Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 72b84d0735cd0ab995da3dcac1472ed3e497f97c / . / docs / LangRef.html
blob: 56ee12654d509514d7b8d20c481d576496329631 [file] [log] [blame]
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
3<html>
4<head>
5 <title>LLVM Assembly Language Reference Manual</title>
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.">
10 <link rel="stylesheet" href="llvm.css" type="text/css">
11</head>
12
13<body>
14
15<div class="doc_title"> LLVM Language Reference Manual </div>
16<ol>
17 <li><a href="#abstract">Abstract</a></li>
18 <li><a href="#introduction">Introduction</a></li>
19 <li><a href="#identifiers">Identifiers</a></li>
20 <li><a href="#highlevel">High Level Structure</a>
21 <ol>
22 <li><a href="#modulestructure">Module Structure</a></li>
23 <li><a href="#linkage">Linkage Types</a></li>
24 <li><a href="#callingconv">Calling Conventions</a></li>
25 <li><a href="#globalvars">Global Variables</a></li>
26 <li><a href="#functionstructure">Functions</a></li>
27 <li><a href="#aliasstructure">Aliases</a>
28 <li><a href="#paramattrs">Parameter Attributes</a></li>
Gordon Henriksen13fe5e32007-12-10 03:18:06 +0000[diff] [blame]29 <li><a href="#gc">Garbage Collector Names</a></li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]30 <li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
31 <li><a href="#datalayout">Data Layout</a></li>
32 </ol>
33 </li>
34 <li><a href="#typesystem">Type System</a>
35 <ol>
36 <li><a href="#t_primitive">Primitive Types</a>
37 <ol>
38 <li><a href="#t_classifications">Type Classifications</a></li>
39 </ol>
40 </li>
41 <li><a href="#t_derived">Derived Types</a>
42 <ol>
Chris Lattner251ab812007-12-18 06:18:21 +0000[diff] [blame]43 <li><a href="#t_integer">Integer Type</a></li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]44 <li><a href="#t_array">Array Type</a></li>
45 <li><a href="#t_function">Function Type</a></li>
46 <li><a href="#t_pointer">Pointer Type</a></li>
47 <li><a href="#t_struct">Structure Type</a></li>
48 <li><a href="#t_pstruct">Packed Structure Type</a></li>
49 <li><a href="#t_vector">Vector Type</a></li>
50 <li><a href="#t_opaque">Opaque Type</a></li>
51 </ol>
52 </li>
53 </ol>
54 </li>
55 <li><a href="#constants">Constants</a>
56 <ol>
57 <li><a href="#simpleconstants">Simple Constants</a>
58 <li><a href="#aggregateconstants">Aggregate Constants</a>
59 <li><a href="#globalconstants">Global Variable and Function Addresses</a>
60 <li><a href="#undefvalues">Undefined Values</a>
61 <li><a href="#constantexprs">Constant Expressions</a>
62 </ol>
63 </li>
64 <li><a href="#othervalues">Other Values</a>
65 <ol>
66 <li><a href="#inlineasm">Inline Assembler Expressions</a>
67 </ol>
68 </li>
69 <li><a href="#instref">Instruction Reference</a>
70 <ol>
71 <li><a href="#terminators">Terminator Instructions</a>
72 <ol>
73 <li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
74 <li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
75 <li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
76 <li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
77 <li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
78 <li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
79 </ol>
80 </li>
81 <li><a href="#binaryops">Binary Operations</a>
82 <ol>
83 <li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
84 <li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
85 <li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
86 <li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
87 <li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
88 <li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
89 <li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
90 <li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
91 <li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
92 </ol>
93 </li>
94 <li><a href="#bitwiseops">Bitwise Binary Operations</a>
95 <ol>
96 <li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
97 <li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
98 <li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
99 <li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
100 <li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
101 <li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
102 </ol>
103 </li>
104 <li><a href="#vectorops">Vector Operations</a>
105 <ol>
106 <li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
107 <li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
108 <li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
109 </ol>
110 </li>
111 <li><a href="#memoryops">Memory Access and Addressing Operations</a>
112 <ol>
113 <li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
114 <li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
115 <li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
116 <li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
117 <li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
118 <li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
119 </ol>
120 </li>
121 <li><a href="#convertops">Conversion Operations</a>
122 <ol>
123 <li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
124 <li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
125 <li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
126 <li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
127 <li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
128 <li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
129 <li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
130 <li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
131 <li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
132 <li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
133 <li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
134 <li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
135 </ol>
136 <li><a href="#otherops">Other Operations</a>
137 <ol>
138 <li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
139 <li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
140 <li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
141 <li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
142 <li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
143 <li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
144 </ol>
145 </li>
146 </ol>
147 </li>
148 <li><a href="#intrinsics">Intrinsic Functions</a>
149 <ol>
150 <li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
151 <ol>
152 <li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
153 <li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
154 <li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
155 </ol>
156 </li>
157 <li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
158 <ol>
159 <li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
160 <li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
161 <li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
162 </ol>
163 </li>
164 <li><a href="#int_codegen">Code Generator Intrinsics</a>
165 <ol>
166 <li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
167 <li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
168 <li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
169 <li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
170 <li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
171 <li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
172 <li><a href="#int_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
173 </ol>
174 </li>
175 <li><a href="#int_libc">Standard C Library Intrinsics</a>
176 <ol>
177 <li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
178 <li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
179 <li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
180 <li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
181 <li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]182 <li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
183 <li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
184 <li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]185 </ol>
186 </li>
187 <li><a href="#int_manip">Bit Manipulation Intrinsics</a>
188 <ol>
189 <li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
190 <li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
191 <li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
192 <li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
193 <li><a href="#int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic </a></li>
194 <li><a href="#int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic </a></li>
195 </ol>
196 </li>
197 <li><a href="#int_debugger">Debugger intrinsics</a></li>
198 <li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]199 <li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]200 <ol>
201 <li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]202 </ol>
203 </li>
Reid Spencerb043f672007-07-20 19:59:11 +0000[diff] [blame]204 <li><a href="#int_general">General intrinsics</a>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]205 <ol>
Reid Spencerb043f672007-07-20 19:59:11 +0000[diff] [blame]206 <li><a href="#int_var_annotation">
Tanya Lattner51369f32007-09-22 00:01:26 +0000[diff] [blame]207 <tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Reid Spencerb043f672007-07-20 19:59:11 +0000[diff] [blame]208 </ol>
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]209 <ol>
210 <li><a href="#int_annotation">
Tanya Lattner51369f32007-09-22 00:01:26 +0000[diff] [blame]211 <tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]212 </ol>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]213 </li>
214 </ol>
215 </li>
216</ol>
217
218<div class="doc_author">
219 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
220 and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
221</div>
222
223<!-- *********************************************************************** -->
224<div class="doc_section"> <a name="abstract">Abstract </a></div>
225<!-- *********************************************************************** -->
226
227<div class="doc_text">
228<p>This document is a reference manual for the LLVM assembly language.
229LLVM is an SSA based representation that provides type safety,
230low-level operations, flexibility, and the capability of representing
231'all' high-level languages cleanly. It is the common code
232representation used throughout all phases of the LLVM compilation
233strategy.</p>
234</div>
235
236<!-- *********************************************************************** -->
237<div class="doc_section"> <a name="introduction">Introduction</a> </div>
238<!-- *********************************************************************** -->
239
240<div class="doc_text">
241
242<p>The LLVM code representation is designed to be used in three
243different forms: as an in-memory compiler IR, as an on-disk bitcode
244representation (suitable for fast loading by a Just-In-Time compiler),
245and as a human readable assembly language representation. This allows
246LLVM to provide a powerful intermediate representation for efficient
247compiler transformations and analysis, while providing a natural means
248to debug and visualize the transformations. The three different forms
249of LLVM are all equivalent. This document describes the human readable
250representation and notation.</p>
251
252<p>The LLVM representation aims to be light-weight and low-level
253while being expressive, typed, and extensible at the same time. It
254aims to be a "universal IR" of sorts, by being at a low enough level
255that high-level ideas may be cleanly mapped to it (similar to how
256microprocessors are "universal IR's", allowing many source languages to
257be mapped to them). By providing type information, LLVM can be used as
258the target of optimizations: for example, through pointer analysis, it
259can be proven that a C automatic variable is never accessed outside of
260the current function... allowing it to be promoted to a simple SSA
261value instead of a memory location.</p>
262
263</div>
264
265<!-- _______________________________________________________________________ -->
266<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
267
268<div class="doc_text">
269
270<p>It is important to note that this document describes 'well formed'
271LLVM assembly language. There is a difference between what the parser
272accepts and what is considered 'well formed'. For example, the
273following instruction is syntactically okay, but not well formed:</p>
274
275<div class="doc_code">
276<pre>
277%x = <a href="#i_add">add</a> i32 1, %x
278</pre>
279</div>
280
281<p>...because the definition of <tt>%x</tt> does not dominate all of
282its uses. The LLVM infrastructure provides a verification pass that may
283be used to verify that an LLVM module is well formed. This pass is
284automatically run by the parser after parsing input assembly and by
285the optimizer before it outputs bitcode. The violations pointed out
286by the verifier pass indicate bugs in transformation passes or input to
287the parser.</p>
288</div>
289
Chris Lattnera83fdc02007-10-03 17:34:29 +0000[diff] [blame]290<!-- Describe the typesetting conventions here. -->
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]291
292<!-- *********************************************************************** -->
293<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
294<!-- *********************************************************************** -->
295
296<div class="doc_text">
297
Reid Spencerc8245b02007-08-07 14:34:28 +0000[diff] [blame]298 <p>LLVM identifiers come in two basic types: global and local. Global
299 identifiers (functions, global variables) begin with the @ character. Local
300 identifiers (register names, types) begin with the % character. Additionally,
301 there are three different formats for identifiers, for different purposes:
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]302
303<ol>
Reid Spencerc8245b02007-08-07 14:34:28 +0000[diff] [blame]304 <li>Named values are represented as a string of characters with their prefix.
305 For example, %foo, @DivisionByZero, %a.really.long.identifier. The actual
306 regular expression used is '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]307 Identifiers which require other characters in their names can be surrounded
Reid Spencerc8245b02007-08-07 14:34:28 +0000[diff] [blame]308 with quotes. In this way, anything except a <tt>&quot;</tt> character can
309 be used in a named value.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]310
Reid Spencerc8245b02007-08-07 14:34:28 +0000[diff] [blame]311 <li>Unnamed values are represented as an unsigned numeric value with their
312 prefix. For example, %12, @2, %44.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]313
314 <li>Constants, which are described in a <a href="#constants">section about
315 constants</a>, below.</li>
316</ol>
317
Reid Spencerc8245b02007-08-07 14:34:28 +0000[diff] [blame]318<p>LLVM requires that values start with a prefix for two reasons: Compilers
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]319don't need to worry about name clashes with reserved words, and the set of
320reserved words may be expanded in the future without penalty. Additionally,
321unnamed identifiers allow a compiler to quickly come up with a temporary
322variable without having to avoid symbol table conflicts.</p>
323
324<p>Reserved words in LLVM are very similar to reserved words in other
325languages. There are keywords for different opcodes
326('<tt><a href="#i_add">add</a></tt>',
327 '<tt><a href="#i_bitcast">bitcast</a></tt>',
328 '<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
329href="#t_void">void</a></tt>', '<tt><a href="#t_primitive">i32</a></tt>', etc...),
330and others. These reserved words cannot conflict with variable names, because
Reid Spencerc8245b02007-08-07 14:34:28 +0000[diff] [blame]331none of them start with a prefix character ('%' or '@').</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]332
333<p>Here is an example of LLVM code to multiply the integer variable
334'<tt>%X</tt>' by 8:</p>
335
336<p>The easy way:</p>
337
338<div class="doc_code">
339<pre>
340%result = <a href="#i_mul">mul</a> i32 %X, 8
341</pre>
342</div>
343
344<p>After strength reduction:</p>
345
346<div class="doc_code">
347<pre>
348%result = <a href="#i_shl">shl</a> i32 %X, i8 3
349</pre>
350</div>
351
352<p>And the hard way:</p>
353
354<div class="doc_code">
355<pre>
356<a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
357<a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
358%result = <a href="#i_add">add</a> i32 %1, %1
359</pre>
360</div>
361
362<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
363important lexical features of LLVM:</p>
364
365<ol>
366
367 <li>Comments are delimited with a '<tt>;</tt>' and go until the end of
368 line.</li>
369
370 <li>Unnamed temporaries are created when the result of a computation is not
371 assigned to a named value.</li>
372
373 <li>Unnamed temporaries are numbered sequentially</li>
374
375</ol>
376
377<p>...and it also shows a convention that we follow in this document. When
378demonstrating instructions, we will follow an instruction with a comment that
379defines the type and name of value produced. Comments are shown in italic
380text.</p>
381
382</div>
383
384<!-- *********************************************************************** -->
385<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
386<!-- *********************************************************************** -->
387
388<!-- ======================================================================= -->
389<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
390</div>
391
392<div class="doc_text">
393
394<p>LLVM programs are composed of "Module"s, each of which is a
395translation unit of the input programs. Each module consists of
396functions, global variables, and symbol table entries. Modules may be
397combined together with the LLVM linker, which merges function (and
398global variable) definitions, resolves forward declarations, and merges
399symbol table entries. Here is an example of the "hello world" module:</p>
400
401<div class="doc_code">
402<pre><i>; Declare the string constant as a global constant...</i>
403<a href="#identifiers">@.LC0</a> = <a href="#linkage_internal">internal</a> <a
404 href="#globalvars">constant</a> <a href="#t_array">[13 x i8]</a> c"hello world\0A\00" <i>; [13 x i8]*</i>
405
406<i>; External declaration of the puts function</i>
407<a href="#functionstructure">declare</a> i32 @puts(i8 *) <i>; i32(i8 *)* </i>
408
409<i>; Definition of main function</i>
410define i32 @main() { <i>; i32()* </i>
411 <i>; Convert [13x i8 ]* to i8 *...</i>
412 %cast210 = <a
413 href="#i_getelementptr">getelementptr</a> [13 x i8 ]* @.LC0, i64 0, i64 0 <i>; i8 *</i>
414
415 <i>; Call puts function to write out the string to stdout...</i>
416 <a
417 href="#i_call">call</a> i32 @puts(i8 * %cast210) <i>; i32</i>
418 <a
419 href="#i_ret">ret</a> i32 0<br>}<br>
420</pre>
421</div>
422
423<p>This example is made up of a <a href="#globalvars">global variable</a>
424named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
425function, and a <a href="#functionstructure">function definition</a>
426for "<tt>main</tt>".</p>
427
428<p>In general, a module is made up of a list of global values,
429where both functions and global variables are global values. Global values are
430represented by a pointer to a memory location (in this case, a pointer to an
431array of char, and a pointer to a function), and have one of the following <a
432href="#linkage">linkage types</a>.</p>
433
434</div>
435
436<!-- ======================================================================= -->
437<div class="doc_subsection">
438 <a name="linkage">Linkage Types</a>
439</div>
440
441<div class="doc_text">
442
443<p>
444All Global Variables and Functions have one of the following types of linkage:
445</p>
446
447<dl>
448
449 <dt><tt><b><a name="linkage_internal">internal</a></b></tt> </dt>
450
451 <dd>Global values with internal linkage are only directly accessible by
452 objects in the current module. In particular, linking code into a module with
453 an internal global value may cause the internal to be renamed as necessary to
454 avoid collisions. Because the symbol is internal to the module, all
455 references can be updated. This corresponds to the notion of the
456 '<tt>static</tt>' keyword in C.
457 </dd>
458
459 <dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
460
461 <dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
462 the same name when linkage occurs. This is typically used to implement
463 inline functions, templates, or other code which must be generated in each
464 translation unit that uses it. Unreferenced <tt>linkonce</tt> globals are
465 allowed to be discarded.
466 </dd>
467
468 <dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
469
470 <dd>"<tt>weak</tt>" linkage is exactly the same as <tt>linkonce</tt> linkage,
471 except that unreferenced <tt>weak</tt> globals may not be discarded. This is
472 used for globals that may be emitted in multiple translation units, but that
473 are not guaranteed to be emitted into every translation unit that uses them.
474 One example of this are common globals in C, such as "<tt>int X;</tt>" at
475 global scope.
476 </dd>
477
478 <dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
479
480 <dd>"<tt>appending</tt>" linkage may only be applied to global variables of
481 pointer to array type. When two global variables with appending linkage are
482 linked together, the two global arrays are appended together. This is the
483 LLVM, typesafe, equivalent of having the system linker append together
484 "sections" with identical names when .o files are linked.
485 </dd>
486
487 <dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt>: </dt>
488 <dd>The semantics of this linkage follow the ELF model: the symbol is weak
489 until linked, if not linked, the symbol becomes null instead of being an
490 undefined reference.
491 </dd>
492
493 <dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
494
495 <dd>If none of the above identifiers are used, the global is externally
496 visible, meaning that it participates in linkage and can be used to resolve
497 external symbol references.
498 </dd>
499</dl>
500
501 <p>
502 The next two types of linkage are targeted for Microsoft Windows platform
503 only. They are designed to support importing (exporting) symbols from (to)
504 DLLs.
505 </p>
506
507 <dl>
508 <dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt>: </dt>
509
510 <dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
511 or variable via a global pointer to a pointer that is set up by the DLL
512 exporting the symbol. On Microsoft Windows targets, the pointer name is
513 formed by combining <code>_imp__</code> and the function or variable name.
514 </dd>
515
516 <dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt>: </dt>
517
518 <dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
519 pointer to a pointer in a DLL, so that it can be referenced with the
520 <tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
521 name is formed by combining <code>_imp__</code> and the function or variable
522 name.
523 </dd>
524
525</dl>
526
527<p><a name="linkage_external"></a>For example, since the "<tt>.LC0</tt>"
528variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
529variable and was linked with this one, one of the two would be renamed,
530preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
531external (i.e., lacking any linkage declarations), they are accessible
532outside of the current module.</p>
533<p>It is illegal for a function <i>declaration</i>
534to have any linkage type other than "externally visible", <tt>dllimport</tt>,
535or <tt>extern_weak</tt>.</p>
536<p>Aliases can have only <tt>external</tt>, <tt>internal</tt> and <tt>weak</tt>
537linkages.
538</div>
539
540<!-- ======================================================================= -->
541<div class="doc_subsection">
542 <a name="callingconv">Calling Conventions</a>
543</div>
544
545<div class="doc_text">
546
547<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
548and <a href="#i_invoke">invokes</a> can all have an optional calling convention
549specified for the call. The calling convention of any pair of dynamic
550caller/callee must match, or the behavior of the program is undefined. The
551following calling conventions are supported by LLVM, and more may be added in
552the future:</p>
553
554<dl>
555 <dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
556
557 <dd>This calling convention (the default if no other calling convention is
558 specified) matches the target C calling conventions. This calling convention
559 supports varargs function calls and tolerates some mismatch in the declared
560 prototype and implemented declaration of the function (as does normal C).
561 </dd>
562
563 <dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
564
565 <dd>This calling convention attempts to make calls as fast as possible
566 (e.g. by passing things in registers). This calling convention allows the
567 target to use whatever tricks it wants to produce fast code for the target,
568 without having to conform to an externally specified ABI. Implementations of
569 this convention should allow arbitrary tail call optimization to be supported.
570 This calling convention does not support varargs and requires the prototype of
571 all callees to exactly match the prototype of the function definition.
572 </dd>
573
574 <dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
575
576 <dd>This calling convention attempts to make code in the caller as efficient
577 as possible under the assumption that the call is not commonly executed. As
578 such, these calls often preserve all registers so that the call does not break
579 any live ranges in the caller side. This calling convention does not support
580 varargs and requires the prototype of all callees to exactly match the
581 prototype of the function definition.
582 </dd>
583
584 <dt><b>"<tt>cc &lt;<em>n</em>&gt;</tt>" - Numbered convention</b>:</dt>
585
586 <dd>Any calling convention may be specified by number, allowing
587 target-specific calling conventions to be used. Target specific calling
588 conventions start at 64.
589 </dd>
590</dl>
591
592<p>More calling conventions can be added/defined on an as-needed basis, to
593support pascal conventions or any other well-known target-independent
594convention.</p>
595
596</div>
597
598<!-- ======================================================================= -->
599<div class="doc_subsection">
600 <a name="visibility">Visibility Styles</a>
601</div>
602
603<div class="doc_text">
604
605<p>
606All Global Variables and Functions have one of the following visibility styles:
607</p>
608
609<dl>
610 <dt><b>"<tt>default</tt>" - Default style</b>:</dt>
611
612 <dd>On ELF, default visibility means that the declaration is visible to other
613 modules and, in shared libraries, means that the declared entity may be
614 overridden. On Darwin, default visibility means that the declaration is
615 visible to other modules. Default visibility corresponds to "external
616 linkage" in the language.
617 </dd>
618
619 <dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
620
621 <dd>Two declarations of an object with hidden visibility refer to the same
622 object if they are in the same shared object. Usually, hidden visibility
623 indicates that the symbol will not be placed into the dynamic symbol table,
624 so no other module (executable or shared library) can reference it
625 directly.
626 </dd>
627
628 <dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
629
630 <dd>On ELF, protected visibility indicates that the symbol will be placed in
631 the dynamic symbol table, but that references within the defining module will
632 bind to the local symbol. That is, the symbol cannot be overridden by another
633 module.
634 </dd>
635</dl>
636
637</div>
638
639<!-- ======================================================================= -->
640<div class="doc_subsection">
641 <a name="globalvars">Global Variables</a>
642</div>
643
644<div class="doc_text">
645
646<p>Global variables define regions of memory allocated at compilation time
647instead of run-time. Global variables may optionally be initialized, may have
648an explicit section to be placed in, and may have an optional explicit alignment
649specified. A variable may be defined as "thread_local", which means that it
650will not be shared by threads (each thread will have a separated copy of the
651variable). A variable may be defined as a global "constant," which indicates
652that the contents of the variable will <b>never</b> be modified (enabling better
653optimization, allowing the global data to be placed in the read-only section of
654an executable, etc). Note that variables that need runtime initialization
655cannot be marked "constant" as there is a store to the variable.</p>
656
657<p>
658LLVM explicitly allows <em>declarations</em> of global variables to be marked
659constant, even if the final definition of the global is not. This capability
660can be used to enable slightly better optimization of the program, but requires
661the language definition to guarantee that optimizations based on the
662'constantness' are valid for the translation units that do not include the
663definition.
