Blame - llvm/docs/LangRef.html - toolchain/llvm-project

blob: 9fc14b4fea560010a12b552ff9286a06c83e5a05 [file] [log] [blame]

Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	1	<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
				2	"http://www.w3.org/TR/html4/strict.dtd">
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3	<html>
				4	<head>
				5	<title>LLVM Assembly Language Reference Manual</title>
Reid Spencer	cb84e43	2004-08-26 20:44:00 +0000	[diff] [blame]	6	<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
				7	<meta name="author" content="Chris Lattner">
				8	<meta name="description"
				9	content="LLVM Assembly Language Reference Manual.">
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	10	<link rel="stylesheet" href="llvm.css" type="text/css">
				11	</head>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	12
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	13	<body>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	14
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	15	<div class="doc_title"> LLVM Language Reference Manual </div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	16	<ol>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	17	<li><a href="#abstract">Abstract</a></li>
				18	<li><a href="#introduction">Introduction</a></li>
				19	<li><a href="#identifiers">Identifiers</a></li>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	20	<li><a href="#highlevel">High Level Structure</a>
				21	<ol>
				22	<li><a href="#modulestructure">Module Structure</a></li>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	23	<li><a href="#linkage">Linkage Types</a></li>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	24	<li><a href="#callingconv">Calling Conventions</a></li>
Chris Lattner	bc08821	2009-01-11 20:53:49 +0000	[diff] [blame]	25	<li><a href="#namedtypes">Named Types</a></li>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	26	<li><a href="#globalvars">Global Variables</a></li>
Chris Lattner	91c15c4	2006-01-23 23:23:47 +0000	[diff] [blame]	27	<li><a href="#functionstructure">Functions</a></li>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	28	<li><a href="#aliasstructure">Aliases</a></li>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	29	<li><a href="#paramattrs">Parameter Attributes</a></li>
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	30	<li><a href="#fnattrs">Function Attributes</a></li>
Gordon Henriksen	71183b6	2007-12-10 03:18:06 +0000	[diff] [blame]	31	<li><a href="#gc">Garbage Collector Names</a></li>
Chris Lattner	91c15c4	2006-01-23 23:23:47 +0000	[diff] [blame]	32	<li><a href="#moduleasm">Module-Level Inline Assembly</a></li>
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	33	<li><a href="#datalayout">Data Layout</a></li>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	34	</ol>
				35	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	36	<li><a href="#typesystem">Type System</a>
				37	<ol>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	38	<li><a href="#t_classifications">Type Classifications</a></li>
Robert Bocchino	820bc75b	2006-02-17 21:18:08 +0000	[diff] [blame]	39	<li><a href="#t_primitive">Primitive Types</a>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	40	<ol>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	41	<li><a href="#t_floating">Floating Point Types</a></li>
				42	<li><a href="#t_void">Void Type</a></li>
				43	<li><a href="#t_label">Label Type</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	44	</ol>
				45	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	46	<li><a href="#t_derived">Derived Types</a>
				47	<ol>
Chris Lattner	9a2e3cb	2007-12-18 06:18:21 +0000	[diff] [blame]	48	<li><a href="#t_integer">Integer Type</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	49	<li><a href="#t_array">Array Type</a></li>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	50	<li><a href="#t_function">Function Type</a></li>
				51	<li><a href="#t_pointer">Pointer Type</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	52	<li><a href="#t_struct">Structure Type</a></li>
Andrew Lenharth	8df88e2	2006-12-08 17:13:00 +0000	[diff] [blame]	53	<li><a href="#t_pstruct">Packed Structure Type</a></li>
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	54	<li><a href="#t_vector">Vector Type</a></li>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	55	<li><a href="#t_opaque">Opaque Type</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	56	</ol>
				57	</li>
Chris Lattner	cf7a584	2009-02-02 07:32:36 +0000	[diff] [blame]	58	<li><a href="#t_uprefs">Type Up-references</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	59	</ol>
				60	</li>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	61	<li><a href="#constants">Constants</a>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	62	<ol>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	63	<li><a href="#simpleconstants">Simple Constants</a></li>
