1740 href="#t_firstclass">first class</a>' type. Notice that a function is
1741not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1742instruction inside of the function that returns a value that does not
1743match the return type of the function.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1744<h5>Semantics:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1745<p>When the '<tt>ret</tt>' instruction is executed, control flow
1746returns back to the calling function's context. If the caller is a "<a
John Criswell40db33f2004-06-25 15:16:57 +0000[diff] [blame]1747 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1748the instruction after the call. If the caller was an "<a
1749 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1750at the beginning of the "normal" destination block. If the instruction
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1751returns a value, that value shall set the call or invoke instruction's
1752return value.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1753<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1754<pre> ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]1755 ret void <i>; Return from a void function</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1756</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1757</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1758<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1759<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1760<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1761<h5>Syntax:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1762<pre> br i1 &lt;cond&gt;, label &lt;iftrue&gt;, label &lt;iffalse&gt;<br> br label &lt;dest&gt; <i>; Unconditional branch</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1763</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1764<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1765<p>The '<tt>br</tt>' instruction is used to cause control flow to
1766transfer to a different basic block in the current function. There are
1767two forms of this instruction, corresponding to a conditional branch
1768and an unconditional branch.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1769<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1770<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1771single '<tt>i1</tt>' value and two '<tt>label</tt>' values. The
Reid Spencer50c723a2007-02-19 23:54:10 +0000[diff] [blame]1772unconditional form of the '<tt>br</tt>' instruction takes a single
1773'<tt>label</tt>' value as a target.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1774<h5>Semantics:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1775<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1776argument is evaluated. If the value is <tt>true</tt>, control flows
1777to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1778control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1779<h5>Example:</h5>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1780<pre>Test:<br> %cond = <a href="#i_icmp">icmp</a> eq, i32 %a, %b<br> br i1 %cond, label %IfEqual, label %IfUnequal<br>IfEqual:<br> <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1781 href="#i_ret">ret</a> i32 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> i32 0<br></pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1782</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1783<!-- _______________________________________________________________________ -->
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1784<div class="doc_subsubsection">
1785 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1786</div>
1787
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1788<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1789<h5>Syntax:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1790
1791<pre>
1792 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1793</pre>
1794
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1795<h5>Overview:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1796
1797<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1798several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1799instruction, allowing a branch to occur to one of many possible
1800destinations.</p>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1801
1802
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1803<h5>Arguments:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1804
1805<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1806comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1807an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1808table is not allowed to contain duplicate constant entries.</p>
1809
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1810<h5>Semantics:</h5>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1811
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1812<p>The <tt>switch</tt> instruction specifies a table of values and
1813destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswellbcbb18c2004-06-25 16:05:06 +0000[diff] [blame]1814table is searched for the given value. If the value is found, control flow is
1815transfered to the corresponding destination; otherwise, control flow is
1816transfered to the default destination.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1817
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1818<h5>Implementation:</h5>
1819
1820<p>Depending on properties of the target machine and the particular
1821<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswellbcbb18c2004-06-25 16:05:06 +0000[diff] [blame]1822ways. For example, it could be generated as a series of chained conditional
1823branches or with a lookup table.</p>
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1824
1825<h5>Example:</h5>
1826
1827<pre>
1828 <i>; Emulate a conditional br instruction</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]1829 %Val = <a href="#i_zext">zext</a> i1 %value to i32
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1830 switch i32 %Val, label %truedest [i32 0, label %falsedest ]
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1831
1832 <i>; Emulate an unconditional br instruction</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1833 switch i32 0, label %dest [ ]
Chris Lattnercf96c6c2004-02-24 04:54:45 +0000[diff] [blame]1834
1835 <i>; Implement a jump table:</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]1836 switch i32 %val, label %otherwise [ i32 0, label %onzero
1837 i32 1, label %onone
1838 i32 2, label %ontwo ]
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1839</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1840</div>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1841
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1842<!-- _______________________________________________________________________ -->
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1843<div class="doc_subsubsection">
1844 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1845</div>
1846
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1847<div class="doc_text">
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1848
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1849<h5>Syntax:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1850
1851<pre>
1852 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)
Chris Lattner6b7a0082006-05-14 18:23:06 +0000[diff] [blame]1853 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1854</pre>
1855
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]1856<h5>Overview:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1857
1858<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1859function, with the possibility of control flow transfer to either the
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1860'<tt>normal</tt>' label or the
1861'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1862"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1863"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswell02fdc6f2005-05-12 16:52:32 +0000[diff] [blame]1864href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1865continued at the dynamically nearest "exception" label.</p>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1866
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1867<h5>Arguments:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1868
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1869<p>This instruction requires several arguments:</p>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1870
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1871<ol>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1872 <li>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]1873 The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1874 convention</a> the call should use. If none is specified, the call defaults
1875 to using C calling conventions.
1876 </li>
1877 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1878 function value being invoked. In most cases, this is a direct function
1879 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1880 an arbitrary pointer to function value.
1881 </li>
1882
1883 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1884 function to be invoked. </li>
1885
1886 <li>'<tt>function args</tt>': argument list whose types match the function
1887 signature argument types. If the function signature indicates the function
1888 accepts a variable number of arguments, the extra arguments can be
1889 specified. </li>
1890
1891 <li>'<tt>normal label</tt>': the label reached when the called function
1892 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1893
1894 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1895 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1896
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1897</ol>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1898
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1899<h5>Semantics:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1900
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1901<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1902href="#i_call">call</a></tt>' instruction in most regards. The primary
1903difference is that it establishes an association with a label, which is used by
1904the runtime library to unwind the stack.</p>
1905
1906<p>This instruction is used in languages with destructors to ensure that proper
1907cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1908exception. Additionally, this is important for implementation of
1909'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1910
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1911<h5>Example:</h5>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]1912<pre>
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]1913 %retval = invoke i32 %Test(i32 15) to label %Continue
1914 unwind label %TestCleanup <i>; {i32}:retval set</i>
1915 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Test(i32 15) to label %Continue
1916 unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1917</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1918</div>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1919
1920
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1921<!-- _______________________________________________________________________ -->
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1922
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1923<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1924Instruction</a> </div>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1925
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1926<div class="doc_text">
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1927
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1928<h5>Syntax:</h5>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1929<pre>
1930 unwind
1931</pre>
1932
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1933<h5>Overview:</h5>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1934
1935<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1936at the first callee in the dynamic call stack which used an <a
1937href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1938primarily used to implement exception handling.</p>
1939
Chris Lattner5ed60612003-09-03 00:41:47 +0000[diff] [blame]1940<h5>Semantics:</h5>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1941
1942<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
1943immediately halt. The dynamic call stack is then searched for the first <a
1944href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
1945execution continues at the "exceptional" destination block specified by the
1946<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
1947dynamic call chain, undefined behavior results.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1948</div>
Chris Lattner08b7d5b2004-10-16 18:04:13 +0000[diff] [blame]1949
1950<!-- _______________________________________________________________________ -->
1951
1952<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
1953Instruction</a> </div>
1954
1955<div class="doc_text">
1956
1957<h5>Syntax:</h5>
1958<pre>
1959 unreachable
1960</pre>
1961
1962<h5>Overview:</h5>
1963
1964<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
1965instruction is used to inform the optimizer that a particular portion of the
1966code is not reachable. This can be used to indicate that the code after a
1967no-return function cannot be reached, and other facts.</p>
1968
1969<h5>Semantics:</h5>
1970
1971<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
1972</div>
1973
1974
1975
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1976<!-- ======================================================================= -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1977<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1978<div class="doc_text">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1979<p>Binary operators are used to do most of the computation in a
1980program. They require two operands, execute an operation on them, and
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]1981produce a single value. The operands might represent
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1982multiple data, as is the case with the <a href="#t_vector">vector</a> data type.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1983The result value of a binary operator is not
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1984necessarily the same type as its operands.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1985<p>There are several different binary operators:</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1986</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1987<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1988<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
1989Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1990<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1991<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]1992<pre> &lt;result&gt; = add &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1993</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1994<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1995<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]1996<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]1997<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]1998 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]1999 This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]2000Both arguments must have identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2001<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2002<p>The value produced is the integer or floating point sum of the two
2003operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2004<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2005<pre> &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2006</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2007</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2008<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2009<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
2010Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2011<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2012<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2013<pre> &lt;result&gt; = sub &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2014</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2015<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2016<p>The '<tt>sub</tt>' instruction returns the difference of its two
2017operands.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2018<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
2019instruction present in most other intermediate representations.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2020<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2021<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2022 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]2023values.
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2024This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]2025Both arguments must have identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2026<h5>Semantics:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2027<p>The value produced is the integer or floating point difference of
2028the two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2029<h5>Example:</h5>
Bill Wendling2d8b9a82007-05-29 09:42:13 +0000[diff] [blame]2030<pre>
2031 &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2032 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2033</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2034</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2035<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2036<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
2037Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2038<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2039<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2040<pre> &lt;result&gt; = mul &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2041</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2042<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2043<p>The '<tt>mul</tt>' instruction returns the product of its two
2044operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2045<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2046<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2047 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]2048values.
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2049This instruction can also take <a href="#t_vector">vector</a> versions of the values.
Chris Lattnerc8cb6952004-08-12 19:12:28 +0000[diff] [blame]2050Both arguments must have identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2051<h5>Semantics:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2052<p>The value produced is the integer or floating point product of the
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2053two operands.</p>
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]2054<p>Because the operands are the same width, the result of an integer
2055multiplication is the same whether the operands should be deemed unsigned or
2056signed.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2057<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2058<pre> &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2059</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2060</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2061<!-- _______________________________________________________________________ -->
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2062<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
2063</a></div>
2064<div class="doc_text">
2065<h5>Syntax:</h5>
2066<pre> &lt;result&gt; = udiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2067</pre>
2068<h5>Overview:</h5>
2069<p>The '<tt>udiv</tt>' instruction returns the quotient of its two
2070operands.</p>
2071<h5>Arguments:</h5>
2072<p>The two arguments to the '<tt>udiv</tt>' instruction must be
2073<a href="#t_integer">integer</a> values. Both arguments must have identical
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2074types. This instruction can also take <a href="#t_vector">vector</a> versions
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2075of the values in which case the elements must be integers.</p>
2076<h5>Semantics:</h5>
2077<p>The value produced is the unsigned integer quotient of the two operands. This
2078instruction always performs an unsigned division operation, regardless of
2079whether the arguments are unsigned or not.</p>
2080<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2081<pre> &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2082</pre>
2083</div>
2084<!-- _______________________________________________________________________ -->
2085<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
2086</a> </div>
2087<div class="doc_text">
2088<h5>Syntax:</h5>
2089<pre> &lt;result&gt; = sdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2090</pre>
2091<h5>Overview:</h5>
2092<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two
2093operands.</p>
2094<h5>Arguments:</h5>
2095<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
2096<a href="#t_integer">integer</a> values. Both arguments must have identical
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2097types. This instruction can also take <a href="#t_vector">vector</a> versions
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2098of the values in which case the elements must be integers.</p>
2099<h5>Semantics:</h5>
2100<p>The value produced is the signed integer quotient of the two operands. This
2101instruction always performs a signed division operation, regardless of whether
2102the arguments are signed or not.</p>
2103<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2104<pre> &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2105</pre>
2106</div>
2107<!-- _______________________________________________________________________ -->
2108<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2109Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2110<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2111<h5>Syntax:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2112<pre> &lt;result&gt; = fdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2113</pre>
2114<h5>Overview:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2115<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2116operands.</p>
2117<h5>Arguments:</h5>
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2118<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2119<a href="#t_floating">floating point</a> values. Both arguments must have
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2120identical types. This instruction can also take <a href="#t_vector">vector</a>
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2121versions of floating point values.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2122<h5>Semantics:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2123<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2124<h5>Example:</h5>
Reid Spencer7e80b0b2006-10-26 06:15:43 +0000[diff] [blame]2125<pre> &lt;result&gt; = fdiv float 4.0, %var <i>; yields {float}:result = 4.0 / %var</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2126</pre>
2127</div>
2128<!-- _______________________________________________________________________ -->
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2129<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
2130</div>
2131<div class="doc_text">
2132<h5>Syntax:</h5>
2133<pre> &lt;result&gt; = urem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2134</pre>
2135<h5>Overview:</h5>
2136<p>The '<tt>urem</tt>' instruction returns the remainder from the
2137unsigned division of its two arguments.</p>
2138<h5>Arguments:</h5>
2139<p>The two arguments to the '<tt>urem</tt>' instruction must be
2140<a href="#t_integer">integer</a> values. Both arguments must have identical
Dan Gohman08143e32007-11-05 23:35:22 +0000[diff] [blame]2141types. This instruction can also take <a href="#t_vector">vector</a> versions
2142of the values in which case the elements must be integers.</p>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2143<h5>Semantics:</h5>
2144<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
2145This instruction always performs an unsigned division to get the remainder,
2146regardless of whether the arguments are unsigned or not.</p>
2147<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2148<pre> &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2149</pre>
2150
2151</div>
2152<!-- _______________________________________________________________________ -->
2153<div class="doc_subsubsection"> <a name="i_srem">'<tt>srem</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2154Instruction</a> </div>
2155<div class="doc_text">
2156<h5>Syntax:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2157<pre> &lt;result&gt; = srem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2158</pre>
2159<h5>Overview:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2160<p>The '<tt>srem</tt>' instruction returns the remainder from the
Dan Gohman08143e32007-11-05 23:35:22 +0000[diff] [blame]2161signed division of its two operands. This instruction can also take
2162<a href="#t_vector">vector</a> versions of the values in which case
2163the elements must be integers.</p>
2164</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2165<h5>Arguments:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2166<p>The two arguments to the '<tt>srem</tt>' instruction must be
2167<a href="#t_integer">integer</a> values. Both arguments must have identical
2168types.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2169<h5>Semantics:</h5>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2170<p>This instruction returns the <i>remainder</i> of a division (where the result
Reid Spencer806ad6a2007-03-24 22:23:39 +0000[diff] [blame]2171has the same sign as the dividend, <tt>var1</tt>), not the <i>modulo</i>
2172operator (where the result has the same sign as the divisor, <tt>var2</tt>) of
2173a value. For more information about the difference, see <a
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2174 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
Reid Spencer806ad6a2007-03-24 22:23:39 +0000[diff] [blame]2175Math Forum</a>. For a table of how this is implemented in various languages,
Reid Spencerdb3b93b2007-03-24 22:40:44 +0000[diff] [blame]2176please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
Reid Spencer806ad6a2007-03-24 22:23:39 +0000[diff] [blame]2177Wikipedia: modulo operation</a>.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2178<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2179<pre> &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2180</pre>
2181
2182</div>
2183<!-- _______________________________________________________________________ -->
2184<div class="doc_subsubsection"> <a name="i_frem">'<tt>frem</tt>'
2185Instruction</a> </div>
2186<div class="doc_text">
2187<h5>Syntax:</h5>
2188<pre> &lt;result&gt; = frem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2189</pre>
2190<h5>Overview:</h5>
2191<p>The '<tt>frem</tt>' instruction returns the remainder from the
2192division of its two operands.</p>
2193<h5>Arguments:</h5>
2194<p>The two arguments to the '<tt>frem</tt>' instruction must be
2195<a href="#t_floating">floating point</a> values. Both arguments must have
Dan Gohman08143e32007-11-05 23:35:22 +0000[diff] [blame]2196identical types. This instruction can also take <a href="#t_vector">vector</a>
2197versions of floating point values.</p>
Reid Spencer7eb55b32006-11-02 01:53:59 +0000[diff] [blame]2198<h5>Semantics:</h5>
2199<p>This instruction returns the <i>remainder</i> of a division.</p>
2200<h5>Example:</h5>
2201<pre> &lt;result&gt; = frem float 4.0, %var <i>; yields {float}:result = 4.0 % %var</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2202</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2203</div>
Robert Bocchino820bc75b2006-02-17 21:18:08 +0000[diff] [blame]2204
Reid Spencer2ab01932007-02-02 13:57:07 +0000[diff] [blame]2205<!-- ======================================================================= -->
2206<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
2207Operations</a> </div>
2208<div class="doc_text">
2209<p>Bitwise binary operators are used to do various forms of
2210bit-twiddling in a program. They are generally very efficient
2211instructions and can commonly be strength reduced from other
2212instructions. They require two operands, execute an operation on them,
2213and produce a single value. The resulting value of the bitwise binary
2214operators is always the same type as its first operand.</p>
2215</div>
2216
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2217<!-- _______________________________________________________________________ -->
2218<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
2219Instruction</a> </div>
2220<div class="doc_text">
2221<h5>Syntax:</h5>
2222<pre> &lt;result&gt; = shl &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2223</pre>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2224
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2225<h5>Overview:</h5>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2226
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2227<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
2228the left a specified number of bits.</p>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2229
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2230<h5>Arguments:</h5>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2231
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2232<p>Both arguments to the '<tt>shl</tt>' instruction must be the same <a
2233 href="#t_integer">integer</a> type.</p>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2234
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2235<h5>Semantics:</h5>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2236
2237<p>The value produced is <tt>var1</tt> * 2<sup><tt>var2</tt></sup>. If
2238<tt>var2</tt> is (statically or dynamically) equal to or larger than the number
2239of bits in <tt>var1</tt>, the result is undefined.</p>
2240
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2241<h5>Example:</h5><pre>
2242 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
2243 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
2244 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2245 &lt;result&gt; = shl i32 1, 32 <i>; undefined</i>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2246</pre>
2247</div>
2248<!-- _______________________________________________________________________ -->
2249<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
2250Instruction</a> </div>
2251<div class="doc_text">
2252<h5>Syntax:</h5>
2253<pre> &lt;result&gt; = lshr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2254</pre>
2255
2256<h5>Overview:</h5>
2257<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2258operand shifted to the right a specified number of bits with zero fill.</p>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2259
2260<h5>Arguments:</h5>
2261<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
2262<a href="#t_integer">integer</a> type.</p>
2263
2264<h5>Semantics:</h5>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2265
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2266<p>This instruction always performs a logical shift right operation. The most
2267significant bits of the result will be filled with zero bits after the
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2268shift. If <tt>var2</tt> is (statically or dynamically) equal to or larger than
2269the number of bits in <tt>var1</tt>, the result is undefined.</p>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2270
2271<h5>Example:</h5>
2272<pre>
2273 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
2274 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
2275 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
2276 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2277 &lt;result&gt; = lshr i32 1, 32 <i>; undefined</i>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2278</pre>
2279</div>
2280
Reid Spencer2ab01932007-02-02 13:57:07 +0000[diff] [blame]2281<!-- _______________________________________________________________________ -->
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2282<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
2283Instruction</a> </div>
2284<div class="doc_text">
2285
2286<h5>Syntax:</h5>
2287<pre> &lt;result&gt; = ashr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2288</pre>
2289
2290<h5>Overview:</h5>
2291<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2292operand shifted to the right a specified number of bits with sign extension.</p>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2293
2294<h5>Arguments:</h5>
2295<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
2296<a href="#t_integer">integer</a> type.</p>
2297
2298<h5>Semantics:</h5>
2299<p>This instruction always performs an arithmetic shift right operation,
2300The most significant bits of the result will be filled with the sign bit
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2301of <tt>var1</tt>. If <tt>var2</tt> is (statically or dynamically) equal to or
2302larger than the number of bits in <tt>var1</tt>, the result is undefined.