664</p>
665
666<p>As SSA values, global variables define pointer values that are in
667scope (i.e. they dominate) all basic blocks in the program. Global
668variables always define a pointer to their "content" type because they
669describe a region of memory, and all memory objects in LLVM are
670accessed through pointers.</p>
671
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]672<p>A global variable may be declared to reside in a target-specifc numbered
673address space. For targets that support them, address spaces may affect how
674optimizations are performed and/or what target instructions are used to access
Christopher Lamb20a39e92007-12-12 08:44:39 +0000[diff] [blame]675the variable. The default address space is zero. The address space qualifier
676must precede any other attributes.</p>
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]677
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]678<p>LLVM allows an explicit section to be specified for globals. If the target
679supports it, it will emit globals to the section specified.</p>
680
681<p>An explicit alignment may be specified for a global. If not present, or if
682the alignment is set to zero, the alignment of the global is set by the target
683to whatever it feels convenient. If an explicit alignment is specified, the
684global is forced to have at least that much alignment. All alignments must be
685a power of 2.</p>
686
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]687<p>For example, the following defines a global in a numbered address space with
688an initializer, section, and alignment:</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]689
690<div class="doc_code">
691<pre>
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]692@G = constant float 1.0 addrspace(5), section "foo", align 4
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]693</pre>
694</div>
695
696</div>
697
698
699<!-- ======================================================================= -->
700<div class="doc_subsection">
701 <a name="functionstructure">Functions</a>
702</div>
703
704<div class="doc_text">
705
706<p>LLVM function definitions consist of the "<tt>define</tt>" keyord,
707an optional <a href="#linkage">linkage type</a>, an optional
708<a href="#visibility">visibility style</a>, an optional
709<a href="#callingconv">calling convention</a>, a return type, an optional
710<a href="#paramattrs">parameter attribute</a> for the return type, a function
711name, a (possibly empty) argument list (each with optional
712<a href="#paramattrs">parameter attributes</a>), an optional section, an
Gordon Henriksend606f5b2007-12-10 03:30:21 +0000[diff] [blame]713optional alignment, an optional <a href="#gc">garbage collector name</a>, an
Gordon Henriksen13fe5e32007-12-10 03:18:06 +0000[diff] [blame]714opening curly brace, a list of basic blocks, and a closing curly brace.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]715
716LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
717optional <a href="#linkage">linkage type</a>, an optional
718<a href="#visibility">visibility style</a>, an optional
719<a href="#callingconv">calling convention</a>, a return type, an optional
720<a href="#paramattrs">parameter attribute</a> for the return type, a function
Gordon Henriksen13fe5e32007-12-10 03:18:06 +0000[diff] [blame]721name, a possibly empty list of arguments, an optional alignment, and an optional
Gordon Henriksend606f5b2007-12-10 03:30:21 +0000[diff] [blame]722<a href="#gc">garbage collector name</a>.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]723
724<p>A function definition contains a list of basic blocks, forming the CFG for
725the function. Each basic block may optionally start with a label (giving the
726basic block a symbol table entry), contains a list of instructions, and ends
727with a <a href="#terminators">terminator</a> instruction (such as a branch or
728function return).</p>
729
730<p>The first basic block in a function is special in two ways: it is immediately
731executed on entrance to the function, and it is not allowed to have predecessor
732basic blocks (i.e. there can not be any branches to the entry block of a
733function). Because the block can have no predecessors, it also cannot have any
734<a href="#i_phi">PHI nodes</a>.</p>
735
736<p>LLVM allows an explicit section to be specified for functions. If the target
737supports it, it will emit functions to the section specified.</p>
738
739<p>An explicit alignment may be specified for a function. If not present, or if
740the alignment is set to zero, the alignment of the function is set by the target
741to whatever it feels convenient. If an explicit alignment is specified, the
742function is forced to have at least that much alignment. All alignments must be
743a power of 2.</p>
744
745</div>
746
747
748<!-- ======================================================================= -->
749<div class="doc_subsection">
750 <a name="aliasstructure">Aliases</a>
751</div>
752<div class="doc_text">
753 <p>Aliases act as "second name" for the aliasee value (which can be either
754 function or global variable or bitcast of global value). Aliases may have an
755 optional <a href="#linkage">linkage type</a>, and an
756 optional <a href="#visibility">visibility style</a>.</p>
757
758 <h5>Syntax:</h5>
759
760<div class="doc_code">
761<pre>
762@&lt;Name&gt; = [Linkage] [Visibility] alias &lt;AliaseeTy&gt; @&lt;Aliasee&gt;
763</pre>
764</div>
765
766</div>
767
768
769
770<!-- ======================================================================= -->
771<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
772<div class="doc_text">
773 <p>The return type and each parameter of a function type may have a set of
774 <i>parameter attributes</i> associated with them. Parameter attributes are
775 used to communicate additional information about the result or parameters of
Duncan Sandsf5588dc2007-11-27 13:23:08 +0000[diff] [blame]776 a function. Parameter attributes are considered to be part of the function,
777 not of the function type, so functions with different parameter attributes
778 can have the same function type.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]779
780 <p>Parameter attributes are simple keywords that follow the type specified. If
781 multiple parameter attributes are needed, they are space separated. For
782 example:</p>
783
784<div class="doc_code">
785<pre>
Duncan Sandsf5588dc2007-11-27 13:23:08 +0000[diff] [blame]786declare i32 @printf(i8* noalias , ...) nounwind
787declare i32 @atoi(i8*) nounwind readonly
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]788</pre>
789</div>
790
Duncan Sandsf5588dc2007-11-27 13:23:08 +0000[diff] [blame]791 <p>Note that any attributes for the function result (<tt>nounwind</tt>,
792 <tt>readonly</tt>) come immediately after the argument list.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]793
794 <p>Currently, only the following parameter attributes are defined:</p>
795 <dl>
Reid Spencerf234bed2007-07-19 23:13:04 +0000[diff] [blame]796 <dt><tt>zeroext</tt></dt>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]797 <dd>This indicates that the parameter should be zero extended just before
798 a call to this function.</dd>
Reid Spencerf234bed2007-07-19 23:13:04 +0000[diff] [blame]799 <dt><tt>signext</tt></dt>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]800 <dd>This indicates that the parameter should be sign extended just before
801 a call to this function.</dd>
802 <dt><tt>inreg</tt></dt>
803 <dd>This indicates that the parameter should be placed in register (if
804 possible) during assembling function call. Support for this attribute is
805 target-specific</dd>
806 <dt><tt>sret</tt></dt>
807 <dd>This indicates that the parameter specifies the address of a structure
808 that is the return value of the function in the source program.</dd>
809 <dt><tt>noalias</tt></dt>
810 <dd>This indicates that the parameter not alias any other object or any
811 other "noalias" objects during the function call.
812 <dt><tt>noreturn</tt></dt>
813 <dd>This function attribute indicates that the function never returns. This
814 indicates to LLVM that every call to this function should be treated as if
815 an <tt>unreachable</tt> instruction immediately followed the call.</dd>
816 <dt><tt>nounwind</tt></dt>
817 <dd>This function attribute indicates that the function type does not use
818 the unwind instruction and does not allow stack unwinding to propagate
819 through it.</dd>
Duncan Sands4ee46812007-07-27 19:57:41 +0000[diff] [blame]820 <dt><tt>nest</tt></dt>
821 <dd>This indicates that the parameter can be excised using the
822 <a href="#int_trampoline">trampoline intrinsics</a>.</dd>
Duncan Sands13e13f82007-11-22 20:23:04 +0000[diff] [blame]823 <dt><tt>readonly</tt></dt>
Duncan Sandsd69c0e62007-11-14 21:14:02 +0000[diff] [blame]824 <dd>This function attribute indicates that the function has no side-effects
Duncan Sands13e13f82007-11-22 20:23:04 +0000[diff] [blame]825 except for producing a return value or throwing an exception. The value
826 returned must only depend on the function arguments and/or global variables.
827 It may use values obtained by dereferencing pointers.</dd>
828 <dt><tt>readnone</tt></dt>
829 <dd>A <tt>readnone</tt> function has the same restrictions as a <tt>readonly</tt>
Duncan Sandsd69c0e62007-11-14 21:14:02 +0000[diff] [blame]830 function, but in addition it is not allowed to dereference any pointer arguments
831 or global variables.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]832 </dl>
833
834</div>
835
836<!-- ======================================================================= -->
837<div class="doc_subsection">
Gordon Henriksen13fe5e32007-12-10 03:18:06 +0000[diff] [blame]838 <a name="gc">Garbage Collector Names</a>
839</div>
840
841<div class="doc_text">
842<p>Each function may specify a garbage collector name, which is simply a
843string.</p>
844
845<div class="doc_code"><pre
846>define void @f() gc "name" { ...</pre></div>
847
848<p>The compiler declares the supported values of <i>name</i>. Specifying a
849collector which will cause the compiler to alter its output in order to support
850the named garbage collection algorithm.</p>
851</div>
852
853<!-- ======================================================================= -->
854<div class="doc_subsection">
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]855 <a name="moduleasm">Module-Level Inline Assembly</a>
856</div>
857
858<div class="doc_text">
859<p>
860Modules may contain "module-level inline asm" blocks, which corresponds to the
861GCC "file scope inline asm" blocks. These blocks are internally concatenated by
862LLVM and treated as a single unit, but may be separated in the .ll file if
863desired. The syntax is very simple:
864</p>
865
866<div class="doc_code">
867<pre>
868module asm "inline asm code goes here"
869module asm "more can go here"
870</pre>
871</div>
872
873<p>The strings can contain any character by escaping non-printable characters.
874 The escape sequence used is simply "\xx" where "xx" is the two digit hex code
875 for the number.
876</p>
877
878<p>
879 The inline asm code is simply printed to the machine code .s file when
880 assembly code is generated.
881</p>
882</div>
883
884<!-- ======================================================================= -->
885<div class="doc_subsection">
886 <a name="datalayout">Data Layout</a>
887</div>
888
889<div class="doc_text">
890<p>A module may specify a target specific data layout string that specifies how
891data is to be laid out in memory. The syntax for the data layout is simply:</p>
892<pre> target datalayout = "<i>layout specification</i>"</pre>
893<p>The <i>layout specification</i> consists of a list of specifications
894separated by the minus sign character ('-'). Each specification starts with a
895letter and may include other information after the letter to define some
896aspect of the data layout. The specifications accepted are as follows: </p>
897<dl>
898 <dt><tt>E</tt></dt>
899 <dd>Specifies that the target lays out data in big-endian form. That is, the
900 bits with the most significance have the lowest address location.</dd>
901 <dt><tt>e</tt></dt>
902 <dd>Specifies that hte target lays out data in little-endian form. That is,
903 the bits with the least significance have the lowest address location.</dd>
904 <dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
905 <dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
906 <i>preferred</i> alignments. All sizes are in bits. Specifying the <i>pref</i>
907 alignment is optional. If omitted, the preceding <tt>:</tt> should be omitted
908 too.</dd>
909 <dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
910 <dd>This specifies the alignment for an integer type of a given bit
911 <i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
912 <dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
913 <dd>This specifies the alignment for a vector type of a given bit
914 <i>size</i>.</dd>
915 <dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
916 <dd>This specifies the alignment for a floating point type of a given bit
917 <i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
918 (double).</dd>
919 <dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
920 <dd>This specifies the alignment for an aggregate type of a given bit
921 <i>size</i>.</dd>
922</dl>
923<p>When constructing the data layout for a given target, LLVM starts with a
924default set of specifications which are then (possibly) overriden by the
925specifications in the <tt>datalayout</tt> keyword. The default specifications
926are given in this list:</p>
927<ul>
928 <li><tt>E</tt> - big endian</li>
929 <li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
930 <li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
931 <li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
932 <li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
933 <li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
934 <li><tt>i64:32:64</tt> - i64 has abi alignment of 32-bits but preferred
935 alignment of 64-bits</li>
936 <li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
937 <li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
938 <li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
939 <li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
940 <li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
941</ul>
942<p>When llvm is determining the alignment for a given type, it uses the
943following rules:
944<ol>
945 <li>If the type sought is an exact match for one of the specifications, that
946 specification is used.</li>
947 <li>If no match is found, and the type sought is an integer type, then the
948 smallest integer type that is larger than the bitwidth of the sought type is
949 used. If none of the specifications are larger than the bitwidth then the the
950 largest integer type is used. For example, given the default specifications
951 above, the i7 type will use the alignment of i8 (next largest) while both
952 i65 and i256 will use the alignment of i64 (largest specified).</li>
953 <li>If no match is found, and the type sought is a vector type, then the
954 largest vector type that is smaller than the sought vector type will be used
955 as a fall back. This happens because <128 x double> can be implemented in
956 terms of 64 <2 x double>, for example.</li>
957</ol>
958</div>
959
960<!-- *********************************************************************** -->
961<div class="doc_section"> <a name="typesystem">Type System</a> </div>
962<!-- *********************************************************************** -->
963
964<div class="doc_text">
965
966<p>The LLVM type system is one of the most important features of the
967intermediate representation. Being typed enables a number of
968optimizations to be performed on the IR directly, without having to do
969extra analyses on the side before the transformation. A strong type
970system makes it easier to read the generated code and enables novel
971analyses and transformations that are not feasible to perform on normal
972three address code representations.</p>
973
974</div>
975
976<!-- ======================================================================= -->
977<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
978<div class="doc_text">
979<p>The primitive types are the fundamental building blocks of the LLVM
980system. The current set of primitive types is as follows:</p>
981
982<table class="layout">
983 <tr class="layout">
984 <td class="left">
985 <table>
986 <tbody>
987 <tr><th>Type</th><th>Description</th></tr>
988 <tr><td><tt><a name="t_void">void</a></tt></td><td>No value</td></tr>
989 <tr><td><tt>label</tt></td><td>Branch destination</td></tr>
990 </tbody>
991 </table>
992 </td>
993 <td class="right">
994 <table>
995 <tbody>
996 <tr><th>Type</th><th>Description</th></tr>
997 <tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
998 <tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
999 </tbody>
1000 </table>
1001 </td>
1002 </tr>
1003</table>
1004</div>
1005
1006<!-- _______________________________________________________________________ -->
1007<div class="doc_subsubsection"> <a name="t_classifications">Type
1008Classifications</a> </div>
1009<div class="doc_text">
1010<p>These different primitive types fall into a few useful
1011classifications:</p>
1012
1013<table border="1" cellspacing="0" cellpadding="4">
1014 <tbody>
1015 <tr><th>Classification</th><th>Types</th></tr>
1016 <tr>
1017 <td><a name="t_integer">integer</a></td>
1018 <td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
1019 </tr>
1020 <tr>
1021 <td><a name="t_floating">floating point</a></td>
1022 <td><tt>float, double</tt></td>
1023 </tr>
1024 <tr>
1025 <td><a name="t_firstclass">first class</a></td>
1026 <td><tt>i1, ..., float, double, <br/>
1027 <a href="#t_pointer">pointer</a>,<a href="#t_vector">vector</a></tt>
1028 </td>
1029 </tr>
1030 </tbody>
1031</table>
1032
1033<p>The <a href="#t_firstclass">first class</a> types are perhaps the
1034most important. Values of these types are the only ones which can be
1035produced by instructions, passed as arguments, or used as operands to
1036instructions. This means that all structures and arrays must be
1037manipulated either by pointer or by component.</p>
1038</div>
1039
1040<!-- ======================================================================= -->
1041<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
1042
1043<div class="doc_text">
1044
1045<p>The real power in LLVM comes from the derived types in the system.
1046This is what allows a programmer to represent arrays, functions,
1047pointers, and other useful types. Note that these derived types may be
1048recursive: For example, it is possible to have a two dimensional array.</p>
1049
1050</div>
1051
1052<!-- _______________________________________________________________________ -->
1053<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
1054
1055<div class="doc_text">
1056
1057<h5>Overview:</h5>
1058<p>The integer type is a very simple derived type that simply specifies an
1059arbitrary bit width for the integer type desired. Any bit width from 1 bit to
10602^23-1 (about 8 million) can be specified.</p>
1061
1062<h5>Syntax:</h5>
1063
1064<pre>
1065 iN
1066</pre>
1067
1068<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
1069value.</p>
1070
1071<h5>Examples:</h5>
1072<table class="layout">
Chris Lattner251ab812007-12-18 06:18:21 +0000[diff] [blame]1073 <tbody>
1074 <tr>
1075 <td><tt>i1</tt></td>
1076 <td>a single-bit integer.</td>
1077 </tr><tr>
1078 <td><tt>i32</tt></td>
1079 <td>a 32-bit integer.</td>
1080 </tr><tr>
1081 <td><tt>i1942652</tt></td>
1082 <td>a really big integer of over 1 million bits.</td>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1083 </tr>
Chris Lattner251ab812007-12-18 06:18:21 +0000[diff] [blame]1084 </tbody>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1085</table>
1086</div>
1087
1088<!-- _______________________________________________________________________ -->
1089<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
1090
1091<div class="doc_text">
1092
1093<h5>Overview:</h5>
1094
1095<p>The array type is a very simple derived type that arranges elements
1096sequentially in memory. The array type requires a size (number of
1097elements) and an underlying data type.</p>
1098
1099<h5>Syntax:</h5>
1100
1101<pre>
1102 [&lt;# elements&gt; x &lt;elementtype&gt;]
1103</pre>
1104
1105<p>The number of elements is a constant integer value; elementtype may
1106be any type with a size.</p>
1107
1108<h5>Examples:</h5>
1109<table class="layout">
1110 <tr class="layout">
1111 <td class="left">
1112 <tt>[40 x i32 ]</tt><br/>
1113 <tt>[41 x i32 ]</tt><br/>
1114 <tt>[40 x i8]</tt><br/>
1115 </td>
1116 <td class="left">
1117 Array of 40 32-bit integer values.<br/>
1118 Array of 41 32-bit integer values.<br/>
1119 Array of 40 8-bit integer values.<br/>
1120 </td>
1121 </tr>
1122</table>
1123<p>Here are some examples of multidimensional arrays:</p>
1124<table class="layout">
1125 <tr class="layout">
1126 <td class="left">
1127 <tt>[3 x [4 x i32]]</tt><br/>
1128 <tt>[12 x [10 x float]]</tt><br/>
1129 <tt>[2 x [3 x [4 x i16]]]</tt><br/>
1130 </td>
1131 <td class="left">
1132 3x4 array of 32-bit integer values.<br/>
1133 12x10 array of single precision floating point values.<br/>
1134 2x3x4 array of 16-bit integer values.<br/>
1135 </td>
1136 </tr>
1137</table>
1138
1139<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
1140length array. Normally, accesses past the end of an array are undefined in
1141LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
1142As a special case, however, zero length arrays are recognized to be variable
1143length. This allows implementation of 'pascal style arrays' with the LLVM
1144type "{ i32, [0 x float]}", for example.</p>
1145
1146</div>
1147
1148<!-- _______________________________________________________________________ -->
1149<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
1150<div class="doc_text">
1151<h5>Overview:</h5>
1152<p>The function type can be thought of as a function signature. It
1153consists of a return type and a list of formal parameter types.
1154Function types are usually used to build virtual function tables
1155(which are structures of pointers to functions), for indirect function
1156calls, and when defining a function.</p>
1157<p>
1158The return type of a function type cannot be an aggregate type.
1159</p>
1160<h5>Syntax:</h5>
1161<pre> &lt;returntype&gt; (&lt;parameter list&gt;)<br></pre>
1162<p>...where '<tt>&lt;parameter list&gt;</tt>' is a comma-separated list of type
1163specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
1164which indicates that the function takes a variable number of arguments.
1165Variable argument functions can access their arguments with the <a
1166 href="#int_varargs">variable argument handling intrinsic</a> functions.</p>
1167<h5>Examples:</h5>
1168<table class="layout">
1169 <tr class="layout">
1170 <td class="left"><tt>i32 (i32)</tt></td>
1171 <td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
1172 </td>
1173 </tr><tr class="layout">
Reid Spencerf234bed2007-07-19 23:13:04 +0000[diff] [blame]1174 <td class="left"><tt>float&nbsp;(i16&nbsp;signext,&nbsp;i32&nbsp;*)&nbsp;*
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1175 </tt></td>
1176 <td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
1177 an <tt>i16</tt> that should be sign extended and a
1178 <a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
1179 <tt>float</tt>.
1180 </td>
1181 </tr><tr class="layout">
1182 <td class="left"><tt>i32 (i8*, ...)</tt></td>
1183 <td class="left">A vararg function that takes at least one
1184 <a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
1185 which returns an integer. This is the signature for <tt>printf</tt> in
1186 LLVM.
1187 </td>
1188 </tr>
1189</table>
1190
1191</div>
1192<!-- _______________________________________________________________________ -->
1193<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
1194<div class="doc_text">
1195<h5>Overview:</h5>
1196<p>The structure type is used to represent a collection of data members
1197together in memory. The packing of the field types is defined to match
1198the ABI of the underlying processor. The elements of a structure may
1199be any type that has a size.</p>
1200<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1201and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1202field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1203instruction.</p>
1204<h5>Syntax:</h5>
1205<pre> { &lt;type list&gt; }<br></pre>
1206<h5>Examples:</h5>
1207<table class="layout">
1208 <tr class="layout">
1209 <td class="left"><tt>{ i32, i32, i32 }</tt></td>
1210 <td class="left">A triple of three <tt>i32</tt> values</td>
1211 </tr><tr class="layout">
1212 <td class="left"><tt>{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}</tt></td>
1213 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1214 second element is a <a href="#t_pointer">pointer</a> to a
1215 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1216 an <tt>i32</tt>.</td>
1217 </tr>
1218</table>
1219</div>
1220
1221<!-- _______________________________________________________________________ -->
1222<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
1223</div>
1224<div class="doc_text">
1225<h5>Overview:</h5>
1226<p>The packed structure type is used to represent a collection of data members
1227together in memory. There is no padding between fields. Further, the alignment
1228of a packed structure is 1 byte. The elements of a packed structure may
1229be any type that has a size.</p>
1230<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
1231and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
1232field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
1233instruction.</p>
1234<h5>Syntax:</h5>
1235<pre> &lt; { &lt;type list&gt; } &gt; <br></pre>
1236<h5>Examples:</h5>
1237<table class="layout">
1238 <tr class="layout">
1239 <td class="left"><tt>&lt; { i32, i32, i32 } &gt;</tt></td>
1240 <td class="left">A triple of three <tt>i32</tt> values</td>
1241 </tr><tr class="layout">
1242 <td class="left"><tt>&lt;&nbsp;{&nbsp;float,&nbsp;i32&nbsp;(i32)&nbsp;*&nbsp;}&nbsp;&gt;</tt></td>
1243 <td class="left">A pair, where the first element is a <tt>float</tt> and the
1244 second element is a <a href="#t_pointer">pointer</a> to a
1245 <a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
1246 an <tt>i32</tt>.</td>
1247 </tr>
1248</table>
1249</div>
1250
1251<!-- _______________________________________________________________________ -->
1252<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
1253<div class="doc_text">
1254<h5>Overview:</h5>
1255<p>As in many languages, the pointer type represents a pointer or
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]1256reference to another object, which must live in memory. Pointer types may have
1257an optional address space attribute defining the target-specific numbered
1258address space where the pointed-to object resides. The default address space is
1259zero.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1260<h5>Syntax:</h5>
1261<pre> &lt;type&gt; *<br></pre>
1262<h5>Examples:</h5>
1263<table class="layout">
1264 <tr class="layout">
1265 <td class="left">
1266 <tt>[4x i32]*</tt><br/>
1267 <tt>i32 (i32 *) *</tt><br/>
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]1268 <tt>i32 addrspace(5)*</tt><br/>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1269 </td>
1270 <td class="left">
1271 A <a href="#t_pointer">pointer</a> to <a href="#t_array">array</a> of
1272 four <tt>i32</tt> values<br/>
1273 A <a href="#t_pointer">pointer</a> to a <a
1274 href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
1275 <tt>i32</tt>.<br/>
Christopher Lambdd0049d2007-12-11 09:31:00 +0000[diff] [blame]1276 A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value that resides
1277 in address space 5.<br/>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1278 </td>
1279 </tr>
1280</table>
1281</div>
1282
1283<!-- _______________________________________________________________________ -->
1284<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
1285<div class="doc_text">
1286
1287<h5>Overview:</h5>
1288
1289<p>A vector type is a simple derived type that represents a vector
1290of elements. Vector types are used when multiple primitive data
1291are operated in parallel using a single instruction (SIMD).