Chris Lattner	361bfcd	2009-02-28 18:32:25 +0000	[diff] [blame]	64	<li><a href="#complexconstants">Complex Constants</a></li>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	65	<li><a href="#globalconstants">Global Variable and Function Addresses</a></li>
				66	<li><a href="#undefvalues">Undefined Values</a></li>
				67	<li><a href="#constantexprs">Constant Expressions</a></li>
Nick Lewycky	49f8919	2009-04-04 07:22:01 +0000	[diff] [blame]	68	<li><a href="#metadata">Embedded Metadata</a></li>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	69	</ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	70	</li>
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	71	<li><a href="#othervalues">Other Values</a>
				72	<ol>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	73	<li><a href="#inlineasm">Inline Assembler Expressions</a></li>
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	74	</ol>
				75	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	76	<li><a href="#instref">Instruction Reference</a>
				77	<ol>
				78	<li><a href="#terminators">Terminator Instructions</a>
				79	<ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	80	<li><a href="#i_ret">'<tt>ret</tt>' Instruction</a></li>
				81	<li><a href="#i_br">'<tt>br</tt>' Instruction</a></li>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	82	<li><a href="#i_switch">'<tt>switch</tt>' Instruction</a></li>
				83	<li><a href="#i_invoke">'<tt>invoke</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	84	<li><a href="#i_unwind">'<tt>unwind</tt>' Instruction</a></li>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	85	<li><a href="#i_unreachable">'<tt>unreachable</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	86	</ol>
				87	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	88	<li><a href="#binaryops">Binary Operations</a>
				89	<ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	90	<li><a href="#i_add">'<tt>add</tt>' Instruction</a></li>
				91	<li><a href="#i_sub">'<tt>sub</tt>' Instruction</a></li>
				92	<li><a href="#i_mul">'<tt>mul</tt>' Instruction</a></li>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	93	<li><a href="#i_udiv">'<tt>udiv</tt>' Instruction</a></li>
				94	<li><a href="#i_sdiv">'<tt>sdiv</tt>' Instruction</a></li>
				95	<li><a href="#i_fdiv">'<tt>fdiv</tt>' Instruction</a></li>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	96	<li><a href="#i_urem">'<tt>urem</tt>' Instruction</a></li>
				97	<li><a href="#i_srem">'<tt>srem</tt>' Instruction</a></li>
				98	<li><a href="#i_frem">'<tt>frem</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	99	</ol>
				100	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	101	<li><a href="#bitwiseops">Bitwise Binary Operations</a>
				102	<ol>
Reid Spencer	2ab0193	2007-02-02 13:57:07 +0000	[diff] [blame]	103	<li><a href="#i_shl">'<tt>shl</tt>' Instruction</a></li>
				104	<li><a href="#i_lshr">'<tt>lshr</tt>' Instruction</a></li>
				105	<li><a href="#i_ashr">'<tt>ashr</tt>' Instruction</a></li>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	106	<li><a href="#i_and">'<tt>and</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	107	<li><a href="#i_or">'<tt>or</tt>' Instruction</a></li>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	108	<li><a href="#i_xor">'<tt>xor</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	109	</ol>
				110	</li>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	111	<li><a href="#vectorops">Vector Operations</a>
				112	<ol>
				113	<li><a href="#i_extractelement">'<tt>extractelement</tt>' Instruction</a></li>
				114	<li><a href="#i_insertelement">'<tt>insertelement</tt>' Instruction</a></li>
				115	<li><a href="#i_shufflevector">'<tt>shufflevector</tt>' Instruction</a></li>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	116	</ol>
				117	</li>
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	118	<li><a href="#aggregateops">Aggregate Operations</a>
				119	<ol>
				120	<li><a href="#i_extractvalue">'<tt>extractvalue</tt>' Instruction</a></li>
				121	<li><a href="#i_insertvalue">'<tt>insertvalue</tt>' Instruction</a></li>
				122	</ol>
				123	</li>
Chris Lattner	6ab6672	2006-08-15 00:45:58 +0000	[diff] [blame]	124	<li><a href="#memoryops">Memory Access and Addressing Operations</a>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	125	<ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	126	<li><a href="#i_malloc">'<tt>malloc</tt>' Instruction</a></li>
				127	<li><a href="#i_free">'<tt>free</tt>' Instruction</a></li>
				128	<li><a href="#i_alloca">'<tt>alloca</tt>' Instruction</a></li>
Robert Bocchino	820bc75b	2006-02-17 21:18:08 +0000	[diff] [blame]	129	<li><a href="#i_load">'<tt>load</tt>' Instruction</a></li>
				130	<li><a href="#i_store">'<tt>store</tt>' Instruction</a></li>
				131	<li><a href="#i_getelementptr">'<tt>getelementptr</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	132	</ol>
				133	</li>
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	134	<li><a href="#convertops">Conversion Operations</a>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	135	<ol>