2303</p>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2304
2305<h5>Example:</h5>
2306<pre>
2307 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
2308 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
2309 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
2310 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattnerf0e50112007-10-03 21:01:14 +0000[diff] [blame]2311 &lt;result&gt; = ashr i32 1, 32 <i>; undefined</i>
Reid Spencer04e259b2007-01-31 21:39:12 +0000[diff] [blame]2312</pre>
2313</div>
2314
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2315<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2316<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
2317Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2318<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2319<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2320<pre> &lt;result&gt; = and &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2321</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2322<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2323<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
2324its two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2325<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2326<p>The two arguments to the '<tt>and</tt>' instruction must be <a
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2327 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2328identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2329<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2330<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2331<p> </p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2332<div style="align: center">
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2333<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2334 <tbody>
2335 <tr>
2336 <td>In0</td>
2337 <td>In1</td>
2338 <td>Out</td>
2339 </tr>
2340 <tr>
2341 <td>0</td>
2342 <td>0</td>
2343 <td>0</td>
2344 </tr>
2345 <tr>
2346 <td>0</td>
2347 <td>1</td>
2348 <td>0</td>
2349 </tr>
2350 <tr>
2351 <td>1</td>
2352 <td>0</td>
2353 <td>0</td>
2354 </tr>
2355 <tr>
2356 <td>1</td>
2357 <td>1</td>
2358 <td>1</td>
2359 </tr>
2360 </tbody>
2361</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2362</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2363<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2364<pre> &lt;result&gt; = and i32 4, %var <i>; yields {i32}:result = 4 &amp; %var</i>
2365 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
2366 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2367</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2368</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2369<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2370<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2371<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2372<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2373<pre> &lt;result&gt; = or &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2374</pre>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2375<h5>Overview:</h5>
2376<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
2377or of its two operands.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2378<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2379<p>The two arguments to the '<tt>or</tt>' instruction must be <a
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2380 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2381identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2382<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2383<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2384<p> </p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2385<div style="align: center">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2386<table border="1" cellspacing="0" cellpadding="4">
2387 <tbody>
2388 <tr>
2389 <td>In0</td>
2390 <td>In1</td>
2391 <td>Out</td>
2392 </tr>
2393 <tr>
2394 <td>0</td>
2395 <td>0</td>
2396 <td>0</td>
2397 </tr>
2398 <tr>
2399 <td>0</td>
2400 <td>1</td>
2401 <td>1</td>
2402 </tr>
2403 <tr>
2404 <td>1</td>
2405 <td>0</td>
2406 <td>1</td>
2407 </tr>
2408 <tr>
2409 <td>1</td>
2410 <td>1</td>
2411 <td>1</td>
2412 </tr>
2413 </tbody>
2414</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2415</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2416<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2417<pre> &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
2418 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
2419 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2420</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2421</div>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2422<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2423<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
2424Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2425<div class="doc_text">
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2426<h5>Syntax:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2427<pre> &lt;result&gt; = xor &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2428</pre>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2429<h5>Overview:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2430<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
2431or of its two operands. The <tt>xor</tt> is used to implement the
2432"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2433<h5>Arguments:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2434<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2435 href="#t_integer">integer</a> values. Both arguments must have
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2436identical types.</p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2437<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2438<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2439<p> </p>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2440<div style="align: center">
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2441<table border="1" cellspacing="0" cellpadding="4">
2442 <tbody>
2443 <tr>
2444 <td>In0</td>
2445 <td>In1</td>
2446 <td>Out</td>
2447 </tr>
2448 <tr>
2449 <td>0</td>
2450 <td>0</td>
2451 <td>0</td>
2452 </tr>
2453 <tr>
2454 <td>0</td>
2455 <td>1</td>
2456 <td>1</td>
2457 </tr>
2458 <tr>
2459 <td>1</td>
2460 <td>0</td>
2461 <td>1</td>
2462 </tr>
2463 <tr>
2464 <td>1</td>
2465 <td>1</td>
2466 <td>0</td>
2467 </tr>
2468 </tbody>
2469</table>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]2470</div>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2471<p> </p>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2472<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2473<pre> &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
2474 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
2475 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
2476 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2477</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2478</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2479
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2480<!-- ======================================================================= -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2481<div class="doc_subsection">
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2482 <a name="vectorops">Vector Operations</a>
2483</div>
2484
2485<div class="doc_text">
2486
2487<p>LLVM supports several instructions to represent vector operations in a
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2488target-independent manner. These instructions cover the element-access and
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2489vector-specific operations needed to process vectors effectively. While LLVM
2490does directly support these vector operations, many sophisticated algorithms
2491will want to use target-specific intrinsics to take full advantage of a specific
2492target.</p>
2493
2494</div>
2495
2496<!-- _______________________________________________________________________ -->
2497<div class="doc_subsubsection">
2498 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
2499</div>
2500
2501<div class="doc_text">
2502
2503<h5>Syntax:</h5>
2504
2505<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2506 &lt;result&gt; = extractelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, i32 &lt;idx&gt; <i>; yields &lt;ty&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2507</pre>
2508
2509<h5>Overview:</h5>
2510
2511<p>
2512The '<tt>extractelement</tt>' instruction extracts a single scalar
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2513element from a vector at a specified index.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2514</p>
2515
2516
2517<h5>Arguments:</h5>
2518
2519<p>
2520The first operand of an '<tt>extractelement</tt>' instruction is a
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2521value of <a href="#t_vector">vector</a> type. The second operand is
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2522an index indicating the position from which to extract the element.
2523The index may be a variable.</p>
2524
2525<h5>Semantics:</h5>
2526
2527<p>
2528The result is a scalar of the same type as the element type of
2529<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
2530<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
2531results are undefined.
2532</p>
2533
2534<h5>Example:</h5>
2535
2536<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2537 %result = extractelement &lt;4 x i32&gt; %vec, i32 0 <i>; yields i32</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2538</pre>
2539</div>
2540
2541
2542<!-- _______________________________________________________________________ -->
2543<div class="doc_subsubsection">
2544 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
2545</div>
2546
2547<div class="doc_text">
2548
2549<h5>Syntax:</h5>
2550
2551<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2552 &lt;result&gt; = insertelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, &lt;ty&gt; &lt;elt&gt, i32 &lt;idx&gt; <i>; yields &lt;n x &lt;ty&gt;&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2553</pre>
2554
2555<h5>Overview:</h5>
2556
2557<p>
2558The '<tt>insertelement</tt>' instruction inserts a scalar
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2559element into a vector at a specified index.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2560</p>
2561
2562
2563<h5>Arguments:</h5>
2564
2565<p>
2566The first operand of an '<tt>insertelement</tt>' instruction is a
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2567value of <a href="#t_vector">vector</a> type. The second operand is a
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2568scalar value whose type must equal the element type of the first
2569operand. The third operand is an index indicating the position at
2570which to insert the value. The index may be a variable.</p>
2571
2572<h5>Semantics:</h5>
2573
2574<p>
Reid Spencer404a3252007-02-15 03:07:05 +0000[diff] [blame]2575The result is a vector of the same type as <tt>val</tt>. Its
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2576element values are those of <tt>val</tt> except at position
2577<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
2578exceeds the length of <tt>val</tt>, the results are undefined.
2579</p>
2580
2581<h5>Example:</h5>
2582
2583<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2584 %result = insertelement &lt;4 x i32&gt; %vec, i32 1, i32 0 <i>; yields &lt;4 x i32&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2585</pre>
2586</div>
2587
2588<!-- _______________________________________________________________________ -->
2589<div class="doc_subsubsection">
2590 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
2591</div>
2592
2593<div class="doc_text">
2594
2595<h5>Syntax:</h5>
2596
2597<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2598 &lt;result&gt; = shufflevector &lt;n x &lt;ty&gt;&gt; &lt;v1&gt;, &lt;n x &lt;ty&gt;&gt; &lt;v2&gt;, &lt;n x i32&gt; &lt;mask&gt; <i>; yields &lt;n x &lt;ty&gt;&gt;</i>
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2599</pre>
2600
2601<h5>Overview:</h5>
2602
2603<p>
2604The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
2605from two input vectors, returning a vector of the same type.
2606</p>
2607
2608<h5>Arguments:</h5>
2609
2610<p>
2611The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
2612with types that match each other and types that match the result of the
2613instruction. The third argument is a shuffle mask, which has the same number
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2614of elements as the other vector type, but whose element type is always 'i32'.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2615</p>
2616
2617<p>
2618The shuffle mask operand is required to be a constant vector with either
2619constant integer or undef values.
2620</p>
2621
2622<h5>Semantics:</h5>
2623
2624<p>
2625The elements of the two input vectors are numbered from left to right across
2626both of the vectors. The shuffle mask operand specifies, for each element of
2627the result vector, which element of the two input registers the result element
2628gets. The element selector may be undef (meaning "don't care") and the second
2629operand may be undef if performing a shuffle from only one vector.
2630</p>
2631
2632<h5>Example:</h5>
2633
2634<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2635 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2636 &lt;4 x i32&gt; &lt;i32 0, i32 4, i32 1, i32 5&gt; <i>; yields &lt;4 x i32&gt;</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2637 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; undef,
2638 &lt;4 x i32&gt; &lt;i32 0, i32 1, i32 2, i32 3&gt; <i>; yields &lt;4 x i32&gt;</i> - Identity shuffle.