1292A vector type requires a size (number of
1293elements) and an underlying primitive data type. Vectors must have a power
1294of two length (1, 2, 4, 8, 16 ...). Vector types are
1295considered <a href="#t_firstclass">first class</a>.</p>
1296
1297<h5>Syntax:</h5>
1298
1299<pre>
1300 &lt; &lt;# elements&gt; x &lt;elementtype&gt; &gt;
1301</pre>
1302
1303<p>The number of elements is a constant integer value; elementtype may
1304be any integer or floating point type.</p>
1305
1306<h5>Examples:</h5>
1307
1308<table class="layout">
1309 <tr class="layout">
1310 <td class="left">
1311 <tt>&lt;4 x i32&gt;</tt><br/>
1312 <tt>&lt;8 x float&gt;</tt><br/>
1313 <tt>&lt;2 x i64&gt;</tt><br/>
1314 </td>
1315 <td class="left">
1316 Vector of 4 32-bit integer values.<br/>
1317 Vector of 8 floating-point values.<br/>
1318 Vector of 2 64-bit integer values.<br/>
1319 </td>
1320 </tr>
1321</table>
1322</div>
1323
1324<!-- _______________________________________________________________________ -->
1325<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
1326<div class="doc_text">
1327
1328<h5>Overview:</h5>
1329
1330<p>Opaque types are used to represent unknown types in the system. This
Gordon Henriksenda0706e2007-10-14 00:34:53 +0000[diff] [blame]1331corresponds (for example) to the C notion of a forward declared structure type.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1332In LLVM, opaque types can eventually be resolved to any type (not just a
1333structure type).</p>
1334
1335<h5>Syntax:</h5>
1336
1337<pre>
1338 opaque
1339</pre>
1340
1341<h5>Examples:</h5>
1342
1343<table class="layout">
1344 <tr class="layout">
1345 <td class="left">
1346 <tt>opaque</tt>
1347 </td>
1348 <td class="left">
1349 An opaque type.<br/>
1350 </td>
1351 </tr>
1352</table>
1353</div>
1354
1355
1356<!-- *********************************************************************** -->
1357<div class="doc_section"> <a name="constants">Constants</a> </div>
1358<!-- *********************************************************************** -->
1359
1360<div class="doc_text">
1361
1362<p>LLVM has several different basic types of constants. This section describes
1363them all and their syntax.</p>
1364
1365</div>
1366
1367<!-- ======================================================================= -->
1368<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
1369
1370<div class="doc_text">
1371
1372<dl>
1373 <dt><b>Boolean constants</b></dt>
1374
1375 <dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
1376 constants of the <tt><a href="#t_primitive">i1</a></tt> type.
1377 </dd>
1378
1379 <dt><b>Integer constants</b></dt>
1380
1381 <dd>Standard integers (such as '4') are constants of the <a
1382 href="#t_integer">integer</a> type. Negative numbers may be used with
1383 integer types.
1384 </dd>
1385
1386 <dt><b>Floating point constants</b></dt>
1387
1388 <dd>Floating point constants use standard decimal notation (e.g. 123.421),
1389 exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
1390 notation (see below). Floating point constants must have a <a
1391 href="#t_floating">floating point</a> type. </dd>
1392
1393 <dt><b>Null pointer constants</b></dt>
1394
1395 <dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
1396 and must be of <a href="#t_pointer">pointer type</a>.</dd>
1397
1398</dl>
1399
1400<p>The one non-intuitive notation for constants is the optional hexadecimal form
1401of floating point constants. For example, the form '<tt>double
14020x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
14034.5e+15</tt>'. The only time hexadecimal floating point constants are required
1404(and the only time that they are generated by the disassembler) is when a
1405floating point constant must be emitted but it cannot be represented as a
1406decimal floating point number. For example, NaN's, infinities, and other
1407special values are represented in their IEEE hexadecimal format so that
1408assembly and disassembly do not cause any bits to change in the constants.</p>
1409
1410</div>
1411
1412<!-- ======================================================================= -->
1413<div class="doc_subsection"><a name="aggregateconstants">Aggregate Constants</a>
1414</div>
1415
1416<div class="doc_text">
1417<p>Aggregate constants arise from aggregation of simple constants
1418and smaller aggregate constants.</p>
1419
1420<dl>
1421 <dt><b>Structure constants</b></dt>
1422
1423 <dd>Structure constants are represented with notation similar to structure
1424 type definitions (a comma separated list of elements, surrounded by braces
1425 (<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* %G }</tt>",
1426 where "<tt>%G</tt>" is declared as "<tt>@G = external global i32</tt>". Structure constants
1427 must have <a href="#t_struct">structure type</a>, and the number and
1428 types of elements must match those specified by the type.
1429 </dd>
1430
1431 <dt><b>Array constants</b></dt>
1432
1433 <dd>Array constants are represented with notation similar to array type
1434 definitions (a comma separated list of elements, surrounded by square brackets
1435 (<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74 ]</tt>". Array
1436 constants must have <a href="#t_array">array type</a>, and the number and
1437 types of elements must match those specified by the type.
1438 </dd>
1439
1440 <dt><b>Vector constants</b></dt>
1441
1442 <dd>Vector constants are represented with notation similar to vector type
1443 definitions (a comma separated list of elements, surrounded by
1444 less-than/greater-than's (<tt>&lt;&gt;</tt>)). For example: "<tt>&lt; i32 42,
1445 i32 11, i32 74, i32 100 &gt;</tt>". Vector constants must have <a
1446 href="#t_vector">vector type</a>, and the number and types of elements must
1447 match those specified by the type.
1448 </dd>
1449
1450 <dt><b>Zero initialization</b></dt>
1451
1452 <dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
1453 value to zero of <em>any</em> type, including scalar and aggregate types.
1454 This is often used to avoid having to print large zero initializers (e.g. for
1455 large arrays) and is always exactly equivalent to using explicit zero
1456 initializers.
1457 </dd>
1458</dl>
1459
1460</div>
1461
1462<!-- ======================================================================= -->
1463<div class="doc_subsection">
1464 <a name="globalconstants">Global Variable and Function Addresses</a>
1465</div>
1466
1467<div class="doc_text">
1468
1469<p>The addresses of <a href="#globalvars">global variables</a> and <a
1470href="#functionstructure">functions</a> are always implicitly valid (link-time)
1471constants. These constants are explicitly referenced when the <a
1472href="#identifiers">identifier for the global</a> is used and always have <a
1473href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
1474file:</p>
1475
1476<div class="doc_code">
1477<pre>
1478@X = global i32 17
1479@Y = global i32 42
1480@Z = global [2 x i32*] [ i32* @X, i32* @Y ]
1481</pre>
1482</div>
1483
1484</div>
1485
1486<!-- ======================================================================= -->
1487<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
1488<div class="doc_text">
1489 <p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
1490 no specific value. Undefined values may be of any type and be used anywhere
1491 a constant is permitted.</p>
1492
1493 <p>Undefined values indicate to the compiler that the program is well defined
1494 no matter what value is used, giving the compiler more freedom to optimize.
1495 </p>
1496</div>
1497
1498<!-- ======================================================================= -->
1499<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
1500</div>
1501
1502<div class="doc_text">
1503
1504<p>Constant expressions are used to allow expressions involving other constants
1505to be used as constants. Constant expressions may be of any <a
1506href="#t_firstclass">first class</a> type and may involve any LLVM operation
1507that does not have side effects (e.g. load and call are not supported). The
1508following is the syntax for constant expressions:</p>
1509
1510<dl>
1511 <dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
1512 <dd>Truncate a constant to another type. The bit size of CST must be larger
1513 than the bit size of TYPE. Both types must be integers.</dd>
1514
1515 <dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
1516 <dd>Zero extend a constant to another type. The bit size of CST must be
1517 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
1518
1519 <dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
1520 <dd>Sign extend a constant to another type. The bit size of CST must be
1521 smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
1522
1523 <dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
1524 <dd>Truncate a floating point constant to another floating point type. The
1525 size of CST must be larger than the size of TYPE. Both types must be
1526 floating point.</dd>
1527
1528 <dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
1529 <dd>Floating point extend a constant to another type. The size of CST must be
1530 smaller or equal to the size of TYPE. Both types must be floating point.</dd>
1531
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]1532 <dt><b><tt>fptoui ( CST to TYPE )</tt></b></dt>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1533 <dd>Convert a floating point constant to the corresponding unsigned integer
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]1534 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1535 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1536 of the same number of elements. If the value won't fit in the integer type,
1537 the results are undefined.</dd>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1538
1539 <dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
1540 <dd>Convert a floating point constant to the corresponding signed integer
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]1541 constant. TYPE must be a scalar or vector integer type. CST must be of scalar
1542 or vector floating point type. Both CST and TYPE must be scalars, or vectors
1543 of the same number of elements. If the value won't fit in the integer type,
1544 the results are undefined.</dd>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1545
1546 <dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
1547 <dd>Convert an unsigned integer constant to the corresponding floating point
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]1548 constant. TYPE must be a scalar or vector floating point type. CST must be of
1549 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1550 of the same number of elements. If the value won't fit in the floating point
1551 type, the results are undefined.</dd>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1552
1553 <dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
1554 <dd>Convert a signed integer constant to the corresponding floating point
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]1555 constant. TYPE must be a scalar or vector floating point type. CST must be of
1556 scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
1557 of the same number of elements. If the value won't fit in the floating point
1558 type, the results are undefined.</dd>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]1559
1560 <dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
1561 <dd>Convert a pointer typed constant to the corresponding integer constant
1562 TYPE must be an integer type. CST must be of pointer type. The CST value is
1563 zero extended, truncated, or unchanged to make it fit in TYPE.</dd>
1564
1565 <dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
1566 <dd>Convert a integer constant to a pointer constant. TYPE must be a
1567 pointer type. CST must be of integer type. The CST value is zero extended,
1568 truncated, or unchanged to make it fit in a pointer size. This one is
1569 <i>really</i> dangerous!</dd>
1570
1571 <dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
1572 <dd>Convert a constant, CST, to another TYPE. The size of CST and TYPE must be
1573 identical (same number of bits). The conversion is done as if the CST value
1574 was stored to memory and read back as TYPE. In other words, no bits change
1575 with this operator, just the type. This can be used for conversion of
1576 vector types to any other type, as long as they have the same bit width. For
1577 pointers it is only valid to cast to another pointer type.
1578 </dd>
1579
1580 <dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
1581
1582 <dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
1583 constants. As with the <a href="#i_getelementptr">getelementptr</a>
1584 instruction, the index list may have zero or more indexes, which are required
1585 to make sense for the type of "CSTPTR".</dd>
1586
1587 <dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
1588
1589 <dd>Perform the <a href="#i_select">select operation</a> on
1590 constants.</dd>
1591
1592 <dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
1593 <dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
1594
1595 <dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
1596 <dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
1597
1598 <dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
1599
1600 <dd>Perform the <a href="#i_extractelement">extractelement
1601 operation</a> on constants.
1602
1603 <dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
1604
1605 <dd>Perform the <a href="#i_insertelement">insertelement
1606 operation</a> on constants.</dd>
1607
1608
1609 <dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
1610
1611 <dd>Perform the <a href="#i_shufflevector">shufflevector
1612 operation</a> on constants.</dd>
1613
1614 <dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
1615
1616 <dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
1617 be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
1618 binary</a> operations. The constraints on operands are the same as those for
1619 the corresponding instruction (e.g. no bitwise operations on floating point
1620 values are allowed).</dd>
1621</dl>
1622</div>
1623
1624<!-- *********************************************************************** -->
1625<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
1626<!-- *********************************************************************** -->
1627
1628<!-- ======================================================================= -->
1629<div class="doc_subsection">
1630<a name="inlineasm">Inline Assembler Expressions</a>
1631</div>
1632
1633<div class="doc_text">
1634
1635<p>
1636LLVM supports inline assembler expressions (as opposed to <a href="#moduleasm">
1637Module-Level Inline Assembly</a>) through the use of a special value. This
1638value represents the inline assembler as a string (containing the instructions
1639to emit), a list of operand constraints (stored as a string), and a flag that
1640indicates whether or not the inline asm expression has side effects. An example
1641inline assembler expression is:
1642</p>
1643
1644<div class="doc_code">
1645<pre>
1646i32 (i32) asm "bswap $0", "=r,r"
1647</pre>
1648</div>
1649
1650<p>
1651Inline assembler expressions may <b>only</b> be used as the callee operand of
1652a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we have:
1653</p>
1654
1655<div class="doc_code">
1656<pre>
1657%X = call i32 asm "<a href="#int_bswap">bswap</a> $0", "=r,r"(i32 %Y)
1658</pre>
1659</div>
1660
1661<p>
1662Inline asms with side effects not visible in the constraint list must be marked
1663as having side effects. This is done through the use of the
1664'<tt>sideeffect</tt>' keyword, like so:
1665</p>
1666
1667<div class="doc_code">
1668<pre>
1669call void asm sideeffect "eieio", ""()
1670</pre>
1671</div>
1672
1673<p>TODO: The format of the asm and constraints string still need to be
1674documented here. Constraints on what can be done (e.g. duplication, moving, etc
1675need to be documented).
1676</p>
1677
1678</div>
1679
1680<!-- *********************************************************************** -->
1681<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
1682<!-- *********************************************************************** -->
1683
1684<div class="doc_text">
1685
1686<p>The LLVM instruction set consists of several different
1687classifications of instructions: <a href="#terminators">terminator
1688instructions</a>, <a href="#binaryops">binary instructions</a>,
1689<a href="#bitwiseops">bitwise binary instructions</a>, <a
1690 href="#memoryops">memory instructions</a>, and <a href="#otherops">other
1691instructions</a>.</p>
1692
1693</div>
1694
1695<!-- ======================================================================= -->
1696<div class="doc_subsection"> <a name="terminators">Terminator
1697Instructions</a> </div>
1698
1699<div class="doc_text">
1700
1701<p>As mentioned <a href="#functionstructure">previously</a>, every
1702basic block in a program ends with a "Terminator" instruction, which
1703indicates which block should be executed after the current block is
1704finished. These terminator instructions typically yield a '<tt>void</tt>'
1705value: they produce control flow, not values (the one exception being
1706the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
1707<p>There are six different terminator instructions: the '<a
1708 href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
1709instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
1710the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
1711 href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
1712 href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
1713
1714</div>
1715
1716<!-- _______________________________________________________________________ -->
1717<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
1718Instruction</a> </div>
1719<div class="doc_text">
1720<h5>Syntax:</h5>
1721<pre> ret &lt;type&gt; &lt;value&gt; <i>; Return a value from a non-void function</i>
1722 ret void <i>; Return from void function</i>
1723</pre>
1724<h5>Overview:</h5>
1725<p>The '<tt>ret</tt>' instruction is used to return control flow (and a
1726value) from a function back to the caller.</p>
1727<p>There are two forms of the '<tt>ret</tt>' instruction: one that
1728returns a value and then causes control flow, and one that just causes
1729control flow to occur.</p>
1730<h5>Arguments:</h5>
1731<p>The '<tt>ret</tt>' instruction may return any '<a
1732 href="#t_firstclass">first class</a>' type. Notice that a function is
1733not <a href="#wellformed">well formed</a> if there exists a '<tt>ret</tt>'
1734instruction inside of the function that returns a value that does not
1735match the return type of the function.</p>
1736<h5>Semantics:</h5>
1737<p>When the '<tt>ret</tt>' instruction is executed, control flow
1738returns back to the calling function's context. If the caller is a "<a
1739 href="#i_call"><tt>call</tt></a>" instruction, execution continues at
1740the instruction after the call. If the caller was an "<a
1741 href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
1742at the beginning of the "normal" destination block. If the instruction
1743returns a value, that value shall set the call or invoke instruction's
1744return value.</p>
1745<h5>Example:</h5>
1746<pre> ret i32 5 <i>; Return an integer value of 5</i>
1747 ret void <i>; Return from a void function</i>
1748</pre>
1749</div>
1750<!-- _______________________________________________________________________ -->
1751<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
1752<div class="doc_text">
1753<h5>Syntax:</h5>
1754<pre> br i1 &lt;cond&gt;, label &lt;iftrue&gt;, label &lt;iffalse&gt;<br> br label &lt;dest&gt; <i>; Unconditional branch</i>
1755</pre>
1756<h5>Overview:</h5>
1757<p>The '<tt>br</tt>' instruction is used to cause control flow to
1758transfer to a different basic block in the current function. There are
1759two forms of this instruction, corresponding to a conditional branch
1760and an unconditional branch.</p>
1761<h5>Arguments:</h5>
1762<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
1763single '<tt>i1</tt>' value and two '<tt>label</tt>' values. The
1764unconditional form of the '<tt>br</tt>' instruction takes a single
1765'<tt>label</tt>' value as a target.</p>
1766<h5>Semantics:</h5>
1767<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
1768argument is evaluated. If the value is <tt>true</tt>, control flows
1769to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
1770control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
1771<h5>Example:</h5>
1772<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
1773 href="#i_ret">ret</a> i32 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> i32 0<br></pre>
1774</div>
1775<!-- _______________________________________________________________________ -->
1776<div class="doc_subsubsection">
1777 <a name="i_switch">'<tt>switch</tt>' Instruction</a>
1778</div>
1779
1780<div class="doc_text">
1781<h5>Syntax:</h5>
1782
1783<pre>
1784 switch &lt;intty&gt; &lt;value&gt;, label &lt;defaultdest&gt; [ &lt;intty&gt; &lt;val&gt;, label &lt;dest&gt; ... ]
1785</pre>
1786
1787<h5>Overview:</h5>
1788
1789<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
1790several different places. It is a generalization of the '<tt>br</tt>'
1791instruction, allowing a branch to occur to one of many possible
1792destinations.</p>
1793
1794
1795<h5>Arguments:</h5>
1796
1797<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
1798comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
1799an array of pairs of comparison value constants and '<tt>label</tt>'s. The
1800table is not allowed to contain duplicate constant entries.</p>
1801
1802<h5>Semantics:</h5>
1803
1804<p>The <tt>switch</tt> instruction specifies a table of values and
1805destinations. When the '<tt>switch</tt>' instruction is executed, this
1806table is searched for the given value. If the value is found, control flow is
1807transfered to the corresponding destination; otherwise, control flow is
1808transfered to the default destination.</p>
1809
1810<h5>Implementation:</h5>
1811
1812<p>Depending on properties of the target machine and the particular
1813<tt>switch</tt> instruction, this instruction may be code generated in different
1814ways. For example, it could be generated as a series of chained conditional
1815branches or with a lookup table.</p>
1816
1817<h5>Example:</h5>
1818
1819<pre>
1820 <i>; Emulate a conditional br instruction</i>
1821 %Val = <a href="#i_zext">zext</a> i1 %value to i32
1822 switch i32 %Val, label %truedest [i32 0, label %falsedest ]
1823
1824 <i>; Emulate an unconditional br instruction</i>
1825 switch i32 0, label %dest [ ]
1826
1827 <i>; Implement a jump table:</i>
1828 switch i32 %val, label %otherwise [ i32 0, label %onzero
1829 i32 1, label %onone
1830 i32 2, label %ontwo ]
1831</pre>
1832</div>
1833
1834<!-- _______________________________________________________________________ -->
1835<div class="doc_subsubsection">
1836 <a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
1837</div>
1838
1839<div class="doc_text">
1840
1841<h5>Syntax:</h5>
1842
1843<pre>
1844 &lt;result&gt; = invoke [<a href="#callingconv">cconv</a>] &lt;ptr to function ty&gt; %&lt;function ptr val&gt;(&lt;function args&gt;)
1845 to label &lt;normal label&gt; unwind label &lt;exception label&gt;
1846</pre>
1847
1848<h5>Overview:</h5>
1849
1850<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
1851function, with the possibility of control flow transfer to either the
1852'<tt>normal</tt>' label or the
1853'<tt>exception</tt>' label. If the callee function returns with the
1854"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
1855"normal" label. If the callee (or any indirect callees) returns with the "<a
1856href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
1857continued at the dynamically nearest "exception" label.</p>
1858
1859<h5>Arguments:</h5>
1860
1861<p>This instruction requires several arguments:</p>
1862
1863<ol>
1864 <li>
1865 The optional "cconv" marker indicates which <a href="#callingconv">calling
1866 convention</a> the call should use. If none is specified, the call defaults
1867 to using C calling conventions.
1868 </li>
1869 <li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
1870 function value being invoked. In most cases, this is a direct function
1871 invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
1872 an arbitrary pointer to function value.