				136	<li><a href="#i_trunc">'<tt>trunc .. to</tt>' Instruction</a></li>
				137	<li><a href="#i_zext">'<tt>zext .. to</tt>' Instruction</a></li>
				138	<li><a href="#i_sext">'<tt>sext .. to</tt>' Instruction</a></li>
				139	<li><a href="#i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a></li>
				140	<li><a href="#i_fpext">'<tt>fpext .. to</tt>' Instruction</a></li>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	141	<li><a href="#i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a></li>
				142	<li><a href="#i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a></li>
				143	<li><a href="#i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a></li>
				144	<li><a href="#i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a></li>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	145	<li><a href="#i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a></li>
				146	<li><a href="#i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a></li>
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	147	<li><a href="#i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a></li>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	148	</ol>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	149	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	150	<li><a href="#otherops">Other Operations</a>
				151	<ol>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	152	<li><a href="#i_icmp">'<tt>icmp</tt>' Instruction</a></li>
				153	<li><a href="#i_fcmp">'<tt>fcmp</tt>' Instruction</a></li>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	154	<li><a href="#i_vicmp">'<tt>vicmp</tt>' Instruction</a></li>
				155	<li><a href="#i_vfcmp">'<tt>vfcmp</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	156	<li><a href="#i_phi">'<tt>phi</tt>' Instruction</a></li>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	157	<li><a href="#i_select">'<tt>select</tt>' Instruction</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	158	<li><a href="#i_call">'<tt>call</tt>' Instruction</a></li>
Chris Lattner	3333747	2006-01-13 23:26:01 +0000	[diff] [blame]	159	<li><a href="#i_va_arg">'<tt>va_arg</tt>' Instruction</a></li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	160	</ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	161	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	162	</ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	163	</li>
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	164	<li><a href="#intrinsics">Intrinsic Functions</a>
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	165	<ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	166	<li><a href="#int_varargs">Variable Argument Handling Intrinsics</a>
				167	<ol>
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	168	<li><a href="#int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a></li>
				169	<li><a href="#int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a></li>
				170	<li><a href="#int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a></li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	171	</ol>
				172	</li>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	173	<li><a href="#int_gc">Accurate Garbage Collection Intrinsics</a>
				174	<ol>
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	175	<li><a href="#int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a></li>
				176	<li><a href="#int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a></li>
				177	<li><a href="#int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a></li>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	178	</ol>
				179	</li>
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	180	<li><a href="#int_codegen">Code Generator Intrinsics</a>
				181	<ol>
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	182	<li><a href="#int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a></li>
				183	<li><a href="#int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a></li>
				184	<li><a href="#int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a></li>
				185	<li><a href="#int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a></li>
				186	<li><a href="#int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a></li>
				187	<li><a href="#int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a></li>
				188	<li><a href="#int_readcyclecounter"><tt>llvm.readcyclecounter</tt>' Intrinsic</a></li>
John Criswell	aa1c3c1	2004-04-09 16:43:20 +0000	[diff] [blame]	189	</ol>
				190	</li>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	191	<li><a href="#int_libc">Standard C Library Intrinsics</a>
				192	<ol>
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	193	<li><a href="#int_memcpy">'<tt>llvm.memcpy.*</tt>' Intrinsic</a></li>
				194	<li><a href="#int_memmove">'<tt>llvm.memmove.*</tt>' Intrinsic</a></li>
				195	<li><a href="#int_memset">'<tt>llvm.memset.*</tt>' Intrinsic</a></li>