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2639</pre>
2640</div>
2641
Tanya Lattnerb138bbe2006-04-14 19:24:33 +0000[diff] [blame]2642
Chris Lattnerce83bff2006-04-08 23:07:04 +0000[diff] [blame]2643<!-- ======================================================================= -->
2644<div class="doc_subsection">
Chris Lattner6ab66722006-08-15 00:45:58 +0000[diff] [blame]2645 <a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2646</div>
2647
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2648<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2649
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2650<p>A key design point of an SSA-based representation is how it
2651represents memory. In LLVM, no memory locations are in SSA form, which
2652makes things very simple. This section describes how to read, write,
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]2653allocate, and free memory in LLVM.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2654
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2655</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2656
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2657<!-- _______________________________________________________________________ -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2658<div class="doc_subsubsection">
2659 <a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
2660</div>
2661
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2662<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2663
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2664<h5>Syntax:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2665
2666<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2667 &lt;result&gt; = malloc &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2668</pre>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2669
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2670<h5>Overview:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2671
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2672<p>The '<tt>malloc</tt>' instruction allocates memory from the system
2673heap and returns a pointer to it.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2674
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2675<h5>Arguments:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2676
2677<p>The '<tt>malloc</tt>' instruction allocates
2678<tt>sizeof(&lt;type&gt;)*NumElements</tt>
John Criswella92e5862004-02-24 16:13:56 +0000[diff] [blame]2679bytes of memory from the operating system and returns a pointer of the
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2680appropriate type to the program. If "NumElements" is specified, it is the
2681number of elements allocated. If an alignment is specified, the value result
2682of the allocation is guaranteed to be aligned to at least that boundary. If
2683not specified, or if zero, the target can choose to align the allocation on any
2684convenient boundary.</p>
2685
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2686<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2687
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2688<h5>Semantics:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2689
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2690<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
2691a pointer is returned.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2692
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2693<h5>Example:</h5>
2694
2695<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2696 %array = malloc [4 x i8 ] <i>; yields {[%4 x i8]*}:array</i>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2697
Bill Wendling2d8b9a82007-05-29 09:42:13 +0000[diff] [blame]2698 %size = <a href="#i_add">add</a> i32 2, 2 <i>; yields {i32}:size = i32 4</i>
2699 %array1 = malloc i8, i32 4 <i>; yields {i8*}:array1</i>
2700 %array2 = malloc [12 x i8], i32 %size <i>; yields {[12 x i8]*}:array2</i>
2701 %array3 = malloc i32, i32 4, align 1024 <i>; yields {i32*}:array3</i>
2702 %array4 = malloc i32, align 1024 <i>; yields {i32*}:array4</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2703</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2704</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2705
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2706<!-- _______________________________________________________________________ -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2707<div class="doc_subsubsection">
2708 <a name="i_free">'<tt>free</tt>' Instruction</a>
2709</div>
2710
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2711<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2712
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2713<h5>Syntax:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2714
2715<pre>
2716 free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2717</pre>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2718
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2719<h5>Overview:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2720
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2721<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2722memory heap to be reallocated in the future.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2723
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2724<h5>Arguments:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2725
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2726<p>'<tt>value</tt>' shall be a pointer value that points to a value
2727that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
2728instruction.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2729
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2730<h5>Semantics:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2731
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]2732<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2733after this instruction executes.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2734
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2735<h5>Example:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2736
2737<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2738 %array = <a href="#i_malloc">malloc</a> [4 x i8] <i>; yields {[4 x i8]*}:array</i>
2739 free [4 x i8]* %array
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2740</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2741</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2742
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2743<!-- _______________________________________________________________________ -->
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2744<div class="doc_subsubsection">
2745 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
2746</div>
2747
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2748<div class="doc_text">
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2749
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2750<h5>Syntax:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2751
2752<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2753 &lt;result&gt; = alloca &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2754</pre>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2755
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2756<h5>Overview:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2757
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2758<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
2759currently executing function, to be automatically released when this function
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2760returns to its caller.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2761
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2762<h5>Arguments:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2763
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]2764<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2765bytes of memory on the runtime stack, returning a pointer of the
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2766appropriate type to the program. If "NumElements" is specified, it is the
2767number of elements allocated. If an alignment is specified, the value result
2768of the allocation is guaranteed to be aligned to at least that boundary. If
2769not specified, or if zero, the target can choose to align the allocation on any
2770convenient boundary.</p>
2771
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2772<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2773
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2774<h5>Semantics:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2775
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2776<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2777memory is automatically released when the function returns. The '<tt>alloca</tt>'
2778instruction is commonly used to represent automatic variables that must
2779have an address available. When the function returns (either with the <tt><a
John Criswellc932bef2005-05-12 16:55:34 +0000[diff] [blame]2780 href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2781instructions), the memory is reclaimed.</p>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2782
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2783<h5>Example:</h5>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2784
2785<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2786 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2787 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
2788 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2789 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2790</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2791</div>
Chris Lattner54611b42005-11-06 08:02:57 +0000[diff] [blame]2792
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2793<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2794<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
2795Instruction</a> </div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2796<div class="doc_text">
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2797<h5>Syntax:</h5>
Christopher Lambbff50202007-04-21 08:16:25 +0000[diff] [blame]2798<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 Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2799<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2800<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2801<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2802<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]2803address from which to load. The pointer must point to a <a
Chris Lattner10ee9652004-06-03 22:57:15 +0000[diff] [blame]2804 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell4c0cf7f2005-10-24 16:17:18 +0000[diff] [blame]2805marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2806the number or order of execution of this <tt>load</tt> with other
2807volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
2808instructions. </p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2809<h5>Semantics:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2810<p>The location of memory pointed to is loaded.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2811<h5>Examples:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2812<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2813 <a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2814 href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
2815 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2816</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2817</div>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2818<!-- _______________________________________________________________________ -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2819<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
2820Instruction</a> </div>
Reid Spencera89fb182006-11-09 21:18:01 +0000[diff] [blame]2821<div class="doc_text">
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2822<h5>Syntax:</h5>
Christopher Lambbff50202007-04-21 08:16:25 +0000[diff] [blame]2823<pre> store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
2824 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2825</pre>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2826<h5>Overview:</h5>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2827<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2828<h5>Arguments:</h5>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2829<p>There are two arguments to the '<tt>store</tt>' instruction: a value
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2830to store and an address at which to store it. The type of the '<tt>&lt;pointer&gt;</tt>'
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2831operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2832operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2833optimizer is not allowed to modify the number or order of execution of
2834this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
2835 href="#i_store">store</a></tt> instructions.</p>
2836<h5>Semantics:</h5>
2837<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
2838at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2839<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2840<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling8830ffe2007-10-22 05:10:05 +0000[diff] [blame]2841 store i32 3, i32* %ptr <i>; yields {void}</i>
2842 %val = <a href="#i_load">load</a> i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2843</pre>
Reid Spencer443460a2006-11-09 21:15:49 +0000[diff] [blame]2844</div>
2845
Chris Lattner095735d2002-05-06 03:03:22 +0000[diff] [blame]2846<!-- _______________________________________________________________________ -->
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2847<div class="doc_subsubsection">
2848 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
2849</div>
2850
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2851<div class="doc_text">
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2852<h5>Syntax:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2853<pre>
2854 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
2855</pre>
2856
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2857<h5>Overview:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2858
2859<p>
2860The '<tt>getelementptr</tt>' instruction is used to get the address of a
2861subelement of an aggregate data structure.</p>
2862
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2863<h5>Arguments:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2864
Reid Spencercee005c2006-12-04 21:29:24 +0000[diff] [blame]2865<p>This instruction takes a list of integer operands that indicate what
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2866elements of the aggregate object to index to. The actual types of the arguments
2867provided depend on the type of the first pointer argument. The
2868'<tt>getelementptr</tt>' instruction is used to index down through the type
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]2869levels of a structure or to a specific index in an array. When indexing into a
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2870structure, only <tt>i32</tt> integer constants are allowed. When indexing
Reid Spencercee005c2006-12-04 21:29:24 +0000[diff] [blame]2871into an array or pointer, only integers of 32 or 64 bits are allowed, and will
2872be sign extended to 64-bit values.</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2873
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2874<p>For example, let's consider a C code fragment and how it gets
2875compiled to LLVM:</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2876
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2877<div class="doc_code">
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2878<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2879struct RT {
2880 char A;
Chris Lattnera446f1b2007-05-29 15:43:56 +0000[diff] [blame]2881 int B[10][20];
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2882 char C;
2883};
2884struct ST {
Chris Lattnera446f1b2007-05-29 15:43:56 +0000[diff] [blame]2885 int X;
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2886 double Y;
2887 struct RT Z;
2888};
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2889
Chris Lattnera446f1b2007-05-29 15:43:56 +0000[diff] [blame]2890int *foo(struct ST *s) {
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2891 return &amp;s[1].Z.B[5][13];
2892}
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2893</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2894</div>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2895
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2896<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2897
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2898<div class="doc_code">
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2899<pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2900%RT = type { i8 , [10 x [20 x i32]], i8 }
2901%ST = type { i32, double, %RT }
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2902
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2903define i32* %foo(%ST* %s) {
2904entry:
2905 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
2906 ret i32* %reg
2907}
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2908</pre>
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]2909</div>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2910
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2911<h5>Semantics:</h5>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2912
2913<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
John Criswell4a3327e2005-05-13 22:25:59 +0000[diff] [blame]2914on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
Reid Spencercee005c2006-12-04 21:29:24 +0000[diff] [blame]2915and <a href="#t_array">array</a> types can use a 32-bit or 64-bit
Reid Spencerc0312692006-12-03 16:53:48 +0000[diff] [blame]2916<a href="#t_integer">integer</a> type but the value will always be sign extended
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2917to 64-bits. <a href="#t_struct">Structure</a> types require <tt>i32</tt>
Reid Spencerc0312692006-12-03 16:53:48 +0000[diff] [blame]2918<b>constants</b>.</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2919
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2920<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2921type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2922}</tt>' type, a structure. The second index indexes into the third element of
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2923the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
2924i8 }</tt>' type, another structure. The third index indexes into the second
2925element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2926array. The two dimensions of the array are subscripted into, yielding an
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2927'<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
2928to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2929
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]2930<p>Note that it is perfectly legal to index partially through a
2931structure, returning a pointer to an inner element. Because of this,
2932the LLVM code for the given testcase is equivalent to:</p>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2933
2934<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2935 define i32* %foo(%ST* %s) {
2936 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
Jeff Cohen5819f182007-04-22 01:17:39 +0000[diff] [blame]2937 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
2938 %t3 = getelementptr %RT* %t2, i32 0, i32 1 <i>; yields [10 x [20 x i32]]*:%t3</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2939 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
2940 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
2941 ret i32* %t5
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2942 }
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]2943</pre>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]2944
2945<p>Note that it is undefined to access an array out of bounds: array and
2946pointer indexes must always be within the defined bounds of the array type.
2947The one exception for this rules is zero length arrays. These arrays are
2948defined to be accessible as variable length arrays, which requires access
2949beyond the zero'th element.</p>
2950
Chris Lattner6ab66722006-08-15 00:45:58 +0000[diff] [blame]2951<p>The getelementptr instruction is often confusing. For some more insight
2952into how it works, see <a href="GetElementPtr.html">the getelementptr
2953FAQ</a>.</p>
2954
Chris Lattner590645f2002-04-14 06:13:44 +0000[diff] [blame]2955<h5>Example:</h5>
Chris Lattnerc0ad71e2005-06-24 17:22:57 +0000[diff] [blame]2956
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2957<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]2958 <i>; yields [12 x i8]*:aptr</i>
2959 %aptr = getelementptr {i32, [12 x i8]}* %sptr, i64 0, i32 1
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2960</pre>
Chris Lattner33fd7022004-04-05 01:30:49 +0000[diff] [blame]2961</div>
Reid Spencer443460a2006-11-09 21:15:49 +0000[diff] [blame]2962
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]2963<!-- ======================================================================= -->
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]2964<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2965</div>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2966<div class="doc_text">
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]2967<p>The instructions in this category are the conversion instructions (casting)
2968which all take a single operand and a type. They perform various bit conversions
2969on the operand.</p>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]2970</div>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]2971
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]2972<!-- _______________________________________________________________________ -->
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]2973<div class="doc_subsubsection">
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2974 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
2975</div>
2976<div class="doc_text">
2977
2978<h5>Syntax:</h5>
2979<pre>
2980 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2981</pre>
2982
2983<h5>Overview:</h5>
2984<p>
2985The '<tt>trunc</tt>' instruction truncates its operand to the type <tt>ty2</tt>.
2986</p>
2987
2988<h5>Arguments:</h5>
2989<p>
2990The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
2991be an <a href="#t_integer">integer</a> type, and a type that specifies the size
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]2992and type of the result, which must be an <a href="#t_integer">integer</a>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2993type. The bit size of <tt>value</tt> must be larger than the bit size of
2994<tt>ty2</tt>. Equal sized types are not allowed.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]2995
2996<h5>Semantics:</h5>
2997<p>
2998The '<tt>trunc</tt>' instruction truncates the high order bits in <tt>value</tt>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]2999and converts the remaining bits to <tt>ty2</tt>. Since the source size must be
3000larger than the destination size, <tt>trunc</tt> cannot be a <i>no-op cast</i>.
3001It will always truncate bits.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3002
3003<h5>Example:</h5>
3004<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3005 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3006 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
3007 %Y = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3008</pre>
3009</div>
3010
3011<!-- _______________________________________________________________________ -->
3012<div class="doc_subsubsection">
3013 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
3014</div>
3015<div class="doc_text">
3016
3017<h5>Syntax:</h5>
3018<pre>
3019 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3020</pre>
3021
3022<h5>Overview:</h5>
3023<p>The '<tt>zext</tt>' instruction zero extends its operand to type
3024<tt>ty2</tt>.</p>
3025
3026
3027<h5>Arguments:</h5>
3028<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3029<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3030also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3031<tt>value</tt> must be smaller than the bit size of the destination type,
3032<tt>ty2</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3033
3034<h5>Semantics:</h5>
3035<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
Chris Lattnerc87f3df2007-05-24 19:13:27 +0000[diff] [blame]3036bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3037
Reid Spencer07c9c682007-01-12 15:46:11 +0000[diff] [blame]3038<p>When zero extending from i1, the result will always be either 0 or 1.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3039
3040<h5>Example:</h5>
3041<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3042 %X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3043 %Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3044</pre>
3045</div>
3046
3047<!-- _______________________________________________________________________ -->
3048<div class="doc_subsubsection">
3049 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
3050</div>
3051<div class="doc_text">
3052
3053<h5>Syntax:</h5>
3054<pre>
3055 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3056</pre>
3057
3058<h5>Overview:</h5>
3059<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
3060
3061<h5>Arguments:</h5>
3062<p>
3063The '<tt>sext</tt>' instruction takes a value to cast, which must be of
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3064<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3065also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3066<tt>value</tt> must be smaller than the bit size of the destination type,
3067<tt>ty2</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3068
3069<h5>Semantics:</h5>
3070<p>
3071The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
3072bit (highest order bit) of the <tt>value</tt> until it reaches the bit size of
Chris Lattnerc87f3df2007-05-24 19:13:27 +0000[diff] [blame]3073the type <tt>ty2</tt>.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3074
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3075<p>When sign extending from i1, the extension always results in -1 or 0.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3076
3077<h5>Example:</h5>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3078<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3079 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3080 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3081</pre>
3082</div>
3083
3084<!-- _______________________________________________________________________ -->
3085<div class="doc_subsubsection">
Reid Spencer2e2740d2006-11-09 21:48:10 +0000[diff] [blame]3086 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
3087</div>
3088
3089<div class="doc_text">
3090
3091<h5>Syntax:</h5>
3092
3093<pre>
3094 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3095</pre>
3096
3097<h5>Overview:</h5>
3098<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
3099<tt>ty2</tt>.</p>
3100
3101
3102<h5>Arguments:</h5>
3103<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
3104 point</a> value to cast and a <a href="#t_floating">floating point</a> type to
3105cast it to. The size of <tt>value</tt> must be larger than the size of
3106<tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
3107<i>no-op cast</i>.</p>
3108
3109<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3110<p> The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
3111<a href="#t_floating">floating point</a> type to a smaller
3112<a href="#t_floating">floating point</a> type. If the value cannot fit within
3113the destination type, <tt>ty2</tt>, then the results are undefined.</p>
Reid Spencer2e2740d2006-11-09 21:48:10 +0000[diff] [blame]3114
3115<h5>Example:</h5>
3116<pre>
3117 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
3118 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
3119</pre>
3120</div>
3121
3122<!-- _______________________________________________________________________ -->
3123<div class="doc_subsubsection">
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3124 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
3125</div>
3126<div class="doc_text">
3127
3128<h5>Syntax:</h5>
3129<pre>
3130 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3131</pre>
3132
3133<h5>Overview:</h5>
3134<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
3135floating point value.</p>
3136
3137<h5>Arguments:</h5>
3138<p>The '<tt>fpext</tt>' instruction takes a
3139<a href="#t_floating">floating point</a> <tt>value</tt> to cast,
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3140and a <a href="#t_floating">floating point</a> type to cast it to. The source
3141type must be smaller than the destination type.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3142
3143<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3144<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3145<a href="#t_floating">floating point</a> type to a larger
3146<a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3147used to make a <i>no-op cast</i> because it always changes bits. Use
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3148<tt>bitcast</tt> to make a <i>no-op cast</i> for a floating point cast.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3149
3150<h5>Example:</h5>
3151<pre>
3152 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
3153 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
3154</pre>
3155</div>
3156
3157<!-- _______________________________________________________________________ -->
3158<div class="doc_subsubsection">
Reid Spencer2eadb532007-01-21 00:29:26 +0000[diff] [blame]3159 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3160</div>
3161<div class="doc_text">
3162
3163<h5>Syntax:</h5>
3164<pre>
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]3165 &lt;result&gt; = fptoui &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3166</pre>
3167
3168<h5>Overview:</h5>
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]3169<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3170unsigned integer equivalent of type <tt>ty2</tt>.