1873 </li>
1874
1875 <li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
1876 function to be invoked. </li>
1877
1878 <li>'<tt>function args</tt>': argument list whose types match the function
1879 signature argument types. If the function signature indicates the function
1880 accepts a variable number of arguments, the extra arguments can be
1881 specified. </li>
1882
1883 <li>'<tt>normal label</tt>': the label reached when the called function
1884 executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
1885
1886 <li>'<tt>exception label</tt>': the label reached when a callee returns with
1887 the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
1888
1889</ol>
1890
1891<h5>Semantics:</h5>
1892
1893<p>This instruction is designed to operate as a standard '<tt><a
1894href="#i_call">call</a></tt>' instruction in most regards. The primary
1895difference is that it establishes an association with a label, which is used by
1896the runtime library to unwind the stack.</p>
1897
1898<p>This instruction is used in languages with destructors to ensure that proper
1899cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
1900exception. Additionally, this is important for implementation of
1901'<tt>catch</tt>' clauses in high-level languages that support them.</p>
1902
1903<h5>Example:</h5>
1904<pre>
1905 %retval = invoke i32 %Test(i32 15) to label %Continue
1906 unwind label %TestCleanup <i>; {i32}:retval set</i>
1907 %retval = invoke <a href="#callingconv">coldcc</a> i32 %Test(i32 15) to label %Continue
1908 unwind label %TestCleanup <i>; {i32}:retval set</i>
1909</pre>
1910</div>
1911
1912
1913<!-- _______________________________________________________________________ -->
1914
1915<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
1916Instruction</a> </div>
1917
1918<div class="doc_text">
1919
1920<h5>Syntax:</h5>
1921<pre>
1922 unwind
1923</pre>
1924
1925<h5>Overview:</h5>
1926
1927<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
1928at the first callee in the dynamic call stack which used an <a
1929href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
1930primarily used to implement exception handling.</p>
1931
1932<h5>Semantics:</h5>
1933
1934<p>The '<tt>unwind</tt>' intrinsic causes execution of the current function to
1935immediately halt. The dynamic call stack is then searched for the first <a
1936href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
1937execution continues at the "exceptional" destination block specified by the
1938<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
1939dynamic call chain, undefined behavior results.</p>
1940</div>
1941
1942<!-- _______________________________________________________________________ -->
1943
1944<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
1945Instruction</a> </div>
1946
1947<div class="doc_text">
1948
1949<h5>Syntax:</h5>
1950<pre>
1951 unreachable
1952</pre>
1953
1954<h5>Overview:</h5>
1955
1956<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
1957instruction is used to inform the optimizer that a particular portion of the
1958code is not reachable. This can be used to indicate that the code after a
1959no-return function cannot be reached, and other facts.</p>
1960
1961<h5>Semantics:</h5>
1962
1963<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
1964</div>
1965
1966
1967
1968<!-- ======================================================================= -->
1969<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
1970<div class="doc_text">
1971<p>Binary operators are used to do most of the computation in a
1972program. They require two operands, execute an operation on them, and
1973produce a single value. The operands might represent
1974multiple data, as is the case with the <a href="#t_vector">vector</a> data type.
1975The result value of a binary operator is not
1976necessarily the same type as its operands.</p>
1977<p>There are several different binary operators:</p>
1978</div>
1979<!-- _______________________________________________________________________ -->
1980<div class="doc_subsubsection"> <a name="i_add">'<tt>add</tt>'
1981Instruction</a> </div>
1982<div class="doc_text">
1983<h5>Syntax:</h5>
1984<pre> &lt;result&gt; = add &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
1985</pre>
1986<h5>Overview:</h5>
1987<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
1988<h5>Arguments:</h5>
1989<p>The two arguments to the '<tt>add</tt>' instruction must be either <a
1990 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a> values.
1991 This instruction can also take <a href="#t_vector">vector</a> versions of the values.
1992Both arguments must have identical types.</p>
1993<h5>Semantics:</h5>
1994<p>The value produced is the integer or floating point sum of the two
1995operands.</p>
1996<h5>Example:</h5>
1997<pre> &lt;result&gt; = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
1998</pre>
1999</div>
2000<!-- _______________________________________________________________________ -->
2001<div class="doc_subsubsection"> <a name="i_sub">'<tt>sub</tt>'
2002Instruction</a> </div>
2003<div class="doc_text">
2004<h5>Syntax:</h5>
2005<pre> &lt;result&gt; = sub &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2006</pre>
2007<h5>Overview:</h5>
2008<p>The '<tt>sub</tt>' instruction returns the difference of its two
2009operands.</p>
2010<p>Note that the '<tt>sub</tt>' instruction is used to represent the '<tt>neg</tt>'
2011instruction present in most other intermediate representations.</p>
2012<h5>Arguments:</h5>
2013<p>The two arguments to the '<tt>sub</tt>' instruction must be either <a
2014 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
2015values.
2016This instruction can also take <a href="#t_vector">vector</a> versions of the values.
2017Both arguments must have identical types.</p>
2018<h5>Semantics:</h5>
2019<p>The value produced is the integer or floating point difference of
2020the two operands.</p>
2021<h5>Example:</h5>
2022<pre>
2023 &lt;result&gt; = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
2024 &lt;result&gt; = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
2025</pre>
2026</div>
2027<!-- _______________________________________________________________________ -->
2028<div class="doc_subsubsection"> <a name="i_mul">'<tt>mul</tt>'
2029Instruction</a> </div>
2030<div class="doc_text">
2031<h5>Syntax:</h5>
2032<pre> &lt;result&gt; = mul &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2033</pre>
2034<h5>Overview:</h5>
2035<p>The '<tt>mul</tt>' instruction returns the product of its two
2036operands.</p>
2037<h5>Arguments:</h5>
2038<p>The two arguments to the '<tt>mul</tt>' instruction must be either <a
2039 href="#t_integer">integer</a> or <a href="#t_floating">floating point</a>
2040values.
2041This instruction can also take <a href="#t_vector">vector</a> versions of the values.
2042Both arguments must have identical types.</p>
2043<h5>Semantics:</h5>
2044<p>The value produced is the integer or floating point product of the
2045two operands.</p>
2046<p>Because the operands are the same width, the result of an integer
2047multiplication is the same whether the operands should be deemed unsigned or
2048signed.</p>
2049<h5>Example:</h5>
2050<pre> &lt;result&gt; = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
2051</pre>
2052</div>
2053<!-- _______________________________________________________________________ -->
2054<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
2055</a></div>
2056<div class="doc_text">
2057<h5>Syntax:</h5>
2058<pre> &lt;result&gt; = udiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2059</pre>
2060<h5>Overview:</h5>
2061<p>The '<tt>udiv</tt>' instruction returns the quotient of its two
2062operands.</p>
2063<h5>Arguments:</h5>
2064<p>The two arguments to the '<tt>udiv</tt>' instruction must be
2065<a href="#t_integer">integer</a> values. Both arguments must have identical
2066types. This instruction can also take <a href="#t_vector">vector</a> versions
2067of the values in which case the elements must be integers.</p>
2068<h5>Semantics:</h5>
2069<p>The value produced is the unsigned integer quotient of the two operands. This
2070instruction always performs an unsigned division operation, regardless of
2071whether the arguments are unsigned or not.</p>
2072<h5>Example:</h5>
2073<pre> &lt;result&gt; = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
2074</pre>
2075</div>
2076<!-- _______________________________________________________________________ -->
2077<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
2078</a> </div>
2079<div class="doc_text">
2080<h5>Syntax:</h5>
2081<pre> &lt;result&gt; = sdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2082</pre>
2083<h5>Overview:</h5>
2084<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two
2085operands.</p>
2086<h5>Arguments:</h5>
2087<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
2088<a href="#t_integer">integer</a> values. Both arguments must have identical
2089types. This instruction can also take <a href="#t_vector">vector</a> versions
2090of the values in which case the elements must be integers.</p>
2091<h5>Semantics:</h5>
2092<p>The value produced is the signed integer quotient of the two operands. This
2093instruction always performs a signed division operation, regardless of whether
2094the arguments are signed or not.</p>
2095<h5>Example:</h5>
2096<pre> &lt;result&gt; = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
2097</pre>
2098</div>
2099<!-- _______________________________________________________________________ -->
2100<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
2101Instruction</a> </div>
2102<div class="doc_text">
2103<h5>Syntax:</h5>
2104<pre> &lt;result&gt; = fdiv &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2105</pre>
2106<h5>Overview:</h5>
2107<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two
2108operands.</p>
2109<h5>Arguments:</h5>
2110<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
2111<a href="#t_floating">floating point</a> values. Both arguments must have
2112identical types. This instruction can also take <a href="#t_vector">vector</a>
2113versions of floating point values.</p>
2114<h5>Semantics:</h5>
2115<p>The value produced is the floating point quotient of the two operands.</p>
2116<h5>Example:</h5>
2117<pre> &lt;result&gt; = fdiv float 4.0, %var <i>; yields {float}:result = 4.0 / %var</i>
2118</pre>
2119</div>
2120<!-- _______________________________________________________________________ -->
2121<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
2122</div>
2123<div class="doc_text">
2124<h5>Syntax:</h5>
2125<pre> &lt;result&gt; = urem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2126</pre>
2127<h5>Overview:</h5>
2128<p>The '<tt>urem</tt>' instruction returns the remainder from the
2129unsigned division of its two arguments.</p>
2130<h5>Arguments:</h5>
2131<p>The two arguments to the '<tt>urem</tt>' instruction must be
2132<a href="#t_integer">integer</a> values. Both arguments must have identical
Dan Gohman3e3fd8c2007-11-05 23:35:22 +0000[diff] [blame]2133types. This instruction can also take <a href="#t_vector">vector</a> versions
2134of the values in which case the elements must be integers.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2135<h5>Semantics:</h5>
2136<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
2137This instruction always performs an unsigned division to get the remainder,
2138regardless of whether the arguments are unsigned or not.</p>
2139<h5>Example:</h5>
2140<pre> &lt;result&gt; = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
2141</pre>
2142
2143</div>
2144<!-- _______________________________________________________________________ -->
2145<div class="doc_subsubsection"> <a name="i_srem">'<tt>srem</tt>'
2146Instruction</a> </div>
2147<div class="doc_text">
2148<h5>Syntax:</h5>
2149<pre> &lt;result&gt; = srem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2150</pre>
2151<h5>Overview:</h5>
2152<p>The '<tt>srem</tt>' instruction returns the remainder from the
Dan Gohman3e3fd8c2007-11-05 23:35:22 +0000[diff] [blame]2153signed division of its two operands. This instruction can also take
2154<a href="#t_vector">vector</a> versions of the values in which case
2155the elements must be integers.</p>
2156</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2157<h5>Arguments:</h5>
2158<p>The two arguments to the '<tt>srem</tt>' instruction must be
2159<a href="#t_integer">integer</a> values. Both arguments must have identical
2160types.</p>
2161<h5>Semantics:</h5>
2162<p>This instruction returns the <i>remainder</i> of a division (where the result
2163has the same sign as the dividend, <tt>var1</tt>), not the <i>modulo</i>
2164operator (where the result has the same sign as the divisor, <tt>var2</tt>) of
2165a value. For more information about the difference, see <a
2166 href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
2167Math Forum</a>. For a table of how this is implemented in various languages,
2168please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
2169Wikipedia: modulo operation</a>.</p>
2170<h5>Example:</h5>
2171<pre> &lt;result&gt; = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
2172</pre>
2173
2174</div>
2175<!-- _______________________________________________________________________ -->
2176<div class="doc_subsubsection"> <a name="i_frem">'<tt>frem</tt>'
2177Instruction</a> </div>
2178<div class="doc_text">
2179<h5>Syntax:</h5>
2180<pre> &lt;result&gt; = frem &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2181</pre>
2182<h5>Overview:</h5>
2183<p>The '<tt>frem</tt>' instruction returns the remainder from the
2184division of its two operands.</p>
2185<h5>Arguments:</h5>
2186<p>The two arguments to the '<tt>frem</tt>' instruction must be
2187<a href="#t_floating">floating point</a> values. Both arguments must have
Dan Gohman3e3fd8c2007-11-05 23:35:22 +0000[diff] [blame]2188identical types. This instruction can also take <a href="#t_vector">vector</a>
2189versions of floating point values.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2190<h5>Semantics:</h5>
2191<p>This instruction returns the <i>remainder</i> of a division.</p>
2192<h5>Example:</h5>
2193<pre> &lt;result&gt; = frem float 4.0, %var <i>; yields {float}:result = 4.0 % %var</i>
2194</pre>
2195</div>
2196
2197<!-- ======================================================================= -->
2198<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
2199Operations</a> </div>
2200<div class="doc_text">
2201<p>Bitwise binary operators are used to do various forms of
2202bit-twiddling in a program. They are generally very efficient
2203instructions and can commonly be strength reduced from other
2204instructions. They require two operands, execute an operation on them,
2205and produce a single value. The resulting value of the bitwise binary
2206operators is always the same type as its first operand.</p>
2207</div>
2208
2209<!-- _______________________________________________________________________ -->
2210<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
2211Instruction</a> </div>
2212<div class="doc_text">
2213<h5>Syntax:</h5>
2214<pre> &lt;result&gt; = shl &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2215</pre>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2216
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2217<h5>Overview:</h5>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2218
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2219<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
2220the left a specified number of bits.</p>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2221
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2222<h5>Arguments:</h5>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2223
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2224<p>Both arguments to the '<tt>shl</tt>' instruction must be the same <a
2225 href="#t_integer">integer</a> type.</p>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2226
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2227<h5>Semantics:</h5>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2228
2229<p>The value produced is <tt>var1</tt> * 2<sup><tt>var2</tt></sup>. If
2230<tt>var2</tt> is (statically or dynamically) equal to or larger than the number
2231of bits in <tt>var1</tt>, the result is undefined.</p>
2232
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2233<h5>Example:</h5><pre>
2234 &lt;result&gt; = shl i32 4, %var <i>; yields {i32}: 4 &lt;&lt; %var</i>
2235 &lt;result&gt; = shl i32 4, 2 <i>; yields {i32}: 16</i>
2236 &lt;result&gt; = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2237 &lt;result&gt; = shl i32 1, 32 <i>; undefined</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2238</pre>
2239</div>
2240<!-- _______________________________________________________________________ -->
2241<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
2242Instruction</a> </div>
2243<div class="doc_text">
2244<h5>Syntax:</h5>
2245<pre> &lt;result&gt; = lshr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2246</pre>
2247
2248<h5>Overview:</h5>
2249<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
2250operand shifted to the right a specified number of bits with zero fill.</p>
2251
2252<h5>Arguments:</h5>
2253<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
2254<a href="#t_integer">integer</a> type.</p>
2255
2256<h5>Semantics:</h5>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2257
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2258<p>This instruction always performs a logical shift right operation. The most
2259significant bits of the result will be filled with zero bits after the
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2260shift. If <tt>var2</tt> is (statically or dynamically) equal to or larger than
2261the number of bits in <tt>var1</tt>, the result is undefined.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2262
2263<h5>Example:</h5>
2264<pre>
2265 &lt;result&gt; = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
2266 &lt;result&gt; = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
2267 &lt;result&gt; = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
2268 &lt;result&gt; = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2269 &lt;result&gt; = lshr i32 1, 32 <i>; undefined</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2270</pre>
2271</div>
2272
2273<!-- _______________________________________________________________________ -->
2274<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
2275Instruction</a> </div>
2276<div class="doc_text">
2277
2278<h5>Syntax:</h5>
2279<pre> &lt;result&gt; = ashr &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2280</pre>
2281
2282<h5>Overview:</h5>
2283<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
2284operand shifted to the right a specified number of bits with sign extension.</p>
2285
2286<h5>Arguments:</h5>
2287<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
2288<a href="#t_integer">integer</a> type.</p>
2289
2290<h5>Semantics:</h5>
2291<p>This instruction always performs an arithmetic shift right operation,
2292The most significant bits of the result will be filled with the sign bit
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2293of <tt>var1</tt>. If <tt>var2</tt> is (statically or dynamically) equal to or
2294larger than the number of bits in <tt>var1</tt>, the result is undefined.
2295</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2296
2297<h5>Example:</h5>
2298<pre>
2299 &lt;result&gt; = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
2300 &lt;result&gt; = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
2301 &lt;result&gt; = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
2302 &lt;result&gt; = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattnerd939d9f2007-10-03 21:01:14 +0000[diff] [blame]2303 &lt;result&gt; = ashr i32 1, 32 <i>; undefined</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2304</pre>
2305</div>
2306
2307<!-- _______________________________________________________________________ -->
2308<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
2309Instruction</a> </div>
2310<div class="doc_text">
2311<h5>Syntax:</h5>
2312<pre> &lt;result&gt; = and &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2313</pre>
2314<h5>Overview:</h5>
2315<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
2316its two operands.</p>
2317<h5>Arguments:</h5>
2318<p>The two arguments to the '<tt>and</tt>' instruction must be <a
2319 href="#t_integer">integer</a> values. Both arguments must have
2320identical types.</p>
2321<h5>Semantics:</h5>
2322<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
2323<p> </p>
2324<div style="align: center">
2325<table border="1" cellspacing="0" cellpadding="4">
2326 <tbody>
2327 <tr>
2328 <td>In0</td>
2329 <td>In1</td>
2330 <td>Out</td>
2331 </tr>
2332 <tr>
2333 <td>0</td>
2334 <td>0</td>
2335 <td>0</td>
2336 </tr>
2337 <tr>
2338 <td>0</td>
2339 <td>1</td>
2340 <td>0</td>
2341 </tr>
2342 <tr>
2343 <td>1</td>
2344 <td>0</td>
2345 <td>0</td>
2346 </tr>
2347 <tr>
2348 <td>1</td>
2349 <td>1</td>
2350 <td>1</td>
2351 </tr>
2352 </tbody>
2353</table>
2354</div>
2355<h5>Example:</h5>
2356<pre> &lt;result&gt; = and i32 4, %var <i>; yields {i32}:result = 4 &amp; %var</i>
2357 &lt;result&gt; = and i32 15, 40 <i>; yields {i32}:result = 8</i>
2358 &lt;result&gt; = and i32 4, 8 <i>; yields {i32}:result = 0</i>
2359</pre>
2360</div>
2361<!-- _______________________________________________________________________ -->
2362<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
2363<div class="doc_text">
2364<h5>Syntax:</h5>
2365<pre> &lt;result&gt; = or &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2366</pre>
2367<h5>Overview:</h5>
2368<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
2369or of its two operands.</p>
2370<h5>Arguments:</h5>
2371<p>The two arguments to the '<tt>or</tt>' instruction must be <a
2372 href="#t_integer">integer</a> values. Both arguments must have
2373identical types.</p>
2374<h5>Semantics:</h5>
2375<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
2376<p> </p>
2377<div style="align: center">
2378<table border="1" cellspacing="0" cellpadding="4">
2379 <tbody>
2380 <tr>
2381 <td>In0</td>
2382 <td>In1</td>
2383 <td>Out</td>
2384 </tr>
2385 <tr>
2386 <td>0</td>
2387 <td>0</td>
2388 <td>0</td>
2389 </tr>
2390 <tr>
2391 <td>0</td>
2392 <td>1</td>
2393 <td>1</td>
2394 </tr>
2395 <tr>
2396 <td>1</td>
2397 <td>0</td>
2398 <td>1</td>
2399 </tr>
2400 <tr>
2401 <td>1</td>
2402 <td>1</td>
2403 <td>1</td>
2404 </tr>
2405 </tbody>
2406</table>
2407</div>
2408<h5>Example:</h5>
2409<pre> &lt;result&gt; = or i32 4, %var <i>; yields {i32}:result = 4 | %var</i>
2410 &lt;result&gt; = or i32 15, 40 <i>; yields {i32}:result = 47</i>
2411 &lt;result&gt; = or i32 4, 8 <i>; yields {i32}:result = 12</i>
2412</pre>
2413</div>
2414<!-- _______________________________________________________________________ -->
2415<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
2416Instruction</a> </div>
2417<div class="doc_text">
2418<h5>Syntax:</h5>
2419<pre> &lt;result&gt; = xor &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {ty}:result</i>
2420</pre>
2421<h5>Overview:</h5>
2422<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
2423or of its two operands. The <tt>xor</tt> is used to implement the
2424"one's complement" operation, which is the "~" operator in C.</p>
2425<h5>Arguments:</h5>
2426<p>The two arguments to the '<tt>xor</tt>' instruction must be <a
2427 href="#t_integer">integer</a> values. Both arguments must have
2428identical types.</p>
2429<h5>Semantics:</h5>
2430<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
2431<p> </p>
2432<div style="align: center">
2433<table border="1" cellspacing="0" cellpadding="4">
2434 <tbody>
2435 <tr>
2436 <td>In0</td>
2437 <td>In1</td>
2438 <td>Out</td>
2439 </tr>
2440 <tr>
2441 <td>0</td>
2442 <td>0</td>
2443 <td>0</td>
2444 </tr>
2445 <tr>
2446 <td>0</td>
2447 <td>1</td>
2448 <td>1</td>
2449 </tr>
2450 <tr>
2451 <td>1</td>
2452 <td>0</td>
2453 <td>1</td>
2454 </tr>
2455 <tr>
2456 <td>1</td>
2457 <td>1</td>
2458 <td>0</td>
2459 </tr>
2460 </tbody>
2461</table>
2462</div>
2463<p> </p>
2464<h5>Example:</h5>
2465<pre> &lt;result&gt; = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
2466 &lt;result&gt; = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
2467 &lt;result&gt; = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
2468 &lt;result&gt; = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
2469</pre>
2470</div>
2471
2472<!-- ======================================================================= -->
2473<div class="doc_subsection">
2474 <a name="vectorops">Vector Operations</a>
2475</div>
2476
2477<div class="doc_text">
2478
2479<p>LLVM supports several instructions to represent vector operations in a
2480target-independent manner. These instructions cover the element-access and
2481vector-specific operations needed to process vectors effectively. While LLVM
2482does directly support these vector operations, many sophisticated algorithms
2483will want to use target-specific intrinsics to take full advantage of a specific
2484target.</p>
2485
2486</div>
2487
2488<!-- _______________________________________________________________________ -->
2489<div class="doc_subsubsection">
2490 <a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
2491</div>
2492
2493<div class="doc_text">
2494
2495<h5>Syntax:</h5>
2496
2497<pre>
2498 &lt;result&gt; = extractelement &lt;n x &lt;ty&gt;&gt; &lt;val&gt;, i32 &lt;idx&gt; <i>; yields &lt;ty&gt;</i>
2499</pre>
2500
2501<h5>Overview:</h5>
2502
2503<p>
2504The '<tt>extractelement</tt>' instruction extracts a single scalar
2505element from a vector at a specified index.
2506</p>
2507
2508
2509<h5>Arguments:</h5>
2510
2511<p>
2512The first operand of an '<tt>extractelement</tt>' instruction is a
2513value of <a href="#t_vector">vector</a> type. The second operand is
2514an index indicating the position from which to extract the element.
2515The index may be a variable.</p>
2516
2517<h5>Semantics:</h5>
2518
2519<p>
2520The result is a scalar of the same type as the element type of
2521<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
2522<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
2523results are undefined.