3171</p>
3172
3173<h5>Arguments:</h5>
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]3174<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]3175scalar or vector <a href="#t_floating">floating point</a> value, and a type
3176to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3177type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3178vector integer type with the same number of elements as <tt>ty</tt></p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3179
3180<h5>Semantics:</h5>
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]3181<p> The '<tt>fptoui</tt>' instruction converts its
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3182<a href="#t_floating">floating point</a> operand into the nearest (rounding
3183towards zero) unsigned integer value. If the value cannot fit in <tt>ty2</tt>,
3184the results are undefined.</p>
3185
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3186<h5>Example:</h5>
3187<pre>
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]3188 %X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner5b95a172007-09-22 03:17:52 +0000[diff] [blame]3189 %Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Reid Spencer753163d2007-07-31 14:40:14 +0000[diff] [blame]3190 %X = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3191</pre>
3192</div>
3193
3194<!-- _______________________________________________________________________ -->
3195<div class="doc_subsubsection">
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3196 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3197</div>
3198<div class="doc_text">
3199
3200<h5>Syntax:</h5>
3201<pre>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3202 &lt;result&gt; = fptosi &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3203</pre>
3204
3205<h5>Overview:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3206<p>The '<tt>fptosi</tt>' instruction converts
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3207<a href="#t_floating">floating point</a> <tt>value</tt> to type <tt>ty2</tt>.
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3208</p>
3209
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3210<h5>Arguments:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3211<p> The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]3212scalar or vector <a href="#t_floating">floating point</a> value, and a type
3213to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3214type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3215vector integer type with the same number of elements as <tt>ty</tt></p>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3216
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3217<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3218<p>The '<tt>fptosi</tt>' instruction converts its
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3219<a href="#t_floating">floating point</a> operand into the nearest (rounding
3220towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
3221the results are undefined.</p>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3222
Chris Lattner70de6632001-07-09 00:26:23 +0000[diff] [blame]3223<h5>Example:</h5>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3224<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3225 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner5b95a172007-09-22 03:17:52 +0000[diff] [blame]3226 %Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3227 %X = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3228</pre>
3229</div>
3230
3231<!-- _______________________________________________________________________ -->
3232<div class="doc_subsubsection">
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3233 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3234</div>
3235<div class="doc_text">
3236
3237<h5>Syntax:</h5>
3238<pre>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3239 &lt;result&gt; = uitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3240</pre>
3241
3242<h5>Overview:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3243<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3244integer and converts that value to the <tt>ty2</tt> type.</p>
3245
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3246<h5>Arguments:</h5>
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]3247<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be a
3248scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3249to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3250type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3251floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3252
3253<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3254<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3255integer quantity and converts it to the corresponding floating point value. If
Jeff Cohenbeccb742007-04-22 14:56:37 +0000[diff] [blame]3256the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3257
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3258<h5>Example:</h5>
3259<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3260 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3261 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3262</pre>
3263</div>
3264
3265<!-- _______________________________________________________________________ -->
3266<div class="doc_subsubsection">
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3267 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3268</div>
3269<div class="doc_text">
3270
3271<h5>Syntax:</h5>
3272<pre>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3273 &lt;result&gt; = sitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3274</pre>
3275
3276<h5>Overview:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3277<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3278integer and converts that value to the <tt>ty2</tt> type.</p>
3279
3280<h5>Arguments:</h5>
Nate Begemand4d45c22007-11-17 03:58:34 +0000[diff] [blame]3281<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be a
3282scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3283to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3284type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3285floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3286
3287<h5>Semantics:</h5>
Reid Spencer51b07252006-11-09 23:03:26 +0000[diff] [blame]3288<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3289integer quantity and converts it to the corresponding floating point value. If
Jeff Cohenbeccb742007-04-22 14:56:37 +0000[diff] [blame]3290the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3291
3292<h5>Example:</h5>
3293<pre>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3294 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3295 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3296</pre>
3297</div>
3298
3299<!-- _______________________________________________________________________ -->
3300<div class="doc_subsubsection">
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3301 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
3302</div>
3303<div class="doc_text">
3304
3305<h5>Syntax:</h5>
3306<pre>
3307 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3308</pre>
3309
3310<h5>Overview:</h5>
3311<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
3312the integer type <tt>ty2</tt>.</p>
3313
3314<h5>Arguments:</h5>
3315<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3316must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3317<tt>ty2</tt>, which must be an <a href="#t_integer">integer</a> type.
3318
3319<h5>Semantics:</h5>
3320<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
3321<tt>ty2</tt> by interpreting the pointer value as an integer and either
3322truncating or zero extending that value to the size of the integer type. If
3323<tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
3324<tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3325are the same size, then nothing is done (<i>no-op cast</i>) other than a type
3326change.</p>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3327
3328<h5>Example:</h5>
3329<pre>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3330 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
3331 %Y = ptrtoint i32* %x to i64 <i>; yields zero extension on 32-bit architecture</i>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3332</pre>
3333</div>
3334
3335<!-- _______________________________________________________________________ -->
3336<div class="doc_subsubsection">
3337 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
3338</div>
3339<div class="doc_text">
3340
3341<h5>Syntax:</h5>
3342<pre>
3343 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3344</pre>
3345
3346<h5>Overview:</h5>
3347<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to
3348a pointer type, <tt>ty2</tt>.</p>
3349
3350<h5>Arguments:</h5>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3351<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3352value to cast, and a type to cast it to, which must be a
Anton Korobeynikova0554d92007-01-12 19:20:47 +0000[diff] [blame]3353<a href="#t_pointer">pointer</a> type.
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3354
3355<h5>Semantics:</h5>
3356<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
3357<tt>ty2</tt> by applying either a zero extension or a truncation depending on
3358the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
3359size of a pointer then a truncation is done. If <tt>value</tt> is smaller than
3360the size of a pointer then a zero extension is done. If they are the same size,
3361nothing is done (<i>no-op cast</i>).</p>
3362
3363<h5>Example:</h5>
3364<pre>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3365 %X = inttoptr i32 255 to i32* <i>; yields zero extension on 64-bit architecture</i>
3366 %X = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
3367 %Y = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3368</pre>
3369</div>
3370
3371<!-- _______________________________________________________________________ -->
3372<div class="doc_subsubsection">
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3373 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3374</div>
3375<div class="doc_text">
3376
3377<h5>Syntax:</h5>
3378<pre>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3379 &lt;result&gt; = bitcast &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3380</pre>
3381
3382<h5>Overview:</h5>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3383<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3384<tt>ty2</tt> without changing any bits.</p>
3385
3386<h5>Arguments:</h5>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3387<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3388a first class value, and a type to cast it to, which must also be a <a
3389 href="#t_firstclass">first class</a> type. The bit sizes of <tt>value</tt>
Reid Spencere3db84c2007-01-09 20:08:58 +0000[diff] [blame]3390and the destination type, <tt>ty2</tt>, must be identical. If the source
3391type is a pointer, the destination type must also be a pointer.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3392
3393<h5>Semantics:</h5>
Reid Spencer5b950642006-11-11 23:08:07 +0000[diff] [blame]3394<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencerb7344ff2006-11-11 21:00:47 +0000[diff] [blame]3395<tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
3396this conversion. The conversion is done as if the <tt>value</tt> had been
3397stored to memory and read back as type <tt>ty2</tt>. Pointer types may only be
3398converted to other pointer types with this instruction. To convert pointers to
3399other types, use the <a href="#i_inttoptr">inttoptr</a> or
3400<a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer59b6b7d2006-11-08 01:11:31 +0000[diff] [blame]3401
3402<h5>Example:</h5>
3403<pre>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3404 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3405 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
3406 %Z = bitcast <2xint> %V to i64; <i>; yields i64: %V</i>
Chris Lattner70de6632001-07-09 00:26:23 +0000[diff] [blame]3407</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3408</div>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3409
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3410<!-- ======================================================================= -->
3411<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
3412<div class="doc_text">
3413<p>The instructions in this category are the "miscellaneous"
3414instructions, which defy better classification.</p>
3415</div>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3416
3417<!-- _______________________________________________________________________ -->
3418<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
3419</div>
3420<div class="doc_text">
3421<h5>Syntax:</h5>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3422<pre> &lt;result&gt; = icmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {i1}:result</i>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3423</pre>
3424<h5>Overview:</h5>
3425<p>The '<tt>icmp</tt>' instruction returns a boolean value based on comparison
3426of its two integer operands.</p>
3427<h5>Arguments:</h5>
3428<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3429the condition code indicating the kind of comparison to perform. It is not
3430a value, just a keyword. The possible condition code are:
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3431<ol>
3432 <li><tt>eq</tt>: equal</li>
3433 <li><tt>ne</tt>: not equal </li>
3434 <li><tt>ugt</tt>: unsigned greater than</li>
3435 <li><tt>uge</tt>: unsigned greater or equal</li>
3436 <li><tt>ult</tt>: unsigned less than</li>
3437 <li><tt>ule</tt>: unsigned less or equal</li>
3438 <li><tt>sgt</tt>: signed greater than</li>
3439 <li><tt>sge</tt>: signed greater or equal</li>
3440 <li><tt>slt</tt>: signed less than</li>
3441 <li><tt>sle</tt>: signed less or equal</li>
3442</ol>
Chris Lattnerc0f423a2007-01-15 01:54:13 +0000[diff] [blame]3443<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Reid Spencer784ef792007-01-04 05:19:58 +0000[diff] [blame]3444<a href="#t_pointer">pointer</a> typed. They must also be identical types.</p>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3445<h5>Semantics:</h5>
3446<p>The '<tt>icmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3447the condition code given as <tt>cond</tt>. The comparison performed always
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3448yields a <a href="#t_primitive">i1</a> result, as follows:
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3449<ol>
3450 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
3451 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3452 </li>
3453 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
3454 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3455 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
3456 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3457 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
3458 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3459 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
3460 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3461 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
3462 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3463 <li><tt>sgt</tt>: interprets the operands as signed values and yields
3464 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3465 <li><tt>sge</tt>: interprets the operands as signed values and yields
3466 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3467 <li><tt>slt</tt>: interprets the operands as signed values and yields
3468 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3469 <li><tt>sle</tt>: interprets the operands as signed values and yields
3470 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3471</ol>
3472<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3473values are compared as if they were integers.</p>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3474
3475<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3476<pre> &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
3477 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
3478 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
3479 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
3480 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
3481 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3482</pre>
3483</div>
3484
3485<!-- _______________________________________________________________________ -->
3486<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
3487</div>
3488<div class="doc_text">
3489<h5>Syntax:</h5>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3490<pre> &lt;result&gt; = fcmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {i1}:result</i>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3491</pre>
3492<h5>Overview:</h5>
3493<p>The '<tt>fcmp</tt>' instruction returns a boolean value based on comparison
3494of its floating point operands.</p>
3495<h5>Arguments:</h5>
3496<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3497the condition code indicating the kind of comparison to perform. It is not
3498a value, just a keyword. The possible condition code are:
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3499<ol>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3500 <li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3501 <li><tt>oeq</tt>: ordered and equal</li>
3502 <li><tt>ogt</tt>: ordered and greater than </li>
3503 <li><tt>oge</tt>: ordered and greater than or equal</li>
3504 <li><tt>olt</tt>: ordered and less than </li>
3505 <li><tt>ole</tt>: ordered and less than or equal</li>
3506 <li><tt>one</tt>: ordered and not equal</li>
3507 <li><tt>ord</tt>: ordered (no nans)</li>
3508 <li><tt>ueq</tt>: unordered or equal</li>
3509 <li><tt>ugt</tt>: unordered or greater than </li>
3510 <li><tt>uge</tt>: unordered or greater than or equal</li>
3511 <li><tt>ult</tt>: unordered or less than </li>
3512 <li><tt>ule</tt>: unordered or less than or equal</li>
3513 <li><tt>une</tt>: unordered or not equal</li>
3514 <li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3515 <li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3516</ol>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3517<p><i>Ordered</i> means that neither operand is a QNAN while
Reid Spencer02e0d1d2006-12-06 07:08:07 +0000[diff] [blame]3518<i>unordered</i> means that either operand may be a QNAN.</p>
Reid Spencer784ef792007-01-04 05:19:58 +0000[diff] [blame]3519<p>The <tt>val1</tt> and <tt>val2</tt> arguments must be
3520<a href="#t_floating">floating point</a> typed. They must have identical
3521types.</p>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3522<h5>Semantics:</h5>
3523<p>The '<tt>fcmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3524the condition code given as <tt>cond</tt>. The comparison performed always
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3525yields a <a href="#t_primitive">i1</a> result, as follows:
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3526<ol>
3527 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3528 <li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3529 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3530 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3531 <tt>var1</tt> is greather than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3532 <li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3533 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3534 <li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3535 <tt>var1</tt> is less than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3536 <li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3537 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3538 <li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3539 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3540 <li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
3541 <li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3542 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3543 <li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3544 <tt>var1</tt> is greater than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3545 <li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3546 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3547 <li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3548 <tt>var1</tt> is less than <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3549 <li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3550 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3551 <li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3552 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
Reid Spencerf69acf32006-11-19 03:00:14 +0000[diff] [blame]3553 <li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3554 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
3555</ol>
Reid Spencerc828a0e2006-11-18 21:50:54 +0000[diff] [blame]3556
3557<h5>Example:</h5>
3558<pre> &lt;result&gt; = fcmp oeq float 4.0, 5.0 <i>; yields: result=false</i>
3559 &lt;result&gt; = icmp one float 4.0, 5.0 <i>; yields: result=true</i>
3560 &lt;result&gt; = icmp olt float 4.0, 5.0 <i>; yields: result=true</i>
3561 &lt;result&gt; = icmp ueq double 1.0, 2.0 <i>; yields: result=false</i>
3562</pre>
3563</div>
3564
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3565<!-- _______________________________________________________________________ -->
3566<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
3567Instruction</a> </div>
3568<div class="doc_text">
3569<h5>Syntax:</h5>
3570<pre> &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...<br></pre>
3571<h5>Overview:</h5>
3572<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
3573the SSA graph representing the function.</p>
3574<h5>Arguments:</h5>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3575<p>The type of the incoming values is specified with the first type
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3576field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
3577as arguments, with one pair for each predecessor basic block of the
3578current block. Only values of <a href="#t_firstclass">first class</a>
3579type may be used as the value arguments to the PHI node. Only labels
3580may be used as the label arguments.</p>
3581<p>There must be no non-phi instructions between the start of a basic
3582block and the PHI instructions: i.e. PHI instructions must be first in
3583a basic block.</p>
3584<h5>Semantics:</h5>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3585<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
3586specified by the pair corresponding to the predecessor basic block that executed
3587just prior to the current block.</p>
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3588<h5>Example:</h5>
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3589<pre>Loop: ; Infinite loop that counts from 0 on up...<br> %indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]<br> %nextindvar = add i32 %indvar, 1<br> br label %Loop<br></pre>
Reid Spencer97c5fa42006-11-08 01:18:52 +0000[diff] [blame]3590</div>
3591
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3592<!-- _______________________________________________________________________ -->
3593<div class="doc_subsubsection">
3594 <a name="i_select">'<tt>select</tt>' Instruction</a>
3595</div>
3596
3597<div class="doc_text">
3598
3599<h5>Syntax:</h5>
3600
3601<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3602 &lt;result&gt; = select i1 &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3603</pre>
3604
3605<h5>Overview:</h5>
3606
3607<p>
3608The '<tt>select</tt>' instruction is used to choose one value based on a
3609condition, without branching.