2524</p>
2525
2526<h5>Example:</h5>
2527
2528<pre>
2529 %result = extractelement &lt;4 x i32&gt; %vec, i32 0 <i>; yields i32</i>
2530</pre>
2531</div>
2532
2533
2534<!-- _______________________________________________________________________ -->
2535<div class="doc_subsubsection">
2536 <a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
2537</div>
2538
2539<div class="doc_text">
2540
2541<h5>Syntax:</h5>
2542
2543<pre>
2544 &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>
2545</pre>
2546
2547<h5>Overview:</h5>
2548
2549<p>
2550The '<tt>insertelement</tt>' instruction inserts a scalar
2551element into a vector at a specified index.
2552</p>
2553
2554
2555<h5>Arguments:</h5>
2556
2557<p>
2558The first operand of an '<tt>insertelement</tt>' instruction is a
2559value of <a href="#t_vector">vector</a> type. The second operand is a
2560scalar value whose type must equal the element type of the first
2561operand. The third operand is an index indicating the position at
2562which to insert the value. The index may be a variable.</p>
2563
2564<h5>Semantics:</h5>
2565
2566<p>
2567The result is a vector of the same type as <tt>val</tt>. Its
2568element values are those of <tt>val</tt> except at position
2569<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
2570exceeds the length of <tt>val</tt>, the results are undefined.
2571</p>
2572
2573<h5>Example:</h5>
2574
2575<pre>
2576 %result = insertelement &lt;4 x i32&gt; %vec, i32 1, i32 0 <i>; yields &lt;4 x i32&gt;</i>
2577</pre>
2578</div>
2579
2580<!-- _______________________________________________________________________ -->
2581<div class="doc_subsubsection">
2582 <a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
2583</div>
2584
2585<div class="doc_text">
2586
2587<h5>Syntax:</h5>
2588
2589<pre>
2590 &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>
2591</pre>
2592
2593<h5>Overview:</h5>
2594
2595<p>
2596The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
2597from two input vectors, returning a vector of the same type.
2598</p>
2599
2600<h5>Arguments:</h5>
2601
2602<p>
2603The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
2604with types that match each other and types that match the result of the
2605instruction. The third argument is a shuffle mask, which has the same number
2606of elements as the other vector type, but whose element type is always 'i32'.
2607</p>
2608
2609<p>
2610The shuffle mask operand is required to be a constant vector with either
2611constant integer or undef values.
2612</p>
2613
2614<h5>Semantics:</h5>
2615
2616<p>
2617The elements of the two input vectors are numbered from left to right across
2618both of the vectors. The shuffle mask operand specifies, for each element of
2619the result vector, which element of the two input registers the result element
2620gets. The element selector may be undef (meaning "don't care") and the second
2621operand may be undef if performing a shuffle from only one vector.
2622</p>
2623
2624<h5>Example:</h5>
2625
2626<pre>
2627 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; %v2,
2628 &lt;4 x i32&gt; &lt;i32 0, i32 4, i32 1, i32 5&gt; <i>; yields &lt;4 x i32&gt;</i>
2629 %result = shufflevector &lt;4 x i32&gt; %v1, &lt;4 x i32&gt; undef,
2630 &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.
2631</pre>
2632</div>
2633
2634
2635<!-- ======================================================================= -->
2636<div class="doc_subsection">
2637 <a name="memoryops">Memory Access and Addressing Operations</a>
2638</div>
2639
2640<div class="doc_text">
2641
2642<p>A key design point of an SSA-based representation is how it
2643represents memory. In LLVM, no memory locations are in SSA form, which
2644makes things very simple. This section describes how to read, write,
2645allocate, and free memory in LLVM.</p>
2646
2647</div>
2648
2649<!-- _______________________________________________________________________ -->
2650<div class="doc_subsubsection">
2651 <a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
2652</div>
2653
2654<div class="doc_text">
2655
2656<h5>Syntax:</h5>
2657
2658<pre>
2659 &lt;result&gt; = malloc &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
2660</pre>
2661
2662<h5>Overview:</h5>
2663
2664<p>The '<tt>malloc</tt>' instruction allocates memory from the system
Christopher Lambcfe00962007-12-17 01:00:21 +0000[diff] [blame]2665heap and returns a pointer to it. The object is always allocated in the generic
2666address space (address space zero).</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2667
2668<h5>Arguments:</h5>
2669
2670<p>The '<tt>malloc</tt>' instruction allocates
2671<tt>sizeof(&lt;type&gt;)*NumElements</tt>
2672bytes of memory from the operating system and returns a pointer of the
2673appropriate type to the program. If "NumElements" is specified, it is the
2674number of elements allocated. If an alignment is specified, the value result
2675of the allocation is guaranteed to be aligned to at least that boundary. If
2676not specified, or if zero, the target can choose to align the allocation on any
2677convenient boundary.</p>
2678
2679<p>'<tt>type</tt>' must be a sized type.</p>
2680
2681<h5>Semantics:</h5>
2682
2683<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
2684a pointer is returned.</p>
2685
2686<h5>Example:</h5>
2687
2688<pre>
2689 %array = malloc [4 x i8 ] <i>; yields {[%4 x i8]*}:array</i>
2690
2691 %size = <a href="#i_add">add</a> i32 2, 2 <i>; yields {i32}:size = i32 4</i>
2692 %array1 = malloc i8, i32 4 <i>; yields {i8*}:array1</i>
2693 %array2 = malloc [12 x i8], i32 %size <i>; yields {[12 x i8]*}:array2</i>
2694 %array3 = malloc i32, i32 4, align 1024 <i>; yields {i32*}:array3</i>
2695 %array4 = malloc i32, align 1024 <i>; yields {i32*}:array4</i>
2696</pre>
2697</div>
2698
2699<!-- _______________________________________________________________________ -->
2700<div class="doc_subsubsection">
2701 <a name="i_free">'<tt>free</tt>' Instruction</a>
2702</div>
2703
2704<div class="doc_text">
2705
2706<h5>Syntax:</h5>
2707
2708<pre>
2709 free &lt;type&gt; &lt;value&gt; <i>; yields {void}</i>
2710</pre>
2711
2712<h5>Overview:</h5>
2713
2714<p>The '<tt>free</tt>' instruction returns memory back to the unused
2715memory heap to be reallocated in the future.</p>
2716
2717<h5>Arguments:</h5>
2718
2719<p>'<tt>value</tt>' shall be a pointer value that points to a value
2720that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
2721instruction.</p>
2722
2723<h5>Semantics:</h5>
2724
2725<p>Access to the memory pointed to by the pointer is no longer defined
2726after this instruction executes.</p>
2727
2728<h5>Example:</h5>
2729
2730<pre>
2731 %array = <a href="#i_malloc">malloc</a> [4 x i8] <i>; yields {[4 x i8]*}:array</i>
2732 free [4 x i8]* %array
2733</pre>
2734</div>
2735
2736<!-- _______________________________________________________________________ -->
2737<div class="doc_subsubsection">
2738 <a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
2739</div>
2740
2741<div class="doc_text">
2742
2743<h5>Syntax:</h5>
2744
2745<pre>
2746 &lt;result&gt; = alloca &lt;type&gt;[, i32 &lt;NumElements&gt;][, align &lt;alignment&gt;] <i>; yields {type*}:result</i>
2747</pre>
2748
2749<h5>Overview:</h5>
2750
2751<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
2752currently executing function, to be automatically released when this function
Christopher Lambcfe00962007-12-17 01:00:21 +0000[diff] [blame]2753returns to its caller. The object is always allocated in the generic address
2754space (address space zero).</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2755
2756<h5>Arguments:</h5>
2757
2758<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(&lt;type&gt;)*NumElements</tt>
2759bytes of memory on the runtime stack, returning a pointer of the
2760appropriate type to the program. If "NumElements" is specified, it is the
2761number of elements allocated. If an alignment is specified, the value result
2762of the allocation is guaranteed to be aligned to at least that boundary. If
2763not specified, or if zero, the target can choose to align the allocation on any
2764convenient boundary.</p>
2765
2766<p>'<tt>type</tt>' may be any sized type.</p>
2767
2768<h5>Semantics:</h5>
2769
2770<p>Memory is allocated; a pointer is returned. '<tt>alloca</tt>'d
2771memory is automatically released when the function returns. The '<tt>alloca</tt>'
2772instruction is commonly used to represent automatic variables that must
2773have an address available. When the function returns (either with the <tt><a
2774 href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
2775instructions), the memory is reclaimed.</p>
2776
2777<h5>Example:</h5>
2778
2779<pre>
2780 %ptr = alloca i32 <i>; yields {i32*}:ptr</i>
2781 %ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
2782 %ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
2783 %ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
2784</pre>
2785</div>
2786
2787<!-- _______________________________________________________________________ -->
2788<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
2789Instruction</a> </div>
2790<div class="doc_text">
2791<h5>Syntax:</h5>
2792<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>
2793<h5>Overview:</h5>
2794<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
2795<h5>Arguments:</h5>
2796<p>The argument to the '<tt>load</tt>' instruction specifies the memory
2797address from which to load. The pointer must point to a <a
2798 href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
2799marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
2800the number or order of execution of this <tt>load</tt> with other
2801volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
2802instructions. </p>
2803<h5>Semantics:</h5>
2804<p>The location of memory pointed to is loaded.</p>
2805<h5>Examples:</h5>
2806<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
2807 <a
2808 href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
2809 %val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
2810</pre>
2811</div>
2812<!-- _______________________________________________________________________ -->
2813<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
2814Instruction</a> </div>
2815<div class="doc_text">
2816<h5>Syntax:</h5>
2817<pre> store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
2818 volatile store &lt;ty&gt; &lt;value&gt;, &lt;ty&gt;* &lt;pointer&gt;[, align &lt;alignment&gt;] <i>; yields {void}</i>
2819</pre>
2820<h5>Overview:</h5>
2821<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
2822<h5>Arguments:</h5>
2823<p>There are two arguments to the '<tt>store</tt>' instruction: a value
2824to store and an address at which to store it. The type of the '<tt>&lt;pointer&gt;</tt>'
2825operand must be a pointer to the type of the '<tt>&lt;value&gt;</tt>'
2826operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
2827optimizer is not allowed to modify the number or order of execution of
2828this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
2829 href="#i_store">store</a></tt> instructions.</p>
2830<h5>Semantics:</h5>
2831<p>The contents of memory are updated to contain '<tt>&lt;value&gt;</tt>'
2832at the location specified by the '<tt>&lt;pointer&gt;</tt>' operand.</p>
2833<h5>Example:</h5>
2834<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling63ffa142007-10-22 05:10:05 +0000[diff] [blame]2835 store i32 3, i32* %ptr <i>; yields {void}</i>
2836 %val = <a href="#i_load">load</a> i32* %ptr <i>; yields {i32}:val = i32 3</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]2837</pre>
2838</div>
2839
2840<!-- _______________________________________________________________________ -->
2841<div class="doc_subsubsection">
2842 <a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
2843</div>
2844
2845<div class="doc_text">
2846<h5>Syntax:</h5>
2847<pre>
2848 &lt;result&gt; = getelementptr &lt;ty&gt;* &lt;ptrval&gt;{, &lt;ty&gt; &lt;idx&gt;}*
2849</pre>
2850
2851<h5>Overview:</h5>
2852
2853<p>
2854The '<tt>getelementptr</tt>' instruction is used to get the address of a
2855subelement of an aggregate data structure.</p>
2856
2857<h5>Arguments:</h5>
2858
2859<p>This instruction takes a list of integer operands that indicate what
2860elements of the aggregate object to index to. The actual types of the arguments
2861provided depend on the type of the first pointer argument. The
2862'<tt>getelementptr</tt>' instruction is used to index down through the type
2863levels of a structure or to a specific index in an array. When indexing into a
2864structure, only <tt>i32</tt> integer constants are allowed. When indexing
2865into an array or pointer, only integers of 32 or 64 bits are allowed, and will
2866be sign extended to 64-bit values.</p>
2867
2868<p>For example, let's consider a C code fragment and how it gets
2869compiled to LLVM:</p>
2870
2871<div class="doc_code">
2872<pre>
2873struct RT {
2874 char A;
2875 int B[10][20];
2876 char C;
2877};
2878struct ST {
2879 int X;
2880 double Y;
2881 struct RT Z;
2882};
2883
2884int *foo(struct ST *s) {
2885 return &amp;s[1].Z.B[5][13];
2886}
2887</pre>
2888</div>
2889
2890<p>The LLVM code generated by the GCC frontend is:</p>
2891
2892<div class="doc_code">
2893<pre>
2894%RT = type { i8 , [10 x [20 x i32]], i8 }
2895%ST = type { i32, double, %RT }
2896
2897define i32* %foo(%ST* %s) {
2898entry:
2899 %reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
2900 ret i32* %reg
2901}
2902</pre>
2903</div>
2904
2905<h5>Semantics:</h5>
2906
2907<p>The index types specified for the '<tt>getelementptr</tt>' instruction depend
2908on the pointer type that is being indexed into. <a href="#t_pointer">Pointer</a>
2909and <a href="#t_array">array</a> types can use a 32-bit or 64-bit
2910<a href="#t_integer">integer</a> type but the value will always be sign extended
2911to 64-bits. <a href="#t_struct">Structure</a> types require <tt>i32</tt>
2912<b>constants</b>.</p>
2913
2914<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
2915type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
2916}</tt>' type, a structure. The second index indexes into the third element of
2917the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
2918i8 }</tt>' type, another structure. The third index indexes into the second
2919element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
2920array. The two dimensions of the array are subscripted into, yielding an
2921'<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
2922to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
2923
2924<p>Note that it is perfectly legal to index partially through a
2925structure, returning a pointer to an inner element. Because of this,
2926the LLVM code for the given testcase is equivalent to:</p>
2927
2928<pre>
2929 define i32* %foo(%ST* %s) {
2930 %t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
2931 %t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
2932 %t3 = getelementptr %RT* %t2, i32 0, i32 1 <i>; yields [10 x [20 x i32]]*:%t3</i>
2933 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
2934 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
2935 ret i32* %t5
2936 }
2937</pre>
2938
2939<p>Note that it is undefined to access an array out of bounds: array and
2940pointer indexes must always be within the defined bounds of the array type.
2941The one exception for this rules is zero length arrays. These arrays are
2942defined to be accessible as variable length arrays, which requires access
2943beyond the zero'th element.</p>
2944
2945<p>The getelementptr instruction is often confusing. For some more insight
2946into how it works, see <a href="GetElementPtr.html">the getelementptr
2947FAQ</a>.</p>
2948
2949<h5>Example:</h5>
2950
2951<pre>
2952 <i>; yields [12 x i8]*:aptr</i>
2953 %aptr = getelementptr {i32, [12 x i8]}* %sptr, i64 0, i32 1
2954</pre>
2955</div>
2956
2957<!-- ======================================================================= -->
2958<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
2959</div>
2960<div class="doc_text">
2961<p>The instructions in this category are the conversion instructions (casting)
2962which all take a single operand and a type. They perform various bit conversions
2963on the operand.</p>
2964</div>
2965
2966<!-- _______________________________________________________________________ -->
2967<div class="doc_subsubsection">
2968 <a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
2969</div>
2970<div class="doc_text">
2971
2972<h5>Syntax:</h5>
2973<pre>
2974 &lt;result&gt; = trunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
2975</pre>
2976
2977<h5>Overview:</h5>
2978<p>
2979The '<tt>trunc</tt>' instruction truncates its operand to the type <tt>ty2</tt>.
2980</p>
2981
2982<h5>Arguments:</h5>
2983<p>
2984The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
2985be an <a href="#t_integer">integer</a> type, and a type that specifies the size
2986and type of the result, which must be an <a href="#t_integer">integer</a>
2987type. The bit size of <tt>value</tt> must be larger than the bit size of
2988<tt>ty2</tt>. Equal sized types are not allowed.</p>
2989
2990<h5>Semantics:</h5>
2991<p>
2992The '<tt>trunc</tt>' instruction truncates the high order bits in <tt>value</tt>
2993and converts the remaining bits to <tt>ty2</tt>. Since the source size must be
2994larger than the destination size, <tt>trunc</tt> cannot be a <i>no-op cast</i>.
2995It will always truncate bits.</p>
2996
2997<h5>Example:</h5>
2998<pre>
2999 %X = trunc i32 257 to i8 <i>; yields i8:1</i>
3000 %Y = trunc i32 123 to i1 <i>; yields i1:true</i>
3001 %Y = trunc i32 122 to i1 <i>; yields i1:false</i>
3002</pre>
3003</div>
3004
3005<!-- _______________________________________________________________________ -->
3006<div class="doc_subsubsection">
3007 <a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
3008</div>
3009<div class="doc_text">
3010
3011<h5>Syntax:</h5>
3012<pre>
3013 &lt;result&gt; = zext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3014</pre>
3015
3016<h5>Overview:</h5>
3017<p>The '<tt>zext</tt>' instruction zero extends its operand to type
3018<tt>ty2</tt>.</p>
3019
3020
3021<h5>Arguments:</h5>
3022<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
3023<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3024also be of <a href="#t_integer">integer</a> type. The bit size of the
3025<tt>value</tt> must be smaller than the bit size of the destination type,
3026<tt>ty2</tt>.</p>
3027
3028<h5>Semantics:</h5>
3029<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
3030bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
3031
3032<p>When zero extending from i1, the result will always be either 0 or 1.</p>
3033
3034<h5>Example:</h5>
3035<pre>
3036 %X = zext i32 257 to i64 <i>; yields i64:257</i>
3037 %Y = zext i1 true to i32 <i>; yields i32:1</i>
3038</pre>
3039</div>
3040
3041<!-- _______________________________________________________________________ -->
3042<div class="doc_subsubsection">
3043 <a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
3044</div>
3045<div class="doc_text">
3046
3047<h5>Syntax:</h5>
3048<pre>
3049 &lt;result&gt; = sext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3050</pre>
3051
3052<h5>Overview:</h5>
3053<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
3054
3055<h5>Arguments:</h5>
3056<p>
3057The '<tt>sext</tt>' instruction takes a value to cast, which must be of
3058<a href="#t_integer">integer</a> type, and a type to cast it to, which must
3059also be of <a href="#t_integer">integer</a> type. The bit size of the
3060<tt>value</tt> must be smaller than the bit size of the destination type,
3061<tt>ty2</tt>.</p>
3062
3063<h5>Semantics:</h5>
3064<p>
3065The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
3066bit (highest order bit) of the <tt>value</tt> until it reaches the bit size of
3067the type <tt>ty2</tt>.</p>
3068
3069<p>When sign extending from i1, the extension always results in -1 or 0.</p>
3070
3071<h5>Example:</h5>
3072<pre>
3073 %X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
3074 %Y = sext i1 true to i32 <i>; yields i32:-1</i>
3075</pre>
3076</div>
3077
3078<!-- _______________________________________________________________________ -->
3079<div class="doc_subsubsection">
3080 <a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
3081</div>
3082
3083<div class="doc_text">
3084
3085<h5>Syntax:</h5>
3086
3087<pre>
3088 &lt;result&gt; = fptrunc &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3089</pre>
3090
3091<h5>Overview:</h5>
3092<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
3093<tt>ty2</tt>.</p>
3094
3095
3096<h5>Arguments:</h5>
3097<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
3098 point</a> value to cast and a <a href="#t_floating">floating point</a> type to
3099cast it to. The size of <tt>value</tt> must be larger than the size of
3100<tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
3101<i>no-op cast</i>.</p>
3102
3103<h5>Semantics:</h5>
3104<p> The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
3105<a href="#t_floating">floating point</a> type to a smaller
3106<a href="#t_floating">floating point</a> type. If the value cannot fit within
3107the destination type, <tt>ty2</tt>, then the results are undefined.</p>
3108
3109<h5>Example:</h5>
3110<pre>
3111 %X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
3112 %Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
3113</pre>
3114</div>
3115
3116<!-- _______________________________________________________________________ -->
3117<div class="doc_subsubsection">
3118 <a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
3119</div>
3120<div class="doc_text">
3121
3122<h5>Syntax:</h5>
3123<pre>
3124 &lt;result&gt; = fpext &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3125</pre>
3126
3127<h5>Overview:</h5>
3128<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
3129floating point value.</p>
3130
3131<h5>Arguments:</h5>
3132<p>The '<tt>fpext</tt>' instruction takes a
3133<a href="#t_floating">floating point</a> <tt>value</tt> to cast,
3134and a <a href="#t_floating">floating point</a> type to cast it to. The source
3135type must be smaller than the destination type.</p>
3136
3137<h5>Semantics:</h5>
3138<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
3139<a href="#t_floating">floating point</a> type to a larger
3140<a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
3141used to make a <i>no-op cast</i> because it always changes bits. Use
3142<tt>bitcast</tt> to make a <i>no-op cast</i> for a floating point cast.</p>
3143
3144<h5>Example:</h5>
3145<pre>
3146 %X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
3147 %Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
3148</pre>
3149</div>
3150
3151<!-- _______________________________________________________________________ -->
3152<div class="doc_subsubsection">
3153 <a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
3154</div>
3155<div class="doc_text">
3156
3157<h5>Syntax:</h5>
3158<pre>
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]3159 &lt;result&gt; = fptoui &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3160</pre>
3161
3162<h5>Overview:</h5>
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]3163<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3164unsigned integer equivalent of type <tt>ty2</tt>.
3165</p>
3166
3167<h5>Arguments:</h5>
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]3168<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]3169scalar or vector <a href="#t_floating">floating point</a> value, and a type
3170to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3171type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3172vector integer type with the same number of elements as <tt>ty</tt></p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3173
3174<h5>Semantics:</h5>
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]3175<p> The '<tt>fptoui</tt>' instruction converts its
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3176<a href="#t_floating">floating point</a> operand into the nearest (rounding
3177towards zero) unsigned integer value. If the value cannot fit in <tt>ty2</tt>,
3178the results are undefined.</p>
3179
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3180<h5>Example:</h5>
3181<pre>
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]3182 %X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner681f1e82007-09-22 03:17:52 +0000[diff] [blame]3183 %Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Reid Spencere6adee82007-07-31 14:40:14 +0000[diff] [blame]3184 %X = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3185</pre>
3186</div>
3187
3188<!-- _______________________________________________________________________ -->
3189<div class="doc_subsubsection">
3190 <a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
3191</div>
3192<div class="doc_text">
3193
3194<h5>Syntax:</h5>
3195<pre>
3196 &lt;result&gt; = fptosi &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3197</pre>
3198
3199<h5>Overview:</h5>
3200<p>The '<tt>fptosi</tt>' instruction converts
3201<a href="#t_floating">floating point</a> <tt>value</tt> to type <tt>ty2</tt>.