3610</p>
3611
3612
3613<h5>Arguments:</h5>
3614
3615<p>
3616The '<tt>select</tt>' instruction requires a boolean value indicating the condition, and two values of the same <a href="#t_firstclass">first class</a> type.
3617</p>
3618
3619<h5>Semantics:</h5>
3620
3621<p>
3622If the boolean condition evaluates to true, the instruction returns the first
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3623value argument; otherwise, it returns the second value argument.
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3624</p>
3625
3626<h5>Example:</h5>
3627
3628<pre>
Reid Spencer36a15422007-01-12 03:35:51 +0000[diff] [blame]3629 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattnerb53c28d2004-03-12 05:50:16 +0000[diff] [blame]3630</pre>
3631</div>
3632
Robert Bocchinof72fdfe2006-01-15 20:48:27 +0000[diff] [blame]3633
3634<!-- _______________________________________________________________________ -->
3635<div class="doc_subsubsection">
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3636 <a name="i_call">'<tt>call</tt>' Instruction</a>
3637</div>
3638
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3639<div class="doc_text">
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3640
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3641<h5>Syntax:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3642<pre>
Nick Lewyckya9b13d52007-09-08 13:57:50 +0000[diff] [blame]3643 &lt;result&gt; = [tail] call [<a href="#callingconv">cconv</a>] &lt;ty&gt; [&lt;fnty&gt;*] &lt;fnptrval&gt;(&lt;param list&gt;)
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3644</pre>
3645
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3646<h5>Overview:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3647
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3648<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3649
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3650<h5>Arguments:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3651
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3652<p>This instruction requires several arguments:</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3653
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3654<ol>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3655 <li>
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]3656 <p>The optional "tail" marker indicates whether the callee function accesses
3657 any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3658 function call is eligible for tail call optimization. Note that calls may
3659 be marked "tail" even if they do not occur before a <a
3660 href="#i_ret"><tt>ret</tt></a> instruction.
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3661 </li>
3662 <li>
Duncan Sands16f122e2007-03-30 12:22:09 +0000[diff] [blame]3663 <p>The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattner0132aff2005-05-06 22:57:40 +0000[diff] [blame]3664 convention</a> the call should use. If none is specified, the call defaults
3665 to using C calling conventions.
3666 </li>
3667 <li>
Nick Lewyckya9b13d52007-09-08 13:57:50 +0000[diff] [blame]3668 <p>'<tt>ty</tt>': the type of the call instruction itself which is also
3669 the type of the return value. Functions that return no value are marked
3670 <tt><a href="#t_void">void</a></tt>.</p>
3671 </li>
3672 <li>
3673 <p>'<tt>fnty</tt>': shall be the signature of the pointer to function
3674 value being invoked. The argument types must match the types implied by
3675 this signature. This type can be omitted if the function is not varargs
3676 and if the function type does not return a pointer to a function.</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3677 </li>
3678 <li>
3679 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
3680 be invoked. In most cases, this is a direct function invocation, but
3681 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3682 to function value.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3683 </li>
3684 <li>
3685 <p>'<tt>function args</tt>': argument list whose types match the
Reid Spencerd845d162005-05-01 22:22:57 +0000[diff] [blame]3686 function signature argument types. All arguments must be of
3687 <a href="#t_firstclass">first class</a> type. If the function signature
3688 indicates the function accepts a variable number of arguments, the extra
3689 arguments can be specified.</p>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3690 </li>
Chris Lattnera8292f32002-05-06 22:08:29 +0000[diff] [blame]3691</ol>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3692
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3693<h5>Semantics:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3694
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3695<p>The '<tt>call</tt>' instruction is used to cause control flow to
3696transfer to a specified function, with its incoming arguments bound to
3697the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
3698instruction in the called function, control flow continues with the
3699instruction after the function call, and the return value of the
3700function is bound to the result argument. This is a simpler case of
3701the <a href="#i_invoke">invoke</a> instruction.</p>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3702
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]3703<h5>Example:</h5>
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3704
3705<pre>
Nick Lewyckya9b13d52007-09-08 13:57:50 +0000[diff] [blame]3706 %retval = call i32 @test(i32 %argc)
3707 call i32 (i8 *, ...)* @printf(i8 * %msg, i32 12, i8 42);
3708 %X = tail call i32 @foo()
3709 %Y = tail call <a href="#callingconv">fastcc</a> i32 @foo()
3710 %Z = call void %foo(i8 97 signext)
Chris Lattnere23c1392005-05-06 05:47:36 +0000[diff] [blame]3711</pre>
3712
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3713</div>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3714
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3715<!-- _______________________________________________________________________ -->
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3716<div class="doc_subsubsection">
Chris Lattner33337472006-01-13 23:26:01 +0000[diff] [blame]3717 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3718</div>
3719
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3720<div class="doc_text">
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3721
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3722<h5>Syntax:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3723
3724<pre>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3725 &lt;resultval&gt; = va_arg &lt;va_list*&gt; &lt;arglist&gt;, &lt;argty&gt;
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3726</pre>
3727
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3728<h5>Overview:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3729
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3730<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3731the "variable argument" area of a function call. It is used to implement the
3732<tt>va_arg</tt> macro in C.</p>
3733
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3734<h5>Arguments:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3735
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3736<p>This instruction takes a <tt>va_list*</tt> value and the type of
3737the argument. It returns a value of the specified argument type and
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3738increments the <tt>va_list</tt> to point to the next argument. The
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3739actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3740
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3741<h5>Semantics:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3742
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3743<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
3744type from the specified <tt>va_list</tt> and causes the
3745<tt>va_list</tt> to point to the next argument. For more information,
3746see the variable argument handling <a href="#int_varargs">Intrinsic
3747Functions</a>.</p>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3748
3749<p>It is legal for this instruction to be called in a function which does not
3750take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3751function.</p>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3752
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3753<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3754href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3755argument.</p>
3756
Chris Lattner26ca62e2003-10-18 05:51:36 +0000[diff] [blame]3757<h5>Example:</h5>
Chris Lattner6a4a0492004-09-27 21:51:25 +0000[diff] [blame]3758
3759<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
3760
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3761</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3762
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3763<!-- *********************************************************************** -->
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3764<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
3765<!-- *********************************************************************** -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3766
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3767<div class="doc_text">
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3768
3769<p>LLVM supports the notion of an "intrinsic function". These functions have
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]3770well known names and semantics and are required to follow certain restrictions.
3771Overall, these intrinsics represent an extension mechanism for the LLVM
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3772language that does not require changing all of the transformations in LLVM when
Gabor Greifa54634a2007-07-06 22:07:22 +0000[diff] [blame]3773adding to the language (or the bitcode reader/writer, the parser, etc...).</p>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3774
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]3775<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3776prefix is reserved in LLVM for intrinsic names; thus, function names may not
3777begin with this prefix. Intrinsic functions must always be external functions:
3778you cannot define the body of intrinsic functions. Intrinsic functions may
3779only be used in call or invoke instructions: it is illegal to take the address
3780of an intrinsic function. Additionally, because intrinsic functions are part
3781of the LLVM language, it is required if any are added that they be documented
3782here.</p>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3783
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]3784<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents
3785a family of functions that perform the same operation but on different data
3786types. Because LLVM can represent over 8 million different integer types,
3787overloading is used commonly to allow an intrinsic function to operate on any
3788integer type. One or more of the argument types or the result type can be
3789overloaded to accept any integer type. Argument types may also be defined as
3790exactly matching a previous argument's type or the result type. This allows an
3791intrinsic function which accepts multiple arguments, but needs all of them to
3792be of the same type, to only be overloaded with respect to a single argument or
3793the result.</p>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3794
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]3795<p>Overloaded intrinsics will have the names of its overloaded argument types
3796encoded into its function name, each preceded by a period. Only those types
3797which are overloaded result in a name suffix. Arguments whose type is matched
3798against another type do not. For example, the <tt>llvm.ctpop</tt> function can
3799take an integer of any width and returns an integer of exactly the same integer
3800width. This leads to a family of functions such as
3801<tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29 %val)</tt>.
3802Only one type, the return type, is overloaded, and only one type suffix is
3803required. Because the argument's type is matched against the return type, it
3804does not require its own name suffix.</p>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]3805
3806<p>To learn how to add an intrinsic function, please see the
3807<a href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3808</p>
3809
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3810</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3811
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3812<!-- ======================================================================= -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3813<div class="doc_subsection">
3814 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
3815</div>
3816
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3817<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3818
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3819<p>Variable argument support is defined in LLVM with the <a
Chris Lattner33337472006-01-13 23:26:01 +0000[diff] [blame]3820 href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3821intrinsic functions. These functions are related to the similarly
3822named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3823
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3824<p>All of these functions operate on arguments that use a
3825target-specific value type "<tt>va_list</tt>". The LLVM assembly
3826language reference manual does not define what this type is, so all
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3827transformations should be prepared to handle these functions regardless of
3828the type used.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3829
Chris Lattner30b868d2006-05-15 17:26:46 +0000[diff] [blame]3830<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3831instruction and the variable argument handling intrinsic functions are
3832used.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3833
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]3834<div class="doc_code">
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3835<pre>
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3836define i32 @test(i32 %X, ...) {
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3837 ; Initialize variable argument processing
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3838 %ap = alloca i8*
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3839 %ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3840 call void @llvm.va_start(i8* %ap2)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3841
3842 ; Read a single integer argument
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3843 %tmp = va_arg i8** %ap, i32
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3844
3845 ; Demonstrate usage of llvm.va_copy and llvm.va_end
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3846 %aq = alloca i8*
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3847 %aq2 = bitcast i8** %aq to i8*
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3848 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3849 call void @llvm.va_end(i8* %aq2)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3850
3851 ; Stop processing of arguments.
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3852 call void @llvm.va_end(i8* %ap2)
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]3853 ret i32 %tmp
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3854}
Anton Korobeynikov640bbe02007-03-21 23:58:04 +0000[diff] [blame]3855
3856declare void @llvm.va_start(i8*)
3857declare void @llvm.va_copy(i8*, i8*)
3858declare void @llvm.va_end(i8*)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3859</pre>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3860</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3861
Bill Wendling3716c5d2007-05-29 09:04:49 +0000[diff] [blame]3862</div>
3863
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3864<!-- _______________________________________________________________________ -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3865<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3866 <a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3867</div>
3868
3869
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3870<div class="doc_text">
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3871<h5>Syntax:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3872<pre> declare void %llvm.va_start(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3873<h5>Overview:</h5>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3874<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
3875<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
3876href="#i_va_arg">va_arg</a></tt>.</p>
3877
3878<h5>Arguments:</h5>
3879
3880<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
3881
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3882<h5>Semantics:</h5>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3883
3884<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
3885macro available in C. In a target-dependent way, it initializes the
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3886<tt>va_list</tt> element to which the argument points, so that the next call to
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3887<tt>va_arg</tt> will produce the first variable argument passed to the function.
3888Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3889last argument of the function as the compiler can figure that out.</p>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3890
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3891</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3892
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3893<!-- _______________________________________________________________________ -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3894<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3895 <a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3896</div>
3897
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3898<div class="doc_text">
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3899<h5>Syntax:</h5>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3900<pre> declare void @llvm.va_end(i8* &lt;arglist&gt;)<br></pre>
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3901<h5>Overview:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3902
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3903<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*&lt;arglist&gt;</tt>,
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3904which has been initialized previously with <tt><a href="#int_va_start">llvm.va_start</a></tt>
Chris Lattner48b383b02003-11-25 01:02:51 +0000[diff] [blame]3905or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3906
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3907<h5>Arguments:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3908
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3909<p>The argument is a pointer to a <tt>va_list</tt> to destroy.</p>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3910
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3911<h5>Semantics:</h5>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3912
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3913<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3914macro available in C. In a target-dependent way, it destroys the
3915<tt>va_list</tt> element to which the argument points. Calls to <a
3916href="#int_va_start"><tt>llvm.va_start</tt></a> and <a href="#int_va_copy">
3917<tt>llvm.va_copy</tt></a> must be matched exactly with calls to
3918<tt>llvm.va_end</tt>.</p>
Chris Lattnerdb0790c2007-01-08 07:55:15 +0000[diff] [blame]3919
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3920</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3921
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3922<!-- _______________________________________________________________________ -->
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3923<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3924 <a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3925</div>
3926
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3927<div class="doc_text">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3928
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3929<h5>Syntax:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3930
3931<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]3932 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3933</pre>
3934
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3935<h5>Overview:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3936
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3937<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
3938from the source argument list to the destination argument list.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3939
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3940<h5>Arguments:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3941
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3942<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Andrew Lenharth5305ea52005-06-22 20:38:11 +0000[diff] [blame]3943The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
Andrew Lenharth5fb787c2005-06-18 18:28:17 +0000[diff] [blame]3944
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3945
Chris Lattnerbd64b4e2003-05-08 04:57:36 +0000[diff] [blame]3946<h5>Semantics:</h5>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3947
Jeff Cohen222a8a42007-04-29 01:07:00 +0000[diff] [blame]3948<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
3949macro available in C. In a target-dependent way, it copies the source
3950<tt>va_list</tt> element into the destination <tt>va_list</tt> element. This
3951intrinsic is necessary because the <tt><a href="#int_va_start">
3952llvm.va_start</a></tt> intrinsic may be arbitrarily complex and require, for
3953example, memory allocation.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3954
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]3955</div>
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]3956
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]3957<!-- ======================================================================= -->
3958<div class="doc_subsection">
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3959 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
3960</div>
3961
3962<div class="doc_text">
3963
3964<p>
3965LLVM support for <a href="GarbageCollection.html">Accurate Garbage
3966Collection</a> requires the implementation and generation of these intrinsics.
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3967These intrinsics allow identification of <a href="#int_gcroot">GC roots on the
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3968stack</a>, as well as garbage collector implementations that require <a
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3969href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a> barriers.
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3970Front-ends for type-safe garbage collected languages should generate these
3971intrinsics to make use of the LLVM garbage collectors. For more details, see <a
3972href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
3973</p>
3974</div>
3975
3976<!-- _______________________________________________________________________ -->
3977<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]3978 <a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3979</div>
3980
3981<div class="doc_text">
3982
3983<h5>Syntax:</h5>
3984
3985<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]3986 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3987</pre>
3988
3989<h5>Overview:</h5>
3990
John Criswelldfe6a862004-12-10 15:51:16 +0000[diff] [blame]3991<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]3992the code generator, and allows some metadata to be associated with it.</p>
3993
3994<h5>Arguments:</h5>
3995
3996<p>The first argument specifies the address of a stack object that contains the
3997root pointer. The second pointer (which must be either a constant or a global
3998value address) contains the meta-data to be associated with the root.</p>
3999
4000<h5>Semantics:</h5>
4001
4002<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
4003location. At compile-time, the code generator generates information to allow
4004the runtime to find the pointer at GC safe points.
4005</p>
4006
4007</div>
4008
4009
4010<!-- _______________________________________________________________________ -->
4011<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4012 <a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]4013</div>
4014
4015<div class="doc_text">
4016
4017<h5>Syntax:</h5>
4018
4019<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]4020 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]4021</pre>
4022
4023<h5>Overview:</h5>
4024
4025<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
4026locations, allowing garbage collector implementations that require read
4027barriers.</p>
4028
4029<h5>Arguments:</h5>
4030
Chris Lattnerf9228072006-03-14 20:02:51 +0000[diff] [blame]4031<p>The second argument is the address to read from, which should be an address
4032allocated from the garbage collector. The first object is a pointer to the
4033start of the referenced object, if needed by the language runtime (otherwise
4034null).</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]4035
4036<h5>Semantics:</h5>
4037
4038<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
4039instruction, but may be replaced with substantially more complex code by the
4040garbage collector runtime, as needed.</p>
4041
4042</div>
4043
4044
4045<!-- _______________________________________________________________________ -->
4046<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4047 <a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]4048</div>
4049
4050<div class="doc_text">
4051
4052<h5>Syntax:</h5>
4053
4054<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]4055 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]4056</pre>
4057
4058<h5>Overview:</h5>
4059
4060<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
4061locations, allowing garbage collector implementations that require write
4062barriers (such as generational or reference counting collectors).</p>
4063
4064<h5>Arguments:</h5>
4065
Chris Lattnerf9228072006-03-14 20:02:51 +0000[diff] [blame]4066<p>The first argument is the reference to store, the second is the start of the
4067object to store it to, and the third is the address of the field of Obj to
4068store to. If the runtime does not require a pointer to the object, Obj may be
4069null.</p>
Chris Lattner757528b0b2004-05-23 21:06:01 +0000[diff] [blame]4070
4071<h5>Semantics:</h5>
4072
4073<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
4074instruction, but may be replaced with substantially more complex code by the
4075garbage collector runtime, as needed.</p>
4076
4077</div>
4078
4079
4080
4081<!-- ======================================================================= -->
4082<div class="doc_subsection">
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4083 <a name="int_codegen">Code Generator Intrinsics</a>
4084</div>
4085
4086<div class="doc_text">
4087<p>
4088These intrinsics are provided by LLVM to expose special features that may only
4089be implemented with code generator support.
4090</p>
4091
4092</div>
4093
4094<!-- _______________________________________________________________________ -->
4095<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4096 <a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4097</div>
4098
4099<div class="doc_text">
4100
4101<h5>Syntax:</h5>
4102<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4103 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4104</pre>
4105
4106<h5>Overview:</h5>
4107
4108<p>
Chris Lattnerc1fb4262006-10-15 20:05:59 +0000[diff] [blame]4109The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
4110target-specific value indicating the return address of the current function
4111or one of its callers.
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4112</p>
4113
4114<h5>Arguments:</h5>
4115
4116<p>
4117The argument to this intrinsic indicates which function to return the address
4118for. Zero indicates the calling function, one indicates its caller, etc. The
4119argument is <b>required</b> to be a constant integer value.
4120</p>
4121
4122<h5>Semantics:</h5>
4123
4124<p>
4125The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
4126the return address of the specified call frame, or zero if it cannot be
4127identified. The value returned by this intrinsic is likely to be incorrect or 0
4128for arguments other than zero, so it should only be used for debugging purposes.
4129</p>
4130
4131<p>
4132Note that calling this intrinsic does not prevent function inlining or other
Chris Lattner2e6eb5f2005-03-07 20:30:51 +0000[diff] [blame]4133aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4134source-language caller.
4135</p>
4136</div>
4137
4138
4139<!-- _______________________________________________________________________ -->
4140<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4141 <a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4142</div>
4143
4144<div class="doc_text">
4145
4146<h5>Syntax:</h5>
4147<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]4148 declare i8 *@llvm.frameaddress(i32 &lt;level&gt;)
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4149</pre>
4150
4151<h5>Overview:</h5>
4152
4153<p>
Chris Lattnerc1fb4262006-10-15 20:05:59 +0000[diff] [blame]4154The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
4155target-specific frame pointer value for the specified stack frame.
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4156</p>
4157
4158<h5>Arguments:</h5>
4159
4160<p>
4161The argument to this intrinsic indicates which function to return the frame
4162pointer for. Zero indicates the calling function, one indicates its caller,
4163etc. The argument is <b>required</b> to be a constant integer value.
4164</p>
4165
4166<h5>Semantics:</h5>
4167
4168<p>
4169The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
4170the frame address of the specified call frame, or zero if it cannot be
4171identified. The value returned by this intrinsic is likely to be incorrect or 0
4172for arguments other than zero, so it should only be used for debugging purposes.
4173</p>
4174
4175<p>
4176Note that calling this intrinsic does not prevent function inlining or other
Chris Lattner2e6eb5f2005-03-07 20:30:51 +0000[diff] [blame]4177aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4178source-language caller.
4179</p>
4180</div>
4181
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4182<!-- _______________________________________________________________________ -->
4183<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4184 <a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4185</div>
4186
4187<div class="doc_text">
4188
4189<h5>Syntax:</h5>
4190<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]4191 declare i8 *@llvm.stacksave()
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4192</pre>
4193
4194<h5>Overview:</h5>
4195
4196<p>
4197The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4198the function stack, for use with <a href="#int_stackrestore">
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4199<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
4200features like scoped automatic variable sized arrays in C99.
4201</p>
4202
4203<h5>Semantics:</h5>
4204
4205<p>
4206This intrinsic returns a opaque pointer value that can be passed to <a
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4207href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4208<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
4209<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
4210state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
4211practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
4212that were allocated after the <tt>llvm.stacksave</tt> was executed.
4213</p>
4214
4215</div>
4216
4217<!-- _______________________________________________________________________ -->
4218<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4219 <a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4220</div>
4221
4222<div class="doc_text">
4223
4224<h5>Syntax:</h5>
4225<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4226 declare void @llvm.stackrestore(i8 * %ptr)
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4227</pre>
4228
4229<h5>Overview:</h5>
4230
4231<p>
4232The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
4233the function stack to the state it was in when the corresponding <a
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4234href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4235useful for implementing language features like scoped automatic variable sized
4236arrays in C99.
4237</p>
4238
4239<h5>Semantics:</h5>
4240
4241<p>
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4242See the description for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.
Chris Lattner2f0f0012006-01-13 02:03:13 +0000[diff] [blame]4243</p>
4244
4245</div>
4246
4247
4248<!-- _______________________________________________________________________ -->
4249<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4250 <a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4251</div>
4252
4253<div class="doc_text">
4254
4255<h5>Syntax:</h5>
4256<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]4257 declare void @llvm.prefetch(i8* &lt;address&gt;, i32 &lt;rw&gt;, i32 &lt;locality&gt;)
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4258</pre>
4259
4260<h5>Overview:</h5>
4261
4262
4263<p>
4264The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]4265a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
4266no
4267effect on the behavior of the program but can change its performance
Chris Lattnerff851072005-02-28 19:47:14 +0000[diff] [blame]4268characteristics.
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4269</p>
4270
4271<h5>Arguments:</h5>
4272
4273<p>
4274<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
4275determining if the fetch should be for a read (0) or write (1), and
4276<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattnerd3e641c2005-03-07 20:31:38 +0000[diff] [blame]4277locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattnerc8a2c222005-02-28 19:24:19 +0000[diff] [blame]4278<tt>locality</tt> arguments must be constant integers.
4279</p>
4280
4281<h5>Semantics:</h5>
4282
4283<p>
4284This intrinsic does not modify the behavior of the program. In particular,
4285prefetches cannot trap and do not produce a value. On targets that support this
4286intrinsic, the prefetch can provide hints to the processor cache for better
4287performance.
4288</p>
4289
4290</div>
4291
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4292<!-- _______________________________________________________________________ -->
4293<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4294 <a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4295</div>
4296
4297<div class="doc_text">
4298
4299<h5>Syntax:</h5>
4300<pre>
Chris Lattner12477732007-09-21 17:30:40 +0000[diff] [blame]4301 declare void @llvm.pcmarker(i32 &lt;id&gt;)
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4302</pre>
4303
4304<h5>Overview:</h5>
4305
4306
4307<p>
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]4308The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
4309(PC) in a region of
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4310code to simulators and other tools. The method is target specific, but it is
4311expected that the marker will use exported symbols to transmit the PC of the marker.
Jeff Cohendc6bfea2005-11-11 02:15:27 +0000[diff] [blame]4312The marker makes no guarantees that it will remain with any specific instruction
Chris Lattnere64d41d2005-11-15 06:07:55 +0000[diff] [blame]4313after optimizations. It is possible that the presence of a marker will inhibit
Chris Lattnerb40261e2006-03-24 07:16:10 +0000[diff] [blame]4314optimizations. The intended use is to be inserted after optimizations to allow
John Criswell88190562005-05-16 16:17:45 +0000[diff] [blame]4315correlations of simulation runs.
Andrew Lenharthb4427912005-03-28 20:05:49 +0000[diff] [blame]4316</p>
4317
4318<h5>Arguments:</h5>
4319
4320<p>
4321<tt>id</tt> is a numerical id identifying the marker.
4322</p>
4323
4324<h5>Semantics:</h5>
4325
4326<p>
4327This intrinsic does not modify the behavior of the program. Backends that do not
4328support this intrinisic may ignore it.
4329</p>
4330
4331</div>
4332
Andrew Lenharth01aa5632005-11-11 16:47:30 +0000[diff] [blame]4333<!-- _______________________________________________________________________ -->
4334<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4335 <a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
Andrew Lenharth01aa5632005-11-11 16:47:30 +0000[diff] [blame]4336</div>
4337
4338<div class="doc_text">
4339
4340<h5>Syntax:</h5>
4341<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4342 declare i64 @llvm.readcyclecounter( )
Andrew Lenharth01aa5632005-11-11 16:47:30 +0000[diff] [blame]4343</pre>
4344
4345<h5>Overview:</h5>
4346
4347
4348<p>
4349The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
4350counter register (or similar low latency, high accuracy clocks) on those targets
4351that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
4352As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
4353should only be used for small timings.
4354</p>
4355
4356<h5>Semantics:</h5>
4357
4358<p>
4359When directly supported, reading the cycle counter should not modify any memory.
4360Implementations are allowed to either return a application specific value or a
4361system wide value. On backends without support, this is lowered to a constant 0.
4362</p>
4363
4364</div>
4365
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4366<!-- ======================================================================= -->
4367<div class="doc_subsection">
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4368 <a name="int_libc">Standard C Library Intrinsics</a>
4369</div>
4370
4371<div class="doc_text">
4372<p>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4373LLVM provides intrinsics for a few important standard C library functions.
4374These intrinsics allow source-language front-ends to pass information about the
4375alignment of the pointer arguments to the code generator, providing opportunity
4376for more efficient code generation.
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4377</p>
4378
4379</div>
4380
4381<!-- _______________________________________________________________________ -->
4382<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4383 <a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4384</div>
4385
4386<div class="doc_text">
4387
4388<h5>Syntax:</h5>
4389<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4390 declare void @llvm.memcpy.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4391 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4392 declare void @llvm.memcpy.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4393 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4394</pre>
4395
4396<h5>Overview:</h5>
4397
4398<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4399The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4400location to the destination location.
4401</p>
4402
4403<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4404Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
4405intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4406</p>
4407
4408<h5>Arguments:</h5>
4409
4410<p>
4411The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4412the source. The third argument is an integer argument
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4413specifying the number of bytes to copy, and the fourth argument is the alignment
4414of the source and destination locations.
4415</p>
4416
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4417<p>
4418If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner5316e5d2006-03-04 00:02:10 +0000[diff] [blame]4419the caller guarantees that both the source and destination pointers are aligned
4420to that boundary.
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4421</p>
4422
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4423<h5>Semantics:</h5>
4424
4425<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4426The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattnerfee11462004-02-12 17:01:32 +0000[diff] [blame]4427location to the destination location, which are not allowed to overlap. It
4428copies "len" bytes of memory over. If the argument is known to be aligned to
4429some boundary, this can be specified as the fourth argument, otherwise it should
4430be set to 0 or 1.
4431</p>
4432</div>
4433
4434
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4435<!-- _______________________________________________________________________ -->
4436<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4437 <a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4438</div>
4439
4440<div class="doc_text">
4441
4442<h5>Syntax:</h5>
4443<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4444 declare void @llvm.memmove.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4445 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4446 declare void @llvm.memmove.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4447 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4448</pre>
4449
4450<h5>Overview:</h5>
4451
4452<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4453The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the source
4454location to the destination location. It is similar to the
4455'<tt>llvm.memcmp</tt>' intrinsic but allows the two memory locations to overlap.
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4456</p>
4457
4458<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4459Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
4460intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4461</p>
4462
4463<h5>Arguments:</h5>
4464
4465<p>
4466The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4467the source. The third argument is an integer argument
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4468specifying the number of bytes to copy, and the fourth argument is the alignment
4469of the source and destination locations.
4470</p>
4471
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4472<p>
4473If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner5316e5d2006-03-04 00:02:10 +0000[diff] [blame]4474the caller guarantees that the source and destination pointers are aligned to
4475that boundary.
Chris Lattner4c67c482004-02-12 21:18:15 +0000[diff] [blame]4476</p>
4477
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4478<h5>Semantics:</h5>
4479
4480<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4481The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the source
Chris Lattnerf30152e2004-02-12 18:10:10 +0000[diff] [blame]4482location to the destination location, which may overlap. It
4483copies "len" bytes of memory over. If the argument is known to be aligned to
4484some boundary, this can be specified as the fourth argument, otherwise it should
4485be set to 0 or 1.