3202</p>
3203
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3204<h5>Arguments:</h5>
3205<p> The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]3206scalar or vector <a href="#t_floating">floating point</a> value, and a type
3207to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
3208type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
3209vector integer type with the same number of elements as <tt>ty</tt></p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3210
3211<h5>Semantics:</h5>
3212<p>The '<tt>fptosi</tt>' instruction converts its
3213<a href="#t_floating">floating point</a> operand into the nearest (rounding
3214towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
3215the results are undefined.</p>
3216
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3217<h5>Example:</h5>
3218<pre>
3219 %X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner681f1e82007-09-22 03:17:52 +0000[diff] [blame]3220 %Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3221 %X = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
3222</pre>
3223</div>
3224
3225<!-- _______________________________________________________________________ -->
3226<div class="doc_subsubsection">
3227 <a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
3228</div>
3229<div class="doc_text">
3230
3231<h5>Syntax:</h5>
3232<pre>
3233 &lt;result&gt; = uitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3234</pre>
3235
3236<h5>Overview:</h5>
3237<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
3238integer and converts that value to the <tt>ty2</tt> type.</p>
3239
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3240<h5>Arguments:</h5>
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]3241<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be a
3242scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3243to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3244type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3245floating point type with the same number of elements as <tt>ty</tt></p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3246
3247<h5>Semantics:</h5>
3248<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
3249integer quantity and converts it to the corresponding floating point value. If
3250the value cannot fit in the floating point value, the results are undefined.</p>
3251
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3252<h5>Example:</h5>
3253<pre>
3254 %X = uitofp i32 257 to float <i>; yields float:257.0</i>
3255 %Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
3256</pre>
3257</div>
3258
3259<!-- _______________________________________________________________________ -->
3260<div class="doc_subsubsection">
3261 <a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
3262</div>
3263<div class="doc_text">
3264
3265<h5>Syntax:</h5>
3266<pre>
3267 &lt;result&gt; = sitofp &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3268</pre>
3269
3270<h5>Overview:</h5>
3271<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed
3272integer and converts that value to the <tt>ty2</tt> type.</p>
3273
3274<h5>Arguments:</h5>
Nate Begeman78246ca2007-11-17 03:58:34 +0000[diff] [blame]3275<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be a
3276scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
3277to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
3278type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
3279floating point type with the same number of elements as <tt>ty</tt></p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3280
3281<h5>Semantics:</h5>
3282<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed
3283integer quantity and converts it to the corresponding floating point value. If
3284the value cannot fit in the floating point value, the results are undefined.</p>
3285
3286<h5>Example:</h5>
3287<pre>
3288 %X = sitofp i32 257 to float <i>; yields float:257.0</i>
3289 %Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
3290</pre>
3291</div>
3292
3293<!-- _______________________________________________________________________ -->
3294<div class="doc_subsubsection">
3295 <a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
3296</div>
3297<div class="doc_text">
3298
3299<h5>Syntax:</h5>
3300<pre>
3301 &lt;result&gt; = ptrtoint &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3302</pre>
3303
3304<h5>Overview:</h5>
3305<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
3306the integer type <tt>ty2</tt>.</p>
3307
3308<h5>Arguments:</h5>
3309<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
3310must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
3311<tt>ty2</tt>, which must be an <a href="#t_integer">integer</a> type.
3312
3313<h5>Semantics:</h5>
3314<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
3315<tt>ty2</tt> by interpreting the pointer value as an integer and either
3316truncating or zero extending that value to the size of the integer type. If
3317<tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
3318<tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
3319are the same size, then nothing is done (<i>no-op cast</i>) other than a type
3320change.</p>
3321
3322<h5>Example:</h5>
3323<pre>
3324 %X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
3325 %Y = ptrtoint i32* %x to i64 <i>; yields zero extension on 32-bit architecture</i>
3326</pre>
3327</div>
3328
3329<!-- _______________________________________________________________________ -->
3330<div class="doc_subsubsection">
3331 <a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
3332</div>
3333<div class="doc_text">
3334
3335<h5>Syntax:</h5>
3336<pre>
3337 &lt;result&gt; = inttoptr &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3338</pre>
3339
3340<h5>Overview:</h5>
3341<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to
3342a pointer type, <tt>ty2</tt>.</p>
3343
3344<h5>Arguments:</h5>
3345<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
3346value to cast, and a type to cast it to, which must be a
3347<a href="#t_pointer">pointer</a> type.
3348
3349<h5>Semantics:</h5>
3350<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
3351<tt>ty2</tt> by applying either a zero extension or a truncation depending on
3352the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
3353size of a pointer then a truncation is done. If <tt>value</tt> is smaller than
3354the size of a pointer then a zero extension is done. If they are the same size,
3355nothing is done (<i>no-op cast</i>).</p>
3356
3357<h5>Example:</h5>
3358<pre>
3359 %X = inttoptr i32 255 to i32* <i>; yields zero extension on 64-bit architecture</i>
3360 %X = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
3361 %Y = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
3362</pre>
3363</div>
3364
3365<!-- _______________________________________________________________________ -->
3366<div class="doc_subsubsection">
3367 <a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
3368</div>
3369<div class="doc_text">
3370
3371<h5>Syntax:</h5>
3372<pre>
3373 &lt;result&gt; = bitcast &lt;ty&gt; &lt;value&gt; to &lt;ty2&gt; <i>; yields ty2</i>
3374</pre>
3375
3376<h5>Overview:</h5>
3377<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
3378<tt>ty2</tt> without changing any bits.</p>
3379
3380<h5>Arguments:</h5>
3381<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be
3382a first class value, and a type to cast it to, which must also be a <a
3383 href="#t_firstclass">first class</a> type. The bit sizes of <tt>value</tt>
3384and the destination type, <tt>ty2</tt>, must be identical. If the source
3385type is a pointer, the destination type must also be a pointer.</p>
3386
3387<h5>Semantics:</h5>
3388<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
3389<tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
3390this conversion. The conversion is done as if the <tt>value</tt> had been
3391stored to memory and read back as type <tt>ty2</tt>. Pointer types may only be
3392converted to other pointer types with this instruction. To convert pointers to
3393other types, use the <a href="#i_inttoptr">inttoptr</a> or
3394<a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
3395
3396<h5>Example:</h5>
3397<pre>
3398 %X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
3399 %Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
3400 %Z = bitcast <2xint> %V to i64; <i>; yields i64: %V</i>
3401</pre>
3402</div>
3403
3404<!-- ======================================================================= -->
3405<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
3406<div class="doc_text">
3407<p>The instructions in this category are the "miscellaneous"
3408instructions, which defy better classification.</p>
3409</div>
3410
3411<!-- _______________________________________________________________________ -->
3412<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
3413</div>
3414<div class="doc_text">
3415<h5>Syntax:</h5>
3416<pre> &lt;result&gt; = icmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {i1}:result</i>
3417</pre>
3418<h5>Overview:</h5>
3419<p>The '<tt>icmp</tt>' instruction returns a boolean value based on comparison
3420of its two integer operands.</p>
3421<h5>Arguments:</h5>
3422<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
3423the condition code indicating the kind of comparison to perform. It is not
3424a value, just a keyword. The possible condition code are:
3425<ol>
3426 <li><tt>eq</tt>: equal</li>
3427 <li><tt>ne</tt>: not equal </li>
3428 <li><tt>ugt</tt>: unsigned greater than</li>
3429 <li><tt>uge</tt>: unsigned greater or equal</li>
3430 <li><tt>ult</tt>: unsigned less than</li>
3431 <li><tt>ule</tt>: unsigned less or equal</li>
3432 <li><tt>sgt</tt>: signed greater than</li>
3433 <li><tt>sge</tt>: signed greater or equal</li>
3434 <li><tt>slt</tt>: signed less than</li>
3435 <li><tt>sle</tt>: signed less or equal</li>
3436</ol>
3437<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
3438<a href="#t_pointer">pointer</a> typed. They must also be identical types.</p>
3439<h5>Semantics:</h5>
3440<p>The '<tt>icmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3441the condition code given as <tt>cond</tt>. The comparison performed always
3442yields a <a href="#t_primitive">i1</a> result, as follows:
3443<ol>
3444 <li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
3445 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3446 </li>
3447 <li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
3448 <tt>false</tt> otherwise. No sign interpretation is necessary or performed.
3449 <li><tt>ugt</tt>: interprets the operands as unsigned values and yields
3450 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3451 <li><tt>uge</tt>: interprets the operands as unsigned values and yields
3452 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3453 <li><tt>ult</tt>: interprets the operands as unsigned values and yields
3454 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3455 <li><tt>ule</tt>: interprets the operands as unsigned values and yields
3456 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3457 <li><tt>sgt</tt>: interprets the operands as signed values and yields
3458 <tt>true</tt> if <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3459 <li><tt>sge</tt>: interprets the operands as signed values and yields
3460 <tt>true</tt> if <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3461 <li><tt>slt</tt>: interprets the operands as signed values and yields
3462 <tt>true</tt> if <tt>var1</tt> is less than <tt>var2</tt>.</li>
3463 <li><tt>sle</tt>: interprets the operands as signed values and yields
3464 <tt>true</tt> if <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3465</ol>
3466<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
3467values are compared as if they were integers.</p>
3468
3469<h5>Example:</h5>
3470<pre> &lt;result&gt; = icmp eq i32 4, 5 <i>; yields: result=false</i>
3471 &lt;result&gt; = icmp ne float* %X, %X <i>; yields: result=false</i>
3472 &lt;result&gt; = icmp ult i16 4, 5 <i>; yields: result=true</i>
3473 &lt;result&gt; = icmp sgt i16 4, 5 <i>; yields: result=false</i>
3474 &lt;result&gt; = icmp ule i16 -4, 5 <i>; yields: result=false</i>
3475 &lt;result&gt; = icmp sge i16 4, 5 <i>; yields: result=false</i>
3476</pre>
3477</div>
3478
3479<!-- _______________________________________________________________________ -->
3480<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
3481</div>
3482<div class="doc_text">
3483<h5>Syntax:</h5>
3484<pre> &lt;result&gt; = fcmp &lt;cond&gt; &lt;ty&gt; &lt;var1&gt;, &lt;var2&gt; <i>; yields {i1}:result</i>
3485</pre>
3486<h5>Overview:</h5>
3487<p>The '<tt>fcmp</tt>' instruction returns a boolean value based on comparison
3488of its floating point operands.</p>
3489<h5>Arguments:</h5>
3490<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
3491the condition code indicating the kind of comparison to perform. It is not
3492a value, just a keyword. The possible condition code are:
3493<ol>
3494 <li><tt>false</tt>: no comparison, always returns false</li>
3495 <li><tt>oeq</tt>: ordered and equal</li>
3496 <li><tt>ogt</tt>: ordered and greater than </li>
3497 <li><tt>oge</tt>: ordered and greater than or equal</li>
3498 <li><tt>olt</tt>: ordered and less than </li>
3499 <li><tt>ole</tt>: ordered and less than or equal</li>
3500 <li><tt>one</tt>: ordered and not equal</li>
3501 <li><tt>ord</tt>: ordered (no nans)</li>
3502 <li><tt>ueq</tt>: unordered or equal</li>
3503 <li><tt>ugt</tt>: unordered or greater than </li>
3504 <li><tt>uge</tt>: unordered or greater than or equal</li>
3505 <li><tt>ult</tt>: unordered or less than </li>
3506 <li><tt>ule</tt>: unordered or less than or equal</li>
3507 <li><tt>une</tt>: unordered or not equal</li>
3508 <li><tt>uno</tt>: unordered (either nans)</li>
3509 <li><tt>true</tt>: no comparison, always returns true</li>
3510</ol>
3511<p><i>Ordered</i> means that neither operand is a QNAN while
3512<i>unordered</i> means that either operand may be a QNAN.</p>
3513<p>The <tt>val1</tt> and <tt>val2</tt> arguments must be
3514<a href="#t_floating">floating point</a> typed. They must have identical
3515types.</p>
3516<h5>Semantics:</h5>
3517<p>The '<tt>fcmp</tt>' compares <tt>var1</tt> and <tt>var2</tt> according to
3518the condition code given as <tt>cond</tt>. The comparison performed always
3519yields a <a href="#t_primitive">i1</a> result, as follows:
3520<ol>
3521 <li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
3522 <li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
3523 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
3524 <li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
3525 <tt>var1</tt> is greather than <tt>var2</tt>.</li>
3526 <li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
3527 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3528 <li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
3529 <tt>var1</tt> is less than <tt>var2</tt>.</li>
3530 <li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
3531 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3532 <li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
3533 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
3534 <li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
3535 <li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
3536 <tt>var1</tt> is equal to <tt>var2</tt>.</li>
3537 <li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
3538 <tt>var1</tt> is greater than <tt>var2</tt>.</li>
3539 <li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
3540 <tt>var1</tt> is greater than or equal to <tt>var2</tt>.</li>
3541 <li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
3542 <tt>var1</tt> is less than <tt>var2</tt>.</li>
3543 <li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
3544 <tt>var1</tt> is less than or equal to <tt>var2</tt>.</li>
3545 <li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
3546 <tt>var1</tt> is not equal to <tt>var2</tt>.</li>
3547 <li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
3548 <li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
3549</ol>
3550
3551<h5>Example:</h5>
3552<pre> &lt;result&gt; = fcmp oeq float 4.0, 5.0 <i>; yields: result=false</i>
3553 &lt;result&gt; = icmp one float 4.0, 5.0 <i>; yields: result=true</i>
3554 &lt;result&gt; = icmp olt float 4.0, 5.0 <i>; yields: result=true</i>
3555 &lt;result&gt; = icmp ueq double 1.0, 2.0 <i>; yields: result=false</i>
3556</pre>
3557</div>
3558
3559<!-- _______________________________________________________________________ -->
3560<div class="doc_subsubsection"> <a name="i_phi">'<tt>phi</tt>'
3561Instruction</a> </div>
3562<div class="doc_text">
3563<h5>Syntax:</h5>
3564<pre> &lt;result&gt; = phi &lt;ty&gt; [ &lt;val0&gt;, &lt;label0&gt;], ...<br></pre>
3565<h5>Overview:</h5>
3566<p>The '<tt>phi</tt>' instruction is used to implement the &#966; node in
3567the SSA graph representing the function.</p>
3568<h5>Arguments:</h5>
3569<p>The type of the incoming values is specified with the first type
3570field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
3571as arguments, with one pair for each predecessor basic block of the
3572current block. Only values of <a href="#t_firstclass">first class</a>
3573type may be used as the value arguments to the PHI node. Only labels
3574may be used as the label arguments.</p>
3575<p>There must be no non-phi instructions between the start of a basic
3576block and the PHI instructions: i.e. PHI instructions must be first in
3577a basic block.</p>
3578<h5>Semantics:</h5>
3579<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
3580specified by the pair corresponding to the predecessor basic block that executed
3581just prior to the current block.</p>
3582<h5>Example:</h5>
3583<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>
3584</div>
3585
3586<!-- _______________________________________________________________________ -->
3587<div class="doc_subsubsection">
3588 <a name="i_select">'<tt>select</tt>' Instruction</a>
3589</div>
3590
3591<div class="doc_text">
3592
3593<h5>Syntax:</h5>
3594
3595<pre>
3596 &lt;result&gt; = select i1 &lt;cond&gt;, &lt;ty&gt; &lt;val1&gt;, &lt;ty&gt; &lt;val2&gt; <i>; yields ty</i>
3597</pre>
3598
3599<h5>Overview:</h5>
3600
3601<p>
3602The '<tt>select</tt>' instruction is used to choose one value based on a
3603condition, without branching.
3604</p>
3605
3606
3607<h5>Arguments:</h5>
3608
3609<p>
3610The '<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.
3611</p>
3612
3613<h5>Semantics:</h5>
3614
3615<p>
3616If the boolean condition evaluates to true, the instruction returns the first
3617value argument; otherwise, it returns the second value argument.
3618</p>
3619
3620<h5>Example:</h5>
3621
3622<pre>
3623 %X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
3624</pre>
3625</div>
3626
3627
3628<!-- _______________________________________________________________________ -->
3629<div class="doc_subsubsection">
3630 <a name="i_call">'<tt>call</tt>' Instruction</a>
3631</div>
3632
3633<div class="doc_text">
3634
3635<h5>Syntax:</h5>
3636<pre>
Nick Lewycky93082fc2007-09-08 13:57:50 +0000[diff] [blame]3637 &lt;result&gt; = [tail] call [<a href="#callingconv">cconv</a>] &lt;ty&gt; [&lt;fnty&gt;*] &lt;fnptrval&gt;(&lt;param list&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3638</pre>
3639
3640<h5>Overview:</h5>
3641
3642<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
3643
3644<h5>Arguments:</h5>
3645
3646<p>This instruction requires several arguments:</p>
3647
3648<ol>
3649 <li>
3650 <p>The optional "tail" marker indicates whether the callee function accesses
3651 any allocas or varargs in the caller. If the "tail" marker is present, the
3652 function call is eligible for tail call optimization. Note that calls may
3653 be marked "tail" even if they do not occur before a <a
3654 href="#i_ret"><tt>ret</tt></a> instruction.
3655 </li>
3656 <li>
3657 <p>The optional "cconv" marker indicates which <a href="#callingconv">calling
3658 convention</a> the call should use. If none is specified, the call defaults
3659 to using C calling conventions.
3660 </li>
3661 <li>
Nick Lewycky93082fc2007-09-08 13:57:50 +0000[diff] [blame]3662 <p>'<tt>ty</tt>': the type of the call instruction itself which is also
3663 the type of the return value. Functions that return no value are marked
3664 <tt><a href="#t_void">void</a></tt>.</p>
3665 </li>
3666 <li>
3667 <p>'<tt>fnty</tt>': shall be the signature of the pointer to function
3668 value being invoked. The argument types must match the types implied by
3669 this signature. This type can be omitted if the function is not varargs
3670 and if the function type does not return a pointer to a function.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3671 </li>
3672 <li>
3673 <p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
3674 be invoked. In most cases, this is a direct function invocation, but
3675 indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
3676 to function value.</p>
3677 </li>
3678 <li>
3679 <p>'<tt>function args</tt>': argument list whose types match the
3680 function signature argument types. All arguments must be of
3681 <a href="#t_firstclass">first class</a> type. If the function signature
3682 indicates the function accepts a variable number of arguments, the extra
3683 arguments can be specified.</p>
3684 </li>
3685</ol>
3686
3687<h5>Semantics:</h5>
3688
3689<p>The '<tt>call</tt>' instruction is used to cause control flow to
3690transfer to a specified function, with its incoming arguments bound to
3691the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
3692instruction in the called function, control flow continues with the
3693instruction after the function call, and the return value of the
3694function is bound to the result argument. This is a simpler case of
3695the <a href="#i_invoke">invoke</a> instruction.</p>
3696
3697<h5>Example:</h5>
3698
3699<pre>
Nick Lewycky93082fc2007-09-08 13:57:50 +0000[diff] [blame]3700 %retval = call i32 @test(i32 %argc)
3701 call i32 (i8 *, ...)* @printf(i8 * %msg, i32 12, i8 42);
3702 %X = tail call i32 @foo()
3703 %Y = tail call <a href="#callingconv">fastcc</a> i32 @foo()
3704 %Z = call void %foo(i8 97 signext)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3705</pre>
3706
3707</div>
3708
3709<!-- _______________________________________________________________________ -->
3710<div class="doc_subsubsection">
3711 <a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
3712</div>
3713
3714<div class="doc_text">
3715
3716<h5>Syntax:</h5>
3717
3718<pre>
3719 &lt;resultval&gt; = va_arg &lt;va_list*&gt; &lt;arglist&gt;, &lt;argty&gt;
3720</pre>
3721
3722<h5>Overview:</h5>
3723
3724<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
3725the "variable argument" area of a function call. It is used to implement the
3726<tt>va_arg</tt> macro in C.</p>
3727
3728<h5>Arguments:</h5>
3729
3730<p>This instruction takes a <tt>va_list*</tt> value and the type of
3731the argument. It returns a value of the specified argument type and
3732increments the <tt>va_list</tt> to point to the next argument. The
3733actual type of <tt>va_list</tt> is target specific.</p>
3734
3735<h5>Semantics:</h5>
3736
3737<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
3738type from the specified <tt>va_list</tt> and causes the
3739<tt>va_list</tt> to point to the next argument. For more information,
3740see the variable argument handling <a href="#int_varargs">Intrinsic
3741Functions</a>.</p>
3742
3743<p>It is legal for this instruction to be called in a function which does not
3744take a variable number of arguments, for example, the <tt>vfprintf</tt>
3745function.</p>
3746
3747<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
3748href="#intrinsics">intrinsic function</a> because it takes a type as an
3749argument.</p>
3750
3751<h5>Example:</h5>
3752
3753<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
3754
3755</div>
3756
3757<!-- *********************************************************************** -->
3758<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
3759<!-- *********************************************************************** -->
3760
3761<div class="doc_text">
3762
3763<p>LLVM supports the notion of an "intrinsic function". These functions have
3764well known names and semantics and are required to follow certain restrictions.
3765Overall, these intrinsics represent an extension mechanism for the LLVM
3766language that does not require changing all of the transformations in LLVM when
3767adding to the language (or the bitcode reader/writer, the parser, etc...).</p>
3768
3769<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
3770prefix is reserved in LLVM for intrinsic names; thus, function names may not
3771begin with this prefix. Intrinsic functions must always be external functions:
3772you cannot define the body of intrinsic functions. Intrinsic functions may
3773only be used in call or invoke instructions: it is illegal to take the address
3774of an intrinsic function. Additionally, because intrinsic functions are part
3775of the LLVM language, it is required if any are added that they be documented
3776here.</p>
3777
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]3778<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents
3779a family of functions that perform the same operation but on different data
3780types. Because LLVM can represent over 8 million different integer types,
3781overloading is used commonly to allow an intrinsic function to operate on any
3782integer type. One or more of the argument types or the result type can be
3783overloaded to accept any integer type. Argument types may also be defined as
3784exactly matching a previous argument's type or the result type. This allows an
3785intrinsic function which accepts multiple arguments, but needs all of them to
3786be of the same type, to only be overloaded with respect to a single argument or
3787the result.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3788
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]3789<p>Overloaded intrinsics will have the names of its overloaded argument types
3790encoded into its function name, each preceded by a period. Only those types
3791which are overloaded result in a name suffix. Arguments whose type is matched
3792against another type do not. For example, the <tt>llvm.ctpop</tt> function can
3793take an integer of any width and returns an integer of exactly the same integer
3794width. This leads to a family of functions such as
3795<tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29 %val)</tt>.
3796Only one type, the return type, is overloaded, and only one type suffix is
3797required. Because the argument's type is matched against the return type, it
3798does not require its own name suffix.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3799
3800<p>To learn how to add an intrinsic function, please see the
3801<a href="ExtendingLLVM.html">Extending LLVM Guide</a>.
3802</p>
3803
3804</div>
3805
3806<!-- ======================================================================= -->
3807<div class="doc_subsection">
3808 <a name="int_varargs">Variable Argument Handling Intrinsics</a>
3809</div>
3810
3811<div class="doc_text">
3812
3813<p>Variable argument support is defined in LLVM with the <a
3814 href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
3815intrinsic functions. These functions are related to the similarly
3816named macros defined in the <tt>&lt;stdarg.h&gt;</tt> header file.</p>
3817
3818<p>All of these functions operate on arguments that use a
3819target-specific value type "<tt>va_list</tt>". The LLVM assembly
3820language reference manual does not define what this type is, so all
3821transformations should be prepared to handle these functions regardless of
3822the type used.</p>
3823
3824<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
3825instruction and the variable argument handling intrinsic functions are
3826used.</p>
3827
3828<div class="doc_code">
3829<pre>
3830define i32 @test(i32 %X, ...) {
3831 ; Initialize variable argument processing
3832 %ap = alloca i8*
3833 %ap2 = bitcast i8** %ap to i8*
3834 call void @llvm.va_start(i8* %ap2)
3835
3836 ; Read a single integer argument
3837 %tmp = va_arg i8** %ap, i32
3838
3839 ; Demonstrate usage of llvm.va_copy and llvm.va_end
3840 %aq = alloca i8*
3841 %aq2 = bitcast i8** %aq to i8*
3842 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
3843 call void @llvm.va_end(i8* %aq2)
3844
3845 ; Stop processing of arguments.
3846 call void @llvm.va_end(i8* %ap2)
3847 ret i32 %tmp
3848}
3849
3850declare void @llvm.va_start(i8*)
3851declare void @llvm.va_copy(i8*, i8*)
3852declare void @llvm.va_end(i8*)
3853</pre>
3854</div>
3855
3856</div>
3857
3858<!-- _______________________________________________________________________ -->
3859<div class="doc_subsubsection">
3860 <a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
3861</div>
3862
3863
3864<div class="doc_text">
3865<h5>Syntax:</h5>
3866<pre> declare void %llvm.va_start(i8* &lt;arglist&gt;)<br></pre>
3867<h5>Overview:</h5>
3868<P>The '<tt>llvm.va_start</tt>' intrinsic initializes
3869<tt>*&lt;arglist&gt;</tt> for subsequent use by <tt><a
3870href="#i_va_arg">va_arg</a></tt>.</p>
3871
3872<h5>Arguments:</h5>
3873
3874<P>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
3875
3876<h5>Semantics:</h5>
3877
3878<P>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
3879macro available in C. In a target-dependent way, it initializes the
3880<tt>va_list</tt> element to which the argument points, so that the next call to
3881<tt>va_arg</tt> will produce the first variable argument passed to the function.
3882Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
3883last argument of the function as the compiler can figure that out.</p>
3884
3885</div>
3886
3887<!-- _______________________________________________________________________ -->
3888<div class="doc_subsubsection">
3889 <a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
3890</div>
3891
3892<div class="doc_text">
3893<h5>Syntax:</h5>
3894<pre> declare void @llvm.va_end(i8* &lt;arglist&gt;)<br></pre>
3895<h5>Overview:</h5>
3896
3897<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*&lt;arglist&gt;</tt>,
3898which has been initialized previously with <tt><a href="#int_va_start">llvm.va_start</a></tt>
3899or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
3900
3901<h5>Arguments:</h5>
3902
3903<p>The argument is a pointer to a <tt>va_list</tt> to destroy.</p>
3904
3905<h5>Semantics:</h5>
3906
3907<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
3908macro available in C. In a target-dependent way, it destroys the
3909<tt>va_list</tt> element to which the argument points. Calls to <a
3910href="#int_va_start"><tt>llvm.va_start</tt></a> and <a href="#int_va_copy">
3911<tt>llvm.va_copy</tt></a> must be matched exactly with calls to
3912<tt>llvm.va_end</tt>.</p>
3913
3914</div>
3915
3916<!-- _______________________________________________________________________ -->
3917<div class="doc_subsubsection">
3918 <a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
3919</div>
3920
3921<div class="doc_text">
3922
3923<h5>Syntax:</h5>
3924
3925<pre>
3926 declare void @llvm.va_copy(i8* &lt;destarglist&gt;, i8* &lt;srcarglist&gt;)
3927</pre>
3928
3929<h5>Overview:</h5>
3930
3931<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
3932from the source argument list to the destination argument list.</p>
3933
3934<h5>Arguments:</h5>
3935
3936<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
3937The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
3938
3939
3940<h5>Semantics:</h5>
3941
3942<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
3943macro available in C. In a target-dependent way, it copies the source
3944<tt>va_list</tt> element into the destination <tt>va_list</tt> element. This
3945intrinsic is necessary because the <tt><a href="#int_va_start">
3946llvm.va_start</a></tt> intrinsic may be arbitrarily complex and require, for
3947example, memory allocation.</p>
3948
3949</div>
3950
3951<!-- ======================================================================= -->
3952<div class="doc_subsection">
3953 <a name="int_gc">Accurate Garbage Collection Intrinsics</a>
3954</div>
3955
3956<div class="doc_text">
3957
3958<p>
3959LLVM support for <a href="GarbageCollection.html">Accurate Garbage
3960Collection</a> requires the implementation and generation of these intrinsics.
3961These intrinsics allow identification of <a href="#int_gcroot">GC roots on the
3962stack</a>, as well as garbage collector implementations that require <a
3963href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a> barriers.
3964Front-ends for type-safe garbage collected languages should generate these
3965intrinsics to make use of the LLVM garbage collectors. For more details, see <a
3966href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
3967</p>
Christopher Lambcfe00962007-12-17 01:00:21 +0000[diff] [blame]3968
3969<p>The garbage collection intrinsics only operate on objects in the generic
3970 address space (address space zero).</p>
3971
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3972</div>
3973
3974<!-- _______________________________________________________________________ -->
3975<div class="doc_subsubsection">
3976 <a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
3977</div>
3978
3979<div class="doc_text">
3980
3981<h5>Syntax:</h5>
3982
3983<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]3984 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]3985</pre>
3986
3987<h5>Overview:</h5>
3988
3989<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
3990the code generator, and allows some metadata to be associated with it.</p>
3991
3992<h5>Arguments:</h5>
3993
3994<p>The first argument specifies the address of a stack object that contains the
3995root pointer. The second pointer (which must be either a constant or a global
3996value address) contains the meta-data to be associated with the root.</p>
3997
3998<h5>Semantics:</h5>
3999
4000<p>At runtime, a call to this intrinsics stores a null pointer into the "ptrloc"
4001location. At compile-time, the code generator generates information to allow
4002the runtime to find the pointer at GC safe points.
4003</p>
4004
4005</div>
4006
4007
4008<!-- _______________________________________________________________________ -->
4009<div class="doc_subsubsection">
4010 <a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
4011</div>
4012
4013<div class="doc_text">
4014
4015<h5>Syntax:</h5>
4016
4017<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]4018 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4019</pre>
4020
4021<h5>Overview:</h5>
4022
4023<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
4024locations, allowing garbage collector implementations that require read
4025barriers.</p>
4026
4027<h5>Arguments:</h5>
4028
4029<p>The second argument is the address to read from, which should be an address
4030allocated from the garbage collector. The first object is a pointer to the
4031start of the referenced object, if needed by the language runtime (otherwise
4032null).</p>
4033
4034<h5>Semantics:</h5>
4035
4036<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
4037instruction, but may be replaced with substantially more complex code by the
4038garbage collector runtime, as needed.</p>
4039
4040</div>
4041
4042
4043<!-- _______________________________________________________________________ -->
4044<div class="doc_subsubsection">
4045 <a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
4046</div>
4047
4048<div class="doc_text">
4049
4050<h5>Syntax:</h5>
4051
4052<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]4053 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4054</pre>
4055
4056<h5>Overview:</h5>
4057
4058<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
4059locations, allowing garbage collector implementations that require write
4060barriers (such as generational or reference counting collectors).</p>
4061
4062<h5>Arguments:</h5>
4063
4064<p>The first argument is the reference to store, the second is the start of the
4065object to store it to, and the third is the address of the field of Obj to
4066store to. If the runtime does not require a pointer to the object, Obj may be
4067null.</p>
4068
4069<h5>Semantics:</h5>
4070
4071<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
4072instruction, but may be replaced with substantially more complex code by the
4073garbage collector runtime, as needed.</p>
4074
4075</div>
4076
4077
4078
4079<!-- ======================================================================= -->
4080<div class="doc_subsection">
4081 <a name="int_codegen">Code Generator Intrinsics</a>
4082</div>
4083
4084<div class="doc_text">
4085<p>
4086These intrinsics are provided by LLVM to expose special features that may only
4087be implemented with code generator support.
4088</p>
4089
4090</div>
4091
4092<!-- _______________________________________________________________________ -->
4093<div class="doc_subsubsection">
4094 <a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
4095</div>
4096
4097<div class="doc_text">
4098
4099<h5>Syntax:</h5>
4100<pre>
4101 declare i8 *@llvm.returnaddress(i32 &lt;level&gt;)
4102</pre>
4103
4104<h5>Overview:</h5>
4105
4106<p>
4107The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
4108target-specific value indicating the return address of the current function
4109or one of its callers.
4110</p>
4111
4112<h5>Arguments:</h5>
4113
4114<p>
4115The argument to this intrinsic indicates which function to return the address
4116for. Zero indicates the calling function, one indicates its caller, etc. The
4117argument is <b>required</b> to be a constant integer value.
4118</p>
4119
4120<h5>Semantics:</h5>
4121
4122<p>
4123The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
4124the return address of the specified call frame, or zero if it cannot be
4125identified. The value returned by this intrinsic is likely to be incorrect or 0
4126for arguments other than zero, so it should only be used for debugging purposes.
4127</p>
4128
4129<p>
4130Note that calling this intrinsic does not prevent function inlining or other
4131aggressive transformations, so the value returned may not be that of the obvious
4132source-language caller.
4133</p>
4134</div>
4135
4136
4137<!-- _______________________________________________________________________ -->
4138<div class="doc_subsubsection">
4139 <a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
4140</div>
4141
4142<div class="doc_text">
4143
4144<h5>Syntax:</h5>
4145<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]4146 declare i8 *@llvm.frameaddress(i32 &lt;level&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4147</pre>
4148
4149<h5>Overview:</h5>
4150
4151<p>
4152The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
4153target-specific frame pointer value for the specified stack frame.
4154</p>
4155
4156<h5>Arguments:</h5>
4157
4158<p>
4159The argument to this intrinsic indicates which function to return the frame
4160pointer for. Zero indicates the calling function, one indicates its caller,
4161etc. The argument is <b>required</b> to be a constant integer value.
4162</p>
4163
4164<h5>Semantics:</h5>
4165
4166<p>
4167The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
4168the frame address of the specified call frame, or zero if it cannot be
4169identified. The value returned by this intrinsic is likely to be incorrect or 0
4170for arguments other than zero, so it should only be used for debugging purposes.
4171</p>
4172
4173<p>
4174Note that calling this intrinsic does not prevent function inlining or other
4175aggressive transformations, so the value returned may not be that of the obvious
4176source-language caller.
4177</p>
4178</div>
4179
4180<!-- _______________________________________________________________________ -->
4181<div class="doc_subsubsection">
4182 <a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
4183</div>
4184
4185<div class="doc_text">
4186
4187<h5>Syntax:</h5>
4188<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]4189 declare i8 *@llvm.stacksave()
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4190</pre>
4191
4192<h5>Overview:</h5>
4193
4194<p>
4195The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
4196the function stack, for use with <a href="#int_stackrestore">
4197<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
4198features like scoped automatic variable sized arrays in C99.
4199</p>
4200
4201<h5>Semantics:</h5>
4202
4203<p>
4204This intrinsic returns a opaque pointer value that can be passed to <a
4205href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
4206<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
4207<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
4208state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
4209practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
4210that were allocated after the <tt>llvm.stacksave</tt> was executed.
4211</p>
4212
4213</div>
4214
4215<!-- _______________________________________________________________________ -->
4216<div class="doc_subsubsection">
4217 <a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
4218</div>
4219
4220<div class="doc_text">
4221
4222<h5>Syntax:</h5>
4223<pre>
4224 declare void @llvm.stackrestore(i8 * %ptr)
4225</pre>
4226
4227<h5>Overview:</h5>
4228
4229<p>
4230The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
4231the function stack to the state it was in when the corresponding <a
4232href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
4233useful for implementing language features like scoped automatic variable sized
4234arrays in C99.
4235</p>
4236
4237<h5>Semantics:</h5>
4238
4239<p>
4240See the description for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.
4241</p>
4242
4243</div>
4244
4245
4246<!-- _______________________________________________________________________ -->
4247<div class="doc_subsubsection">
4248 <a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
4249</div>
4250
4251<div class="doc_text">
4252
4253<h5>Syntax:</h5>
4254<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]4255 declare void @llvm.prefetch(i8* &lt;address&gt;, i32 &lt;rw&gt;, i32 &lt;locality&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4256</pre>
4257
4258<h5>Overview:</h5>
4259
4260
4261<p>
4262The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
4263a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
4264no
4265effect on the behavior of the program but can change its performance
4266characteristics.
4267</p>
4268
4269<h5>Arguments:</h5>
4270
4271<p>
4272<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
4273determining if the fetch should be for a read (0) or write (1), and
4274<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
4275locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
4276<tt>locality</tt> arguments must be constant integers.
4277</p>
4278
4279<h5>Semantics:</h5>
4280
4281<p>
4282This intrinsic does not modify the behavior of the program. In particular,
4283prefetches cannot trap and do not produce a value. On targets that support this
4284intrinsic, the prefetch can provide hints to the processor cache for better
4285performance.
4286</p>
4287
4288</div>
4289
4290<!-- _______________________________________________________________________ -->
4291<div class="doc_subsubsection">
4292 <a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
4293</div>
4294
4295<div class="doc_text">
4296
4297<h5>Syntax:</h5>
4298<pre>
Chris Lattner38bd5dd2007-09-21 17:30:40 +0000[diff] [blame]4299 declare void @llvm.pcmarker(i32 &lt;id&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4300</pre>
4301
4302<h5>Overview:</h5>
4303
4304
4305<p>
4306The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
4307(PC) in a region of
4308code to simulators and other tools. The method is target specific, but it is
4309expected that the marker will use exported symbols to transmit the PC of the marker.
4310The marker makes no guarantees that it will remain with any specific instruction
4311after optimizations. It is possible that the presence of a marker will inhibit
4312optimizations. The intended use is to be inserted after optimizations to allow
4313correlations of simulation runs.
4314</p>
4315
4316<h5>Arguments:</h5>
4317
4318<p>
4319<tt>id</tt> is a numerical id identifying the marker.
4320</p>
4321
4322<h5>Semantics:</h5>
4323
4324<p>
4325This intrinsic does not modify the behavior of the program. Backends that do not
4326support this intrinisic may ignore it.
4327</p>
4328
4329</div>
4330
4331<!-- _______________________________________________________________________ -->
4332<div class="doc_subsubsection">
4333 <a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
4334</div>
4335
4336<div class="doc_text">
4337
4338<h5>Syntax:</h5>
4339<pre>
4340 declare i64 @llvm.readcyclecounter( )
4341</pre>
4342
4343<h5>Overview:</h5>
4344
4345
4346<p>
4347The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
4348counter register (or similar low latency, high accuracy clocks) on those targets
4349that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
4350As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
4351should only be used for small timings.
4352</p>
4353
4354<h5>Semantics:</h5>
4355
4356<p>
4357When directly supported, reading the cycle counter should not modify any memory.
4358Implementations are allowed to either return a application specific value or a
4359system wide value. On backends without support, this is lowered to a constant 0.
4360</p>
4361
4362</div>
4363
4364<!-- ======================================================================= -->
4365<div class="doc_subsection">
4366 <a name="int_libc">Standard C Library Intrinsics</a>
4367</div>
4368
4369<div class="doc_text">
4370<p>
4371LLVM provides intrinsics for a few important standard C library functions.
4372These intrinsics allow source-language front-ends to pass information about the
4373alignment of the pointer arguments to the code generator, providing opportunity
4374for more efficient code generation.
4375</p>
4376
4377</div>
4378
4379<!-- _______________________________________________________________________ -->
4380<div class="doc_subsubsection">
4381 <a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
4382</div>
4383
4384<div class="doc_text">
4385
4386<h5>Syntax:</h5>
4387<pre>
4388 declare void @llvm.memcpy.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
4389 i32 &lt;len&gt;, i32 &lt;align&gt;)
4390 declare void @llvm.memcpy.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
4391 i64 &lt;len&gt;, i32 &lt;align&gt;)
4392</pre>
4393
4394<h5>Overview:</h5>
4395
4396<p>
4397The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
4398location to the destination location.
4399</p>
4400
4401<p>
4402Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
4403intrinsics do not return a value, and takes an extra alignment argument.
4404</p>
4405
4406<h5>Arguments:</h5>
4407
4408<p>
4409The first argument is a pointer to the destination, the second is a pointer to
4410the source. The third argument is an integer argument
4411specifying the number of bytes to copy, and the fourth argument is the alignment
4412of the source and destination locations.
4413</p>
4414
4415<p>
4416If the call to this intrinisic has an alignment value that is not 0 or 1, then
4417the caller guarantees that both the source and destination pointers are aligned
4418to that boundary.
4419</p>
4420
4421<h5>Semantics:</h5>
4422
4423<p>
4424The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
4425location to the destination location, which are not allowed to overlap. It
4426copies "len" bytes of memory over. If the argument is known to be aligned to
4427some boundary, this can be specified as the fourth argument, otherwise it should
4428be set to 0 or 1.
4429</p>
4430</div>
4431
4432
4433<!-- _______________________________________________________________________ -->
4434<div class="doc_subsubsection">
4435 <a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
4436</div>
4437
4438<div class="doc_text">
4439
4440<h5>Syntax:</h5>
4441<pre>
4442 declare void @llvm.memmove.i32(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
4443 i32 &lt;len&gt;, i32 &lt;align&gt;)
4444 declare void @llvm.memmove.i64(i8 * &lt;dest&gt;, i8 * &lt;src&gt;,
4445 i64 &lt;len&gt;, i32 &lt;align&gt;)
4446</pre>
4447
4448<h5>Overview:</h5>
4449
4450<p>
4451The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the source
4452location to the destination location. It is similar to the
4453'<tt>llvm.memcmp</tt>' intrinsic but allows the two memory locations to overlap.
4454</p>
4455
4456<p>
4457Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
4458intrinsics do not return a value, and takes an extra alignment argument.
4459</p>
4460
4461<h5>Arguments:</h5>
4462
4463<p>
4464The first argument is a pointer to the destination, the second is a pointer to
4465the source. The third argument is an integer argument
4466specifying the number of bytes to copy, and the fourth argument is the alignment
4467of the source and destination locations.
4468</p>
4469
4470<p>
4471If the call to this intrinisic has an alignment value that is not 0 or 1, then
4472the caller guarantees that the source and destination pointers are aligned to
4473that boundary.
4474</p>
4475
4476<h5>Semantics:</h5>
4477
4478<p>
4479The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the source
4480location to the destination location, which may overlap. It
4481copies "len" bytes of memory over. If the argument is known to be aligned to
4482some boundary, this can be specified as the fourth argument, otherwise it should
4483be set to 0 or 1.
4484</p>
4485</div>
4486
4487
4488<!-- _______________________________________________________________________ -->
4489<div class="doc_subsubsection">
4490 <a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
4491</div>
4492
4493<div class="doc_text">
4494
4495<h5>Syntax:</h5>
4496<pre>
4497 declare void @llvm.memset.i32(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
4498 i32 &lt;len&gt;, i32 &lt;align&gt;)
4499 declare void @llvm.memset.i64(i8 * &lt;dest&gt;, i8 &lt;val&gt;,
4500 i64 &lt;len&gt;, i32 &lt;align&gt;)
4501</pre>
4502
4503<h5>Overview:</h5>
4504
4505<p>
4506The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a particular
4507byte value.
4508</p>
4509
4510<p>
4511Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
4512does not return a value, and takes an extra alignment argument.
4513</p>
4514
4515<h5>Arguments:</h5>
4516
4517<p>
4518The first argument is a pointer to the destination to fill, the second is the
4519byte value to fill it with, the third argument is an integer
4520argument specifying the number of bytes to fill, and the fourth argument is the
4521known alignment of destination location.
4522</p>
4523
4524<p>
4525If the call to this intrinisic has an alignment value that is not 0 or 1, then
4526the caller guarantees that the destination pointer is aligned to that boundary.
4527</p>
4528
4529<h5>Semantics:</h5>
4530
4531<p>
4532The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting at
4533the
4534destination location. If the argument is known to be aligned to some boundary,
4535this can be specified as the fourth argument, otherwise it should be set to 0 or
45361.
4537</p>
4538</div>
4539
4540
4541<!-- _______________________________________________________________________ -->
4542<div class="doc_subsubsection">
4543 <a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
4544</div>
4545
4546<div class="doc_text">
4547
4548<h5>Syntax:</h5>
Dale Johannesenf9adbb62007-10-02 17:47:38 +0000[diff] [blame]4549<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4550floating point or vector of floating point type. Not all targets support all
4551types however.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4552<pre>
Dale Johannesenf9adbb62007-10-02 17:47:38 +0000[diff] [blame]4553 declare float @llvm.sqrt.f32(float %Val)
4554 declare double @llvm.sqrt.f64(double %Val)
4555 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
4556 declare fp128 @llvm.sqrt.f128(fp128 %Val)
4557 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4558</pre>
4559
4560<h5>Overview:</h5>
4561
4562<p>
4563The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4564returning the same value as the libm '<tt>sqrt</tt>' functions would. Unlike
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4565<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
4566negative numbers (which allows for better optimization).
4567</p>
4568
4569<h5>Arguments:</h5>
4570
4571<p>
4572The argument and return value are floating point numbers of the same type.
4573</p>
4574
4575<h5>Semantics:</h5>
4576
4577<p>
4578This function returns the sqrt of the specified operand if it is a nonnegative
4579floating point number.
4580</p>
4581</div>
4582
4583<!-- _______________________________________________________________________ -->
4584<div class="doc_subsubsection">
4585 <a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
4586</div>
4587
4588<div class="doc_text">
4589
4590<h5>Syntax:</h5>
Dale Johannesenf9adbb62007-10-02 17:47:38 +0000[diff] [blame]4591<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4592floating point or vector of floating point type. Not all targets support all
4593types however.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4594<pre>
Dale Johannesenf9adbb62007-10-02 17:47:38 +0000[diff] [blame]4595 declare float @llvm.powi.f32(float %Val, i32 %power)
4596 declare double @llvm.powi.f64(double %Val, i32 %power)
4597 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
4598 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
4599 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4600</pre>
4601
4602<h5>Overview:</h5>
4603
4604<p>
4605The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
4606specified (positive or negative) power. The order of evaluation of
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4607multiplications is not defined. When a vector of floating point type is
4608used, the second argument remains a scalar integer value.