4486</p>
4487</div>
4488
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]4489
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4490<!-- _______________________________________________________________________ -->
4491<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4492 <a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4493</div>
4494
4495<div class="doc_text">
4496
4497<h5>Syntax:</h5>
4498<pre>
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4499 declare void @llvm.memset.i32(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4500 i32 &lt;len&gt;, i32 &lt;align&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4501 declare void @llvm.memset.i64(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4502 i64 &lt;len&gt;, i32 &lt;align&gt;)
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4503</pre>
4504
4505<h5>Overview:</h5>
4506
4507<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4508The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a particular
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4509byte value.
4510</p>
4511
4512<p>
4513Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
4514does not return a value, and takes an extra alignment argument.
4515</p>
4516
4517<h5>Arguments:</h5>
4518
4519<p>
4520The first argument is a pointer to the destination to fill, the second is the
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4521byte value to fill it with, the third argument is an integer
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4522argument specifying the number of bytes to fill, and the fourth argument is the
4523known alignment of destination location.
4524</p>
4525
4526<p>
4527If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner5316e5d2006-03-04 00:02:10 +0000[diff] [blame]4528the caller guarantees that the destination pointer is aligned to that boundary.
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4529</p>
4530
4531<h5>Semantics:</h5>
4532
4533<p>
Chris Lattner0c8b2592006-03-03 00:07:20 +0000[diff] [blame]4534The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting at
4535the
Chris Lattner3649c3a2004-02-14 04:08:35 +0000[diff] [blame]4536destination location. If the argument is known to be aligned to some boundary,
4537this can be specified as the fourth argument, otherwise it should be set to 0 or
45381.
4539</p>
4540</div>
4541
4542
Chris Lattner3b4f4372004-06-11 02:28:03 +0000[diff] [blame]4543<!-- _______________________________________________________________________ -->
4544<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4545 <a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4546</div>
4547
4548<div class="doc_text">
4549
4550<h5>Syntax:</h5>
Dale Johannesendd89d272007-10-02 17:47:38 +0000[diff] [blame]4551<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4552floating point or vector of floating point type. Not all targets support all
4553types however.
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4554<pre>
Dale Johannesendd89d272007-10-02 17:47:38 +0000[diff] [blame]4555 declare float @llvm.sqrt.f32(float %Val)
4556 declare double @llvm.sqrt.f64(double %Val)
4557 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
4558 declare fp128 @llvm.sqrt.f128(fp128 %Val)
4559 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4560</pre>
4561
4562<h5>Overview:</h5>
4563
4564<p>
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4565The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4566returning the same value as the libm '<tt>sqrt</tt>' functions would. Unlike
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4567<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
4568negative numbers (which allows for better optimization).
4569</p>
4570
4571<h5>Arguments:</h5>
4572
4573<p>
4574The argument and return value are floating point numbers of the same type.
4575</p>
4576
4577<h5>Semantics:</h5>
4578
4579<p>
Dan Gohman33988db2007-07-16 14:37:41 +0000[diff] [blame]4580This function returns the sqrt of the specified operand if it is a nonnegative
Chris Lattner8a8f2e52005-07-21 01:29:16 +0000[diff] [blame]4581floating point number.
4582</p>
4583</div>
4584
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4585<!-- _______________________________________________________________________ -->
4586<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4587 <a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4588</div>
4589
4590<div class="doc_text">
4591
4592<h5>Syntax:</h5>
Dale Johannesendd89d272007-10-02 17:47:38 +0000[diff] [blame]4593<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4594floating point or vector of floating point type. Not all targets support all
4595types however.
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4596<pre>
Dale Johannesendd89d272007-10-02 17:47:38 +0000[diff] [blame]4597 declare float @llvm.powi.f32(float %Val, i32 %power)
4598 declare double @llvm.powi.f64(double %Val, i32 %power)
4599 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
4600 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
4601 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4602</pre>
4603
4604<h5>Overview:</h5>
4605
4606<p>
4607The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
4608specified (positive or negative) power. The order of evaluation of
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4609multiplications is not defined. When a vector of floating point type is
4610used, the second argument remains a scalar integer value.
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4611</p>
4612
4613<h5>Arguments:</h5>
4614
4615<p>
4616The second argument is an integer power, and the first is a value to raise to
4617that power.
4618</p>
4619
4620<h5>Semantics:</h5>
4621
4622<p>
4623This function returns the first value raised to the second power with an
4624unspecified sequence of rounding operations.</p>
4625</div>
4626
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4627<!-- _______________________________________________________________________ -->
4628<div class="doc_subsubsection">
4629 <a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
4630</div>
4631
4632<div class="doc_text">
4633
4634<h5>Syntax:</h5>
4635<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
4636floating point or vector of floating point type. Not all targets support all
4637types however.
4638<pre>
4639 declare float @llvm.sin.f32(float %Val)
4640 declare double @llvm.sin.f64(double %Val)
4641 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
4642 declare fp128 @llvm.sin.f128(fp128 %Val)
4643 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
4644</pre>
4645
4646<h5>Overview:</h5>
4647
4648<p>
4649The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.
4650</p>
4651
4652<h5>Arguments:</h5>
4653
4654<p>
4655The argument and return value are floating point numbers of the same type.
4656</p>
4657
4658<h5>Semantics:</h5>
4659
4660<p>
4661This function returns the sine of the specified operand, returning the
4662same values as the libm <tt>sin</tt> functions would, and handles error
Dan Gohmand0806a02007-10-17 18:05:13 +0000[diff] [blame]4663conditions in the same way.</p>
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4664</div>
4665
4666<!-- _______________________________________________________________________ -->
4667<div class="doc_subsubsection">
4668 <a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
4669</div>
4670
4671<div class="doc_text">
4672
4673<h5>Syntax:</h5>
4674<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
4675floating point or vector of floating point type. Not all targets support all
4676types however.
4677<pre>
4678 declare float @llvm.cos.f32(float %Val)
4679 declare double @llvm.cos.f64(double %Val)
4680 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
4681 declare fp128 @llvm.cos.f128(fp128 %Val)
4682 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
4683</pre>
4684
4685<h5>Overview:</h5>
4686
4687<p>
4688The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.
4689</p>
4690
4691<h5>Arguments:</h5>
4692
4693<p>
4694The argument and return value are floating point numbers of the same type.
4695</p>
4696
4697<h5>Semantics:</h5>
4698
4699<p>
4700This function returns the cosine of the specified operand, returning the
4701same values as the libm <tt>cos</tt> functions would, and handles error
Dan Gohmand0806a02007-10-17 18:05:13 +0000[diff] [blame]4702conditions in the same way.</p>
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4703</div>
4704
4705<!-- _______________________________________________________________________ -->
4706<div class="doc_subsubsection">
4707 <a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
4708</div>
4709
4710<div class="doc_text">
4711
4712<h5>Syntax:</h5>
4713<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
4714floating point or vector of floating point type. Not all targets support all
4715types however.
4716<pre>
4717 declare float @llvm.pow.f32(float %Val, float %Power)
4718 declare double @llvm.pow.f64(double %Val, double %Power)
4719 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
4720 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
4721 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
4722</pre>
4723
4724<h5>Overview:</h5>
4725
4726<p>
4727The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
4728specified (positive or negative) power.
4729</p>
4730
4731<h5>Arguments:</h5>
4732
4733<p>
4734The second argument is a floating point power, and the first is a value to
4735raise to that power.
4736</p>
4737
4738<h5>Semantics:</h5>
4739
4740<p>
4741This function returns the first value raised to the second power,
4742returning the
4743same values as the libm <tt>pow</tt> functions would, and handles error
Dan Gohmand0806a02007-10-17 18:05:13 +0000[diff] [blame]4744conditions in the same way.</p>
Dan Gohmanb6324c12007-10-15 20:30:11 +0000[diff] [blame]4745</div>
4746
Chris Lattner33b73f92006-09-08 06:34:02 +0000[diff] [blame]4747
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4748<!-- ======================================================================= -->
4749<div class="doc_subsection">
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4750 <a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4751</div>
4752
4753<div class="doc_text">
4754<p>
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4755LLVM provides intrinsics for a few important bit manipulation operations.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4756These allow efficient code generation for some algorithms.
4757</p>
4758
4759</div>
4760
4761<!-- _______________________________________________________________________ -->
4762<div class="doc_subsubsection">
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4763 <a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4764</div>
4765
4766<div class="doc_text">
4767
4768<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4769<p>This is an overloaded intrinsic function. You can use bswap on any integer
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4770type that is an even number of bytes (i.e. BitWidth % 16 == 0).
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4771<pre>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4772 declare i16 @llvm.bswap.i16(i16 &lt;id&gt;)
4773 declare i32 @llvm.bswap.i32(i32 &lt;id&gt;)
4774 declare i64 @llvm.bswap.i64(i64 &lt;id&gt;)
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4775</pre>
4776
4777<h5>Overview:</h5>
4778
4779<p>
Reid Spencerf361c4f2007-04-02 02:25:19 +0000[diff] [blame]4780The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4781values with an even number of bytes (positive multiple of 16 bits). These are
4782useful for performing operations on data that is not in the target's native
4783byte order.
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4784</p>
4785
4786<h5>Semantics:</h5>
4787
4788<p>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4789The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4790and low byte of the input i16 swapped. Similarly, the <tt>llvm.bswap.i32</tt>
4791intrinsic returns an i32 value that has the four bytes of the input i32
4792swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the returned
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4793i32 will have its bytes in 3, 2, 1, 0 order. The <tt>llvm.bswap.i48</tt>,
4794<tt>llvm.bswap.i64</tt> and other intrinsics extend this concept to
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4795additional even-byte lengths (6 bytes, 8 bytes and more, respectively).
Nate Begeman0f223bb2006-01-13 23:26:38 +0000[diff] [blame]4796</p>
4797
4798</div>
4799
4800<!-- _______________________________________________________________________ -->
4801<div class="doc_subsubsection">
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4802 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4803</div>
4804
4805<div class="doc_text">
4806
4807<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4808<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
4809width. Not all targets support all bit widths however.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4810<pre>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4811 declare i8 @llvm.ctpop.i8 (i8 &lt;src&gt;)
4812 declare i16 @llvm.ctpop.i16(i16 &lt;src&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4813 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4814 declare i64 @llvm.ctpop.i64(i64 &lt;src&gt;)
4815 declare i256 @llvm.ctpop.i256(i256 &lt;src&gt;)
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4816</pre>
4817
4818<h5>Overview:</h5>
4819
4820<p>
Chris Lattner069b5bd2006-01-16 22:38:59 +0000[diff] [blame]4821The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set in a
4822value.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4823</p>
4824
4825<h5>Arguments:</h5>
4826
4827<p>
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]4828The only argument is the value to be counted. The argument may be of any
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]4829integer type. The return type must match the argument type.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4830</p>
4831
4832<h5>Semantics:</h5>
4833
4834<p>
4835The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
4836</p>
4837</div>
4838
4839<!-- _______________________________________________________________________ -->
4840<div class="doc_subsubsection">
Chris Lattnerb748c672006-01-16 22:34:14 +0000[diff] [blame]4841 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4842</div>
4843
4844<div class="doc_text">
4845
4846<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4847<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
4848integer bit width. Not all targets support all bit widths however.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4849<pre>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4850 declare i8 @llvm.ctlz.i8 (i8 &lt;src&gt;)
4851 declare i16 @llvm.ctlz.i16(i16 &lt;src&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4852 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4853 declare i64 @llvm.ctlz.i64(i64 &lt;src&gt;)
4854 declare i256 @llvm.ctlz.i256(i256 &lt;src&gt;)
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4855</pre>
4856
4857<h5>Overview:</h5>
4858
4859<p>
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4860The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
4861leading zeros in a variable.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4862</p>
4863
4864<h5>Arguments:</h5>
4865
4866<p>
Chris Lattner573f64e2005-05-07 01:46:40 +0000[diff] [blame]4867The only argument is the value to be counted. The argument may be of any
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]4868integer type. The return type must match the argument type.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4869</p>
4870
4871<h5>Semantics:</h5>
4872
4873<p>
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4874The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
4875in a variable. If the src == 0 then the result is the size in bits of the type
Reid Spencerb5ebf3d2006-12-31 07:07:53 +0000[diff] [blame]4876of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.
Andrew Lenharth1d463522005-05-03 18:01:48 +0000[diff] [blame]4877</p>
4878</div>
Chris Lattner3b4f4372004-06-11 02:28:03 +0000[diff] [blame]4879
4880
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4881
4882<!-- _______________________________________________________________________ -->
4883<div class="doc_subsubsection">
Chris Lattnerb748c672006-01-16 22:34:14 +0000[diff] [blame]4884 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4885</div>
4886
4887<div class="doc_text">
4888
4889<h5>Syntax:</h5>
Reid Spencer4eefaab2007-04-01 08:04:23 +0000[diff] [blame]4890<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
4891integer bit width. Not all targets support all bit widths however.
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4892<pre>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4893 declare i8 @llvm.cttz.i8 (i8 &lt;src&gt;)
4894 declare i16 @llvm.cttz.i16(i16 &lt;src&gt;)
Anton Korobeynikovbe9c93c2007-03-22 00:02:17 +0000[diff] [blame]4895 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4896 declare i64 @llvm.cttz.i64(i64 &lt;src&gt;)
4897 declare i256 @llvm.cttz.i256(i256 &lt;src&gt;)
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4898</pre>
4899
4900<h5>Overview:</h5>
4901
4902<p>
Reid Spencerb4f9a6f2006-01-16 21:12:35 +0000[diff] [blame]4903The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
4904trailing zeros.
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4905</p>
4906
4907<h5>Arguments:</h5>
4908
4909<p>
4910The only argument is the value to be counted. The argument may be of any
Reid Spencer3e628eb92007-01-04 16:43:23 +0000[diff] [blame]4911integer type. The return type must match the argument type.
Chris Lattnerefa20fa2005-05-15 19:39:26 +0000[diff] [blame]4912</p>
4913
4914<h5>Semantics:</h5>
4915
4916<p>
4917The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
4918in a variable. If the src == 0 then the result is the size in bits of the type
4919of src. For example, <tt>llvm.cttz(2) = 1</tt>.
4920</p>
4921</div>
4922
Reid Spencer8a5799f2007-04-01 08:27:01 +0000[diff] [blame]4923<!-- _______________________________________________________________________ -->
4924<div class="doc_subsubsection">
Reid Spencerea2945e2007-04-10 02:51:31 +0000[diff] [blame]4925 <a name="int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic</a>
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4926</div>
4927
4928<div class="doc_text">
4929
4930<h5>Syntax:</h5>
Reid Spencerea2945e2007-04-10 02:51:31 +0000[diff] [blame]4931<p>This is an overloaded intrinsic. You can use <tt>llvm.part.select</tt>
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4932on any integer bit width.