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4609</p>
4610
4611<h5>Arguments:</h5>
4612
4613<p>
4614The second argument is an integer power, and the first is a value to raise to
4615that power.
4616</p>
4617
4618<h5>Semantics:</h5>
4619
4620<p>
4621This function returns the first value raised to the second power with an
4622unspecified sequence of rounding operations.</p>
4623</div>
4624
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4625<!-- _______________________________________________________________________ -->
4626<div class="doc_subsubsection">
4627 <a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
4628</div>
4629
4630<div class="doc_text">
4631
4632<h5>Syntax:</h5>
4633<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
4634floating point or vector of floating point type. Not all targets support all
4635types however.
4636<pre>
4637 declare float @llvm.sin.f32(float %Val)
4638 declare double @llvm.sin.f64(double %Val)
4639 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
4640 declare fp128 @llvm.sin.f128(fp128 %Val)
4641 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
4642</pre>
4643
4644<h5>Overview:</h5>
4645
4646<p>
4647The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.
4648</p>
4649
4650<h5>Arguments:</h5>
4651
4652<p>
4653The argument and return value are floating point numbers of the same type.
4654</p>
4655
4656<h5>Semantics:</h5>
4657
4658<p>
4659This function returns the sine of the specified operand, returning the
4660same values as the libm <tt>sin</tt> functions would, and handles error
Dan Gohmaneaba92e2007-10-17 18:05:13 +0000[diff] [blame]4661conditions in the same way.</p>
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4662</div>
4663
4664<!-- _______________________________________________________________________ -->
4665<div class="doc_subsubsection">
4666 <a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
4667</div>
4668
4669<div class="doc_text">
4670
4671<h5>Syntax:</h5>
4672<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
4673floating point or vector of floating point type. Not all targets support all
4674types however.
4675<pre>
4676 declare float @llvm.cos.f32(float %Val)
4677 declare double @llvm.cos.f64(double %Val)
4678 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
4679 declare fp128 @llvm.cos.f128(fp128 %Val)
4680 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
4681</pre>
4682
4683<h5>Overview:</h5>
4684
4685<p>
4686The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.
4687</p>
4688
4689<h5>Arguments:</h5>
4690
4691<p>
4692The argument and return value are floating point numbers of the same type.
4693</p>
4694
4695<h5>Semantics:</h5>
4696
4697<p>
4698This function returns the cosine of the specified operand, returning the
4699same values as the libm <tt>cos</tt> functions would, and handles error
Dan Gohmaneaba92e2007-10-17 18:05:13 +0000[diff] [blame]4700conditions in the same way.</p>
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4701</div>
4702
4703<!-- _______________________________________________________________________ -->
4704<div class="doc_subsubsection">
4705 <a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
4706</div>
4707
4708<div class="doc_text">
4709
4710<h5>Syntax:</h5>
4711<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
4712floating point or vector of floating point type. Not all targets support all
4713types however.
4714<pre>
4715 declare float @llvm.pow.f32(float %Val, float %Power)
4716 declare double @llvm.pow.f64(double %Val, double %Power)
4717 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
4718 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
4719 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
4720</pre>
4721
4722<h5>Overview:</h5>
4723
4724<p>
4725The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
4726specified (positive or negative) power.
4727</p>
4728
4729<h5>Arguments:</h5>
4730
4731<p>
4732The second argument is a floating point power, and the first is a value to
4733raise to that power.
4734</p>
4735
4736<h5>Semantics:</h5>
4737
4738<p>
4739This function returns the first value raised to the second power,
4740returning the
4741same values as the libm <tt>pow</tt> functions would, and handles error
Dan Gohmaneaba92e2007-10-17 18:05:13 +0000[diff] [blame]4742conditions in the same way.</p>
Dan Gohman361079c2007-10-15 20:30:11 +0000[diff] [blame]4743</div>
4744
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4745
4746<!-- ======================================================================= -->
4747<div class="doc_subsection">
4748 <a name="int_manip">Bit Manipulation Intrinsics</a>
4749</div>
4750
4751<div class="doc_text">
4752<p>
4753LLVM provides intrinsics for a few important bit manipulation operations.
4754These allow efficient code generation for some algorithms.
4755</p>
4756
4757</div>
4758
4759<!-- _______________________________________________________________________ -->
4760<div class="doc_subsubsection">
4761 <a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
4762</div>
4763
4764<div class="doc_text">
4765
4766<h5>Syntax:</h5>
4767<p>This is an overloaded intrinsic function. You can use bswap on any integer
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4768type that is an even number of bytes (i.e. BitWidth % 16 == 0).
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4769<pre>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4770 declare i16 @llvm.bswap.i16(i16 &lt;id&gt;)
4771 declare i32 @llvm.bswap.i32(i32 &lt;id&gt;)
4772 declare i64 @llvm.bswap.i64(i64 &lt;id&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4773</pre>
4774
4775<h5>Overview:</h5>
4776
4777<p>
4778The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
4779values with an even number of bytes (positive multiple of 16 bits). These are
4780useful for performing operations on data that is not in the target's native
4781byte order.
4782</p>
4783
4784<h5>Semantics:</h5>
4785
4786<p>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4787The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4788and low byte of the input i16 swapped. Similarly, the <tt>llvm.bswap.i32</tt>
4789intrinsic returns an i32 value that has the four bytes of the input i32
4790swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the returned
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4791i32 will have its bytes in 3, 2, 1, 0 order. The <tt>llvm.bswap.i48</tt>,
4792<tt>llvm.bswap.i64</tt> and other intrinsics extend this concept to
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4793additional even-byte lengths (6 bytes, 8 bytes and more, respectively).
4794</p>
4795
4796</div>
4797
4798<!-- _______________________________________________________________________ -->
4799<div class="doc_subsubsection">
4800 <a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
4801</div>
4802
4803<div class="doc_text">
4804
4805<h5>Syntax:</h5>
4806<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
4807width. Not all targets support all bit widths however.
4808<pre>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4809 declare i8 @llvm.ctpop.i8 (i8 &lt;src&gt;)
4810 declare i16 @llvm.ctpop.i16(i16 &lt;src&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4811 declare i32 @llvm.ctpop.i32(i32 &lt;src&gt;)
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4812 declare i64 @llvm.ctpop.i64(i64 &lt;src&gt;)
4813 declare i256 @llvm.ctpop.i256(i256 &lt;src&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4814</pre>
4815
4816<h5>Overview:</h5>
4817
4818<p>
4819The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set in a
4820value.
4821</p>
4822
4823<h5>Arguments:</h5>
4824
4825<p>
4826The only argument is the value to be counted. The argument may be of any
4827integer type. The return type must match the argument type.
4828</p>
4829
4830<h5>Semantics:</h5>
4831
4832<p>
4833The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
4834</p>
4835</div>
4836
4837<!-- _______________________________________________________________________ -->
4838<div class="doc_subsubsection">
4839 <a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
4840</div>
4841
4842<div class="doc_text">
4843
4844<h5>Syntax:</h5>
4845<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
4846integer bit width. Not all targets support all bit widths however.
4847<pre>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4848 declare i8 @llvm.ctlz.i8 (i8 &lt;src&gt;)
4849 declare i16 @llvm.ctlz.i16(i16 &lt;src&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4850 declare i32 @llvm.ctlz.i32(i32 &lt;src&gt;)
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4851 declare i64 @llvm.ctlz.i64(i64 &lt;src&gt;)
4852 declare i256 @llvm.ctlz.i256(i256 &lt;src&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4853</pre>
4854
4855<h5>Overview:</h5>
4856
4857<p>
4858The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
4859leading zeros in a variable.
4860</p>
4861
4862<h5>Arguments:</h5>
4863
4864<p>
4865The only argument is the value to be counted. The argument may be of any
4866integer type. The return type must match the argument type.
4867</p>
4868
4869<h5>Semantics:</h5>
4870
4871<p>
4872The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
4873in a variable. If the src == 0 then the result is the size in bits of the type
4874of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.
4875</p>
4876</div>
4877
4878
4879
4880<!-- _______________________________________________________________________ -->
4881<div class="doc_subsubsection">
4882 <a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
4883</div>
4884
4885<div class="doc_text">
4886
4887<h5>Syntax:</h5>
4888<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
4889integer bit width. Not all targets support all bit widths however.
4890<pre>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4891 declare i8 @llvm.cttz.i8 (i8 &lt;src&gt;)
4892 declare i16 @llvm.cttz.i16(i16 &lt;src&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4893 declare i32 @llvm.cttz.i32(i32 &lt;src&gt;)
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4894 declare i64 @llvm.cttz.i64(i64 &lt;src&gt;)
4895 declare i256 @llvm.cttz.i256(i256 &lt;src&gt;)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4896</pre>
4897
4898<h5>Overview:</h5>
4899
4900<p>
4901The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
4902trailing zeros.
4903</p>
4904
4905<h5>Arguments:</h5>
4906
4907<p>
4908The only argument is the value to be counted. The argument may be of any
4909integer type. The return type must match the argument type.
4910</p>
4911
4912<h5>Semantics:</h5>
4913
4914<p>
4915The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
4916in a variable. If the src == 0 then the result is the size in bits of the type
4917of src. For example, <tt>llvm.cttz(2) = 1</tt>.
4918</p>
4919</div>
4920
4921<!-- _______________________________________________________________________ -->
4922<div class="doc_subsubsection">
4923 <a name="int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic</a>
4924</div>
4925
4926<div class="doc_text">
4927
4928<h5>Syntax:</h5>
4929<p>This is an overloaded intrinsic. You can use <tt>llvm.part.select</tt>
4930on any integer bit width.
4931<pre>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4932 declare i17 @llvm.part.select.i17 (i17 %val, i32 %loBit, i32 %hiBit)
4933 declare i29 @llvm.part.select.i29 (i29 %val, i32 %loBit, i32 %hiBit)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4934</pre>
4935
4936<h5>Overview:</h5>
4937<p>The '<tt>llvm.part.select</tt>' family of intrinsic functions selects a
4938range of bits from an integer value and returns them in the same bit width as
4939the original value.</p>
4940
4941<h5>Arguments:</h5>
4942<p>The first argument, <tt>%val</tt> and the result may be integer types of
4943any bit width but they must have the same bit width. The second and third
4944arguments must be <tt>i32</tt> type since they specify only a bit index.</p>
4945
4946<h5>Semantics:</h5>
4947<p>The operation of the '<tt>llvm.part.select</tt>' intrinsic has two modes
4948of operation: forwards and reverse. If <tt>%loBit</tt> is greater than
4949<tt>%hiBits</tt> then the intrinsic operates in reverse mode. Otherwise it
4950operates in forward mode.</p>
4951<p>In forward mode, this intrinsic is the equivalent of shifting <tt>%val</tt>
4952right by <tt>%loBit</tt> bits and then ANDing it with a mask with
4953only the <tt>%hiBit - %loBit</tt> bits set, as follows:</p>
4954<ol>
4955 <li>The <tt>%val</tt> is shifted right (LSHR) by the number of bits specified
4956 by <tt>%loBits</tt>. This normalizes the value to the low order bits.</li>
4957 <li>The <tt>%loBits</tt> value is subtracted from the <tt>%hiBits</tt> value
4958 to determine the number of bits to retain.</li>
4959 <li>A mask of the retained bits is created by shifting a -1 value.</li>
4960 <li>The mask is ANDed with <tt>%val</tt> to produce the result.
4961</ol>
4962<p>In reverse mode, a similar computation is made except that the bits are
4963returned in the reverse order. So, for example, if <tt>X</tt> has the value
4964<tt>i16 0x0ACF (101011001111)</tt> and we apply
4965<tt>part.select(i16 X, 8, 3)</tt> to it, we get back the value
4966<tt>i16 0x0026 (000000100110)</tt>.</p>
4967</div>
4968
4969<div class="doc_subsubsection">
4970 <a name="int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic</a>
4971</div>
4972
4973<div class="doc_text">
4974
4975<h5>Syntax:</h5>
4976<p>This is an overloaded intrinsic. You can use <tt>llvm.part.set</tt>
4977on any integer bit width.
4978<pre>
Chandler Carrutha228e392007-08-04 01:51:18 +0000[diff] [blame]4979 declare i17 @llvm.part.set.i17.i9 (i17 %val, i9 %repl, i32 %lo, i32 %hi)
4980 declare i29 @llvm.part.set.i29.i9 (i29 %val, i9 %repl, i32 %lo, i32 %hi)
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]4981</pre>
4982
4983<h5>Overview:</h5>
4984<p>The '<tt>llvm.part.set</tt>' family of intrinsic functions replaces a range
4985of bits in an integer value with another integer value. It returns the integer
4986with the replaced bits.</p>
4987
4988<h5>Arguments:</h5>
4989<p>The first argument, <tt>%val</tt> and the result may be integer types of
4990any bit width but they must have the same bit width. <tt>%val</tt> is the value
4991whose bits will be replaced. The second argument, <tt>%repl</tt> may be an
4992integer of any bit width. The third and fourth arguments must be <tt>i32</tt>
4993type since they specify only a bit index.</p>
4994
4995<h5>Semantics:</h5>
4996<p>The operation of the '<tt>llvm.part.set</tt>' intrinsic has two modes
4997of operation: forwards and reverse. If <tt>%lo</tt> is greater than
4998<tt>%hi</tt> then the intrinsic operates in reverse mode. Otherwise it
4999operates in forward mode.</p>
5000<p>For both modes, the <tt>%repl</tt> value is prepared for use by either
5001truncating it down to the size of the replacement area or zero extending it
5002up to that size.</p>
5003<p>In forward mode, the bits between <tt>%lo</tt> and <tt>%hi</tt> (inclusive)
5004are replaced with corresponding bits from <tt>%repl</tt>. That is the 0th bit
5005in <tt>%repl</tt> replaces the <tt>%lo</tt>th bit in <tt>%val</tt> and etc. up
5006to the <tt>%hi</tt>th bit.
5007<p>In reverse mode, a similar computation is made except that the bits are
5008reversed. That is, the <tt>0</tt>th bit in <tt>%repl</tt> replaces the
5009<tt>%hi</tt> bit in <tt>%val</tt> and etc. down to the <tt>%lo</tt>th bit.
5010<h5>Examples:</h5>
5011<pre>
5012 llvm.part.set(0xFFFF, 0, 4, 7) -&gt; 0xFF0F
5013 llvm.part.set(0xFFFF, 0, 7, 4) -&gt; 0xFF0F
5014 llvm.part.set(0xFFFF, 1, 7, 4) -&gt; 0xFF8F
5015 llvm.part.set(0xFFFF, F, 8, 3) -&gt; 0xFFE7
5016 llvm.part.set(0xFFFF, 0, 3, 8) -&gt; 0xFE07
5017</pre>
5018</div>
5019
5020<!-- ======================================================================= -->
5021<div class="doc_subsection">
5022 <a name="int_debugger">Debugger Intrinsics</a>
5023</div>
5024
5025<div class="doc_text">
5026<p>
5027The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
5028are described in the <a
5029href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
5030Debugging</a> document.
5031</p>
5032</div>
5033
5034
5035<!-- ======================================================================= -->
5036<div class="doc_subsection">
5037 <a name="int_eh">Exception Handling Intrinsics</a>
5038</div>
5039
5040<div class="doc_text">
5041<p> The LLVM exception handling intrinsics (which all start with
5042<tt>llvm.eh.</tt> prefix), are described in the <a
5043href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
5044Handling</a> document. </p>
5045</div>
5046
5047<!-- ======================================================================= -->
5048<div class="doc_subsection">
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5049 <a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5050</div>
5051
5052<div class="doc_text">
5053<p>
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5054 This intrinsic makes it possible to excise one parameter, marked with
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5055 the <tt>nest</tt> attribute, from a function. The result is a callable
5056 function pointer lacking the nest parameter - the caller does not need
5057 to provide a value for it. Instead, the value to use is stored in
5058 advance in a "trampoline", a block of memory usually allocated
5059 on the stack, which also contains code to splice the nest value into the
5060 argument list. This is used to implement the GCC nested function address
5061 extension.
5062</p>
5063<p>
5064 For example, if the function is
5065 <tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
Bill Wendlinge262b4c2007-09-22 09:23:55 +0000[diff] [blame]5066 pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as follows:</p>
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5067<pre>
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5068 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
5069 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
5070 %p = call i8* @llvm.init.trampoline( i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8*), i8* %nval )
5071 %fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5072</pre>
Bill Wendlinge262b4c2007-09-22 09:23:55 +0000[diff] [blame]5073 <p>The call <tt>%val = call i32 %fp( i32 %x, i32 %y )</tt> is then equivalent
5074 to <tt>%val = call i32 %f( i8* %nval, i32 %x, i32 %y )</tt>.</p>
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5075</div>
5076
5077<!-- _______________________________________________________________________ -->
5078<div class="doc_subsubsection">
5079 <a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
5080</div>
5081<div class="doc_text">
5082<h5>Syntax:</h5>
5083<pre>
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5084declare i8* @llvm.init.trampoline(i8* &lt;tramp&gt;, i8* &lt;func&gt;, i8* &lt;nval&gt;)
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5085</pre>
5086<h5>Overview:</h5>
5087<p>
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5088 This fills the memory pointed to by <tt>tramp</tt> with code
5089 and returns a function pointer suitable for executing it.
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5090</p>
5091<h5>Arguments:</h5>
5092<p>
5093 The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
5094 pointers. The <tt>tramp</tt> argument must point to a sufficiently large
5095 and sufficiently aligned block of memory; this memory is written to by the
Duncan Sands35012212007-08-22 23:39:54 +0000[diff] [blame]5096 intrinsic. Note that the size and the alignment are target-specific - LLVM
5097 currently provides no portable way of determining them, so a front-end that
5098 generates this intrinsic needs to have some target-specific knowledge.
5099 The <tt>func</tt> argument must hold a function bitcast to an <tt>i8*</tt>.
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5100</p>
5101<h5>Semantics:</h5>
5102<p>
5103 The block of memory pointed to by <tt>tramp</tt> is filled with target
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5104 dependent code, turning it into a function. A pointer to this function is
5105 returned, but needs to be bitcast to an
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5106 <a href="#int_trampoline">appropriate function pointer type</a>
Duncan Sands7407a9f2007-09-11 14:10:23 +0000[diff] [blame]5107 before being called. The new function's signature is the same as that of
5108 <tt>func</tt> with any arguments marked with the <tt>nest</tt> attribute
5109 removed. At most one such <tt>nest</tt> argument is allowed, and it must be
5110 of pointer type. Calling the new function is equivalent to calling
5111 <tt>func</tt> with the same argument list, but with <tt>nval</tt> used for the
5112 missing <tt>nest</tt> argument. If, after calling
5113 <tt>llvm.init.trampoline</tt>, the memory pointed to by <tt>tramp</tt> is
5114 modified, then the effect of any later call to the returned function pointer is
5115 undefined.
Duncan Sands38947cd2007-07-27 12:58:54 +0000[diff] [blame]5116</p>
5117</div>
5118
5119<!-- ======================================================================= -->
5120<div class="doc_subsection">
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]5121 <a name="int_general">General Intrinsics</a>
5122</div>
5123
5124<div class="doc_text">
5125<p> This class of intrinsics is designed to be generic and has
5126no specific purpose. </p>
5127</div>
5128
5129<!-- _______________________________________________________________________ -->
5130<div class="doc_subsubsection">
5131 <a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
5132</div>
5133
5134<div class="doc_text">
5135
5136<h5>Syntax:</h5>
5137<pre>
5138 declare void @llvm.var.annotation(i8* &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5139</pre>
5140
5141<h5>Overview:</h5>
5142
5143<p>
5144The '<tt>llvm.var.annotation</tt>' intrinsic
5145</p>
5146
5147<h5>Arguments:</h5>
5148
5149<p>
5150The first argument is a pointer to a value, the second is a pointer to a
5151global string, the third is a pointer to a global string which is the source
5152file name, and the last argument is the line number.
5153</p>
5154
5155<h5>Semantics:</h5>
5156
5157<p>
5158This intrinsic allows annotation of local variables with arbitrary strings.
5159This can be useful for special purpose optimizations that want to look for these
5160 annotations. These have no other defined use, they are ignored by code
5161 generation and optimization.
5162</div>
5163
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]5164<!-- _______________________________________________________________________ -->
5165<div class="doc_subsubsection">
Tanya Lattnerc9869b12007-09-21 23:57:59 +0000[diff] [blame]5166 <a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]5167</div>
5168
5169<div class="doc_text">
5170
5171<h5>Syntax:</h5>
Tanya Lattnere545be72007-09-21 23:56:27 +0000[diff] [blame]5172<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
5173any integer bit width.
5174</p>
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]5175<pre>
Tanya Lattner09161fe2007-09-22 00:03:01 +0000[diff] [blame]5176 declare i8 @llvm.annotation.i8(i8 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5177 declare i16 @llvm.annotation.i16(i16 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5178 declare i32 @llvm.annotation.i32(i32 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5179 declare i64 @llvm.annotation.i64(i64 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
5180 declare i256 @llvm.annotation.i256(i256 &lt;val&gt;, i8* &lt;str&gt;, i8* &lt;str&gt;, i32 &lt;int&gt; )
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]5181</pre>
5182
5183<h5>Overview:</h5>
Tanya Lattnere545be72007-09-21 23:56:27 +0000[diff] [blame]5184
5185<p>
5186The '<tt>llvm.annotation</tt>' intrinsic.
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]5187</p>
5188
5189<h5>Arguments:</h5>
5190
5191<p>
5192The first argument is an integer value (result of some expression),
5193the second is a pointer to a global string, the third is a pointer to a global
5194string which is the source file name, and the last argument is the line number.
Tanya Lattnere545be72007-09-21 23:56:27 +0000[diff] [blame]5195It returns the value of the first argument.
Tanya Lattnerb306a9e2007-09-21 22:59:12 +0000[diff] [blame]5196</p>
5197
5198<h5>Semantics:</h5>
5199
5200<p>
5201This intrinsic allows annotations to be put on arbitrary expressions
5202with arbitrary strings. This can be useful for special purpose optimizations
5203that want to look for these annotations. These have no other defined use, they
5204are ignored by code generation and optimization.
5205</div>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]5206
5207<!-- *********************************************************************** -->
5208<hr>
5209<address>
5210 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
5211 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
5212 <a href="http://validator.w3.org/check/referer"><img
5213 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
5214
5215 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
5216 <a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
5217 Last modified: $Date$
5218</address>
5219</body>
5220</html>