4933<pre>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4934 declare i17 @llvm.part.select.i17 (i17 %val, i32 %loBit, i32 %hiBit)
4935 declare i29 @llvm.part.select.i29 (i29 %val, i32 %loBit, i32 %hiBit)
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4936</pre>
4937
4938<h5>Overview:</h5>
Reid Spencerea2945e2007-04-10 02:51:31 +0000[diff] [blame]4939<p>The '<tt>llvm.part.select</tt>' family of intrinsic functions selects a
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4940range of bits from an integer value and returns them in the same bit width as
4941the original value.</p>
4942
4943<h5>Arguments:</h5>
4944<p>The first argument, <tt>%val</tt> and the result may be integer types of
4945any bit width but they must have the same bit width. The second and third
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4946arguments must be <tt>i32</tt> type since they specify only a bit index.</p>
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4947
4948<h5>Semantics:</h5>
Reid Spencerea2945e2007-04-10 02:51:31 +0000[diff] [blame]4949<p>The operation of the '<tt>llvm.part.select</tt>' intrinsic has two modes
Reid Spencer96a5f022007-04-04 02:42:35 +0000[diff] [blame]4950of operation: forwards and reverse. If <tt>%loBit</tt> is greater than
4951<tt>%hiBits</tt> then the intrinsic operates in reverse mode. Otherwise it
4952operates in forward mode.</p>
4953<p>In forward mode, this intrinsic is the equivalent of shifting <tt>%val</tt>
4954right by <tt>%loBit</tt> bits and then ANDing it with a mask with
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4955only the <tt>%hiBit - %loBit</tt> bits set, as follows:</p>
4956<ol>
4957 <li>The <tt>%val</tt> is shifted right (LSHR) by the number of bits specified
4958 by <tt>%loBits</tt>. This normalizes the value to the low order bits.</li>
4959 <li>The <tt>%loBits</tt> value is subtracted from the <tt>%hiBits</tt> value
4960 to determine the number of bits to retain.</li>
4961 <li>A mask of the retained bits is created by shifting a -1 value.</li>
4962 <li>The mask is ANDed with <tt>%val</tt> to produce the result.
4963</ol>
Reid Spencer70845c02007-05-14 16:14:57 +0000[diff] [blame]4964<p>In reverse mode, a similar computation is made except that the bits are
4965returned in the reverse order. So, for example, if <tt>X</tt> has the value
4966<tt>i16 0x0ACF (101011001111)</tt> and we apply
4967<tt>part.select(i16 X, 8, 3)</tt> to it, we get back the value
4968<tt>i16 0x0026 (000000100110)</tt>.</p>
Reid Spencer8bc7d952007-04-01 19:00:37 +0000[diff] [blame]4969</div>
4970
Reid Spencer5bf54c82007-04-11 23:23:49 +0000[diff] [blame]4971<div class="doc_subsubsection">
4972 <a name="int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic</a>
4973</div>
4974
4975<div class="doc_text">
4976
4977<h5>Syntax:</h5>
4978<p>This is an overloaded intrinsic. You can use <tt>llvm.part.set</tt>
4979on any integer bit width.
4980<pre>
Chandler Carruth7132e002007-08-04 01:51:18 +0000[diff] [blame]4981 declare i17 @llvm.part.set.i17.i9 (i17 %val, i9 %repl, i32 %lo, i32 %hi)
4982 declare i29 @llvm.part.set.i29.i9 (i29 %val, i9 %repl, i32 %lo, i32 %hi)
Reid Spencer5bf54c82007-04-11 23:23:49 +0000[diff] [blame]4983</pre>
4984
4985<h5>Overview:</h5>
4986<p>The '<tt>llvm.part.set</tt>' family of intrinsic functions replaces a range
4987of bits in an integer value with another integer value. It returns the integer
4988with the replaced bits.</p>
4989
4990<h5>Arguments:</h5>
4991<p>The first argument, <tt>%val</tt> and the result may be integer types of
4992any bit width but they must have the same bit width. <tt>%val</tt> is the value
4993whose bits will be replaced. The second argument, <tt>%repl</tt> may be an
4994integer of any bit width. The third and fourth arguments must be <tt>i32</tt>
4995type since they specify only a bit index.</p>
4996
4997<h5>Semantics:</h5>
4998<p>The operation of the '<tt>llvm.part.set</tt>' intrinsic has two modes
4999of operation: forwards and reverse. If <tt>%lo</tt> is greater than
5000<tt>%hi</tt> then the intrinsic operates in reverse mode. Otherwise it
5001operates in forward mode.</p>
5002<p>For both modes, the <tt>%repl</tt> value is prepared for use by either
5003truncating it down to the size of the replacement area or zero extending it
5004up to that size.</p>
5005<p>In forward mode, the bits between <tt>%lo</tt> and <tt>%hi</tt> (inclusive)
5006are replaced with corresponding bits from <tt>%repl</tt>. That is the 0th bit
5007in <tt>%repl</tt> replaces the <tt>%lo</tt>th bit in <tt>%val</tt> and etc. up
5008to the <tt>%hi</tt>th bit.
Reid Spencer146281c2007-05-14 16:50:20 +0000[diff] [blame]5009<p>In reverse mode, a similar computation is made except that the bits are
5010reversed. That is, the <tt>0</tt>th bit in <tt>%repl</tt> replaces the
5011<tt>%hi</tt> bit in <tt>%val</tt> and etc. down to the <tt>%lo</tt>th bit.
Reid Spencer5bf54c82007-04-11 23:23:49 +0000[diff] [blame]5012<h5>Examples:</h5>
5013<pre>
Reid Spencerc70afc32007-04-12 01:03:03 +0000[diff] [blame]5014 llvm.part.set(0xFFFF, 0, 4, 7) -&gt; 0xFF0F
Reid Spencer146281c2007-05-14 16:50:20 +0000[diff] [blame]5015 llvm.part.set(0xFFFF, 0, 7, 4) -&gt; 0xFF0F
5016 llvm.part.set(0xFFFF, 1, 7, 4) -&gt; 0xFF8F
5017 llvm.part.set(0xFFFF, F, 8, 3) -&gt; 0xFFE7
Reid Spencerc70afc32007-04-12 01:03:03 +0000[diff] [blame]5018 llvm.part.set(0xFFFF, 0, 3, 8) -&gt; 0xFE07
Reid Spencer7972c472007-04-11 23:49:50 +0000[diff] [blame]5019</pre>
Reid Spencer5bf54c82007-04-11 23:23:49 +0000[diff] [blame]5020</div>
5021
Chris Lattner941515c2004-01-06 05:31:32 +0000[diff] [blame]5022<!-- ======================================================================= -->
5023<div class="doc_subsection">
5024 <a name="int_debugger">Debugger Intrinsics</a>
5025</div>
5026
5027<div class="doc_text">
5028<p>
5029The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
5030are described in the <a
5031href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
5032Debugging</a> document.
5033</p>
5034</div>
5035
5036
Jim Laskey2211f492007-03-14 19:31:19 +0000[diff] [blame]5037<!-- ======================================================================= -->
5038<div class="doc_subsection">
5039 <a name="int_eh">Exception Handling Intrinsics</a>
5040</div>
5041
5042<div class="doc_text">
5043<p> The LLVM exception handling intrinsics (which all start with
5044<tt>llvm.eh.</tt> prefix), are described in the <a
5045href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
5046Handling</a> document. </p>
5047</div>
5048
Tanya Lattnercb1b9602007-06-15 20:50:54 +0000[diff] [blame]5049<!-- ======================================================================= -->
5050<div class="doc_subsection">
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5051 <a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5052</div>
5053
5054<div class="doc_text">
5055<p>
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5056 This intrinsic makes it possible to excise one parameter, marked with
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5057 the <tt>nest</tt> attribute, from a function. The result is a callable
5058 function pointer lacking the nest parameter - the caller does not need
5059 to provide a value for it. Instead, the value to use is stored in
5060 advance in a "trampoline", a block of memory usually allocated
5061 on the stack, which also contains code to splice the nest value into the
5062 argument list. This is used to implement the GCC nested function address
5063 extension.
5064</p>
5065<p>
5066 For example, if the function is
5067 <tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
Bill Wendling252570f2007-09-22 09:23:55 +0000[diff] [blame]5068 pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as follows:</p>
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5069<pre>
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5070 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
5071 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
5072 %p = call i8* @llvm.init.trampoline( i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8*), i8* %nval )
5073 %fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5074</pre>
Bill Wendling252570f2007-09-22 09:23:55 +0000[diff] [blame]5075 <p>The call <tt>%val = call i32 %fp( i32 %x, i32 %y )</tt> is then equivalent
5076 to <tt>%val = call i32 %f( i8* %nval, i32 %x, i32 %y )</tt>.</p>
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5077</div>
5078
5079<!-- _______________________________________________________________________ -->
5080<div class="doc_subsubsection">
5081 <a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
5082</div>
5083<div class="doc_text">
5084<h5>Syntax:</h5>
5085<pre>
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5086declare i8* @llvm.init.trampoline(i8* &lt;tramp&gt;, i8* &lt;func&gt;, i8* &lt;nval&gt;)
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5087</pre>
5088<h5>Overview:</h5>
5089<p>
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5090 This fills the memory pointed to by <tt>tramp</tt> with code
5091 and returns a function pointer suitable for executing it.
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5092</p>
5093<h5>Arguments:</h5>
5094<p>
5095 The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
5096 pointers. The <tt>tramp</tt> argument must point to a sufficiently large
5097 and sufficiently aligned block of memory; this memory is written to by the
Duncan Sandsf2bcd372007-08-22 23:39:54 +0000[diff] [blame]5098 intrinsic. Note that the size and the alignment are target-specific - LLVM
5099 currently provides no portable way of determining them, so a front-end that
5100 generates this intrinsic needs to have some target-specific knowledge.
5101 The <tt>func</tt> argument must hold a function bitcast to an <tt>i8*</tt>.
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5102</p>
5103<h5>Semantics:</h5>
5104<p>
5105 The block of memory pointed to by <tt>tramp</tt> is filled with target
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5106 dependent code, turning it into a function. A pointer to this function is
5107 returned, but needs to be bitcast to an
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5108 <a href="#int_trampoline">appropriate function pointer type</a>
Duncan Sands86e01192007-09-11 14:10:23 +0000[diff] [blame]5109 before being called. The new function's signature is the same as that of
5110 <tt>func</tt> with any arguments marked with the <tt>nest</tt> attribute
5111 removed. At most one such <tt>nest</tt> argument is allowed, and it must be
5112 of pointer type. Calling the new function is equivalent to calling
5113 <tt>func</tt> with the same argument list, but with <tt>nval</tt> used for the
5114 missing <tt>nest</tt> argument. If, after calling
5115 <tt>llvm.init.trampoline</tt>, the memory pointed to by <tt>tramp</tt> is
5116 modified, then the effect of any later call to the returned function pointer is
5117 undefined.
Duncan Sands644f9172007-07-27 12:58:54 +0000[diff] [blame]5118</p>
5119</div>
5120
5121<!-- ======================================================================= -->
5122<div class="doc_subsection">
Tanya Lattnercb1b9602007-06-15 20:50:54 +0000[diff] [blame]5123 <a name="int_general">General Intrinsics</a>
5124</div>
5125
5126<div class="doc_text">
5127<p> This class of intrinsics is designed to be generic and has
5128no specific purpose. </p>
5129</div>
5130
5131<!-- _______________________________________________________________________ -->
5132<div class="doc_subsubsection">
5133 <a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
5134</div>
5135
5136<div class="doc_text">
5137
5138<h5>Syntax:</h5>
5139<pre>
Tanya Lattnerbed1d4d2007-06-18 23:42:37 +0000[diff] [blame]5140 declare void @llvm.var.annotation(i8* &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
Tanya Lattnercb1b9602007-06-15 20:50:54 +0000[diff] [blame]5141</pre>
5142
5143<h5>Overview:</h5>
5144
5145<p>
5146The '<tt>llvm.var.annotation</tt>' intrinsic
5147</p>
5148
5149<h5>Arguments:</h5>
5150
5151<p>
Tanya Lattnerbed1d4d2007-06-18 23:42:37 +0000[diff] [blame]5152The first argument is a pointer to a value, the second is a pointer to a
5153global string, the third is a pointer to a global string which is the source
5154file name, and the last argument is the line number.
Tanya Lattnercb1b9602007-06-15 20:50:54 +0000[diff] [blame]5155</p>
5156
5157<h5>Semantics:</h5>
5158
5159<p>
5160This intrinsic allows annotation of local variables with arbitrary strings.
5161This can be useful for special purpose optimizations that want to look for these
5162 annotations. These have no other defined use, they are ignored by code
5163 generation and optimization.
5164</div>
5165
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]5166<!-- _______________________________________________________________________ -->
5167<div class="doc_subsubsection">
Tanya Lattner0186a652007-09-21 23:57:59 +0000[diff] [blame]5168 <a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]5169</div>
5170
5171<div class="doc_text">
5172
5173<h5>Syntax:</h5>
Tanya Lattner23dbd572007-09-21 23:56:27 +0000[diff] [blame]5174<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
5175any integer bit width.
5176</p>
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]5177<pre>
Tanya Lattnercf3e26f2007-09-22 00:03:01 +0000[diff] [blame]5178 declare i8 @llvm.annotation.i8(i8 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5179 declare i16 @llvm.annotation.i16(i16 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5180 declare i32 @llvm.annotation.i32(i32 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5181 declare i64 @llvm.annotation.i64(i64 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5182 declare i256 @llvm.annotation.i256(i256 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]5183</pre>
5184
5185<h5>Overview:</h5>
Tanya Lattner23dbd572007-09-21 23:56:27 +0000[diff] [blame]5186
5187<p>
5188The '<tt>llvm.annotation</tt>' intrinsic.
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]5189</p>
5190
5191<h5>Arguments:</h5>
5192
5193<p>
5194The first argument is an integer value (result of some expression),
5195the second is a pointer to a global string, the third is a pointer to a global
5196string which is the source file name, and the last argument is the line number.
Tanya Lattner23dbd572007-09-21 23:56:27 +0000[diff] [blame]5197It returns the value of the first argument.
Tanya Lattner293c0372007-09-21 22:59:12 +0000[diff] [blame]5198</p>
5199
5200<h5>Semantics:</h5>
5201
5202<p>
5203This intrinsic allows annotations to be put on arbitrary expressions
5204with arbitrary strings. This can be useful for special purpose optimizations
5205that want to look for these annotations. These have no other defined use, they
5206are ignored by code generation and optimization.
5207</div>
Jim Laskey2211f492007-03-14 19:31:19 +0000[diff] [blame]5208
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]5209<!-- *********************************************************************** -->
Chris Lattner2f7c9632001-06-06 20:29:01 +0000[diff] [blame]5210<hr>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]5211<address>
5212 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
5213 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
5214 <a href="http://validator.w3.org/check/referer"><img
5215 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
5216
5217 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencerca058542006-03-14 05:39:39 +0000[diff] [blame]5218 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukmanc501f552004-03-01 17:47:27 +0000[diff] [blame]5219 Last modified: $Date$
5220</address>
Misha Brukman76307852003-11-08 01:05:38 +0000[diff] [blame]5221</body>
5222</html>