				196	<li><a href="#int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a></li>
				197	<li><a href="#int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a></li>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	198	<li><a href="#int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a></li>
				199	<li><a href="#int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a></li>
				200	<li><a href="#int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a></li>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	201	</ol>
				202	</li>
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	203	<li><a href="#int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	204	<ol>
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	205	<li><a href="#int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a></li>
Chris Lattner	b748c67	2006-01-16 22:34:14 +0000	[diff] [blame]	206	<li><a href="#int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic </a></li>
				207	<li><a href="#int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic </a></li>
				208	<li><a href="#int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic </a></li>
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	209	<li><a href="#int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic </a></li>
				210	<li><a href="#int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic </a></li>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	211	</ol>
				212	</li>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	213	<li><a href="#int_overflow">Arithmetic with Overflow Intrinsics</a>
				214	<ol>
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	215	<li><a href="#int_sadd_overflow">'<tt>llvm.sadd.with.overflow.*</tt> Intrinsics</a></li>
				216	<li><a href="#int_uadd_overflow">'<tt>llvm.uadd.with.overflow.*</tt> Intrinsics</a></li>
				217	<li><a href="#int_ssub_overflow">'<tt>llvm.ssub.with.overflow.*</tt> Intrinsics</a></li>
				218	<li><a href="#int_usub_overflow">'<tt>llvm.usub.with.overflow.*</tt> Intrinsics</a></li>
				219	<li><a href="#int_smul_overflow">'<tt>llvm.smul.with.overflow.*</tt> Intrinsics</a></li>
Bill Wendling	b9a7327	2009-02-08 23:00:09 +0000	[diff] [blame]	220	<li><a href="#int_umul_overflow">'<tt>llvm.umul.with.overflow.*</tt> Intrinsics</a></li>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	221	</ol>
				222	</li>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	223	<li><a href="#int_debugger">Debugger intrinsics</a></li>
Jim Laskey	2211f49	2007-03-14 19:31:19 +0000	[diff] [blame]	224	<li><a href="#int_eh">Exception Handling intrinsics</a></li>
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	225	<li><a href="#int_trampoline">Trampoline Intrinsic</a>
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	226	<ol>
				227	<li><a href="#int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a></li>
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	228	</ol>
				229	</li>
Bill Wendling	f85850f	2008-11-18 22:10:53 +0000	[diff] [blame]	230	<li><a href="#int_atomics">Atomic intrinsics</a>
				231	<ol>
				232	<li><a href="#int_memory_barrier"><tt>llvm.memory_barrier</tt></a></li>
				233	<li><a href="#int_atomic_cmp_swap"><tt>llvm.atomic.cmp.swap</tt></a></li>
				234	<li><a href="#int_atomic_swap"><tt>llvm.atomic.swap</tt></a></li>
				235	<li><a href="#int_atomic_load_add"><tt>llvm.atomic.load.add</tt></a></li>
				236	<li><a href="#int_atomic_load_sub"><tt>llvm.atomic.load.sub</tt></a></li>
				237	<li><a href="#int_atomic_load_and"><tt>llvm.atomic.load.and</tt></a></li>
				238	<li><a href="#int_atomic_load_nand"><tt>llvm.atomic.load.nand</tt></a></li>
				239	<li><a href="#int_atomic_load_or"><tt>llvm.atomic.load.or</tt></a></li>
				240	<li><a href="#int_atomic_load_xor"><tt>llvm.atomic.load.xor</tt></a></li>
				241	<li><a href="#int_atomic_load_max"><tt>llvm.atomic.load.max</tt></a></li>
				242	<li><a href="#int_atomic_load_min"><tt>llvm.atomic.load.min</tt></a></li>
				243	<li><a href="#int_atomic_load_umax"><tt>llvm.atomic.load.umax</tt></a></li>
				244	<li><a href="#int_atomic_load_umin"><tt>llvm.atomic.load.umin</tt></a></li>
				245	</ol>
				246	</li>
Reid Spencer	5b2cb0f	2007-07-20 19:59:11 +0000	[diff] [blame]	247	<li><a href="#int_general">General intrinsics</a>
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	248	<ol>
Reid Spencer	5b2cb0f	2007-07-20 19:59:11 +0000	[diff] [blame]	249	<li><a href="#int_var_annotation">
Bill Wendling	1431331	2008-11-19 05:56:17 +0000	[diff] [blame]	250	'<tt>llvm.var.annotation</tt>' Intrinsic</a></li>
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	251	<li><a href="#int_annotation">
Bill Wendling	1431331	2008-11-19 05:56:17 +0000	[diff] [blame]	252	'<tt>llvm.annotation.*</tt>' Intrinsic</a></li>
Anton Korobeynikov	06cbb65	2008-01-15 22:31:34 +0000	[diff] [blame]	253	<li><a href="#int_trap">
Bill Wendling	1431331	2008-11-19 05:56:17 +0000	[diff] [blame]	254	'<tt>llvm.trap</tt>' Intrinsic</a></li>
				255	<li><a href="#int_stackprotector">
				256	'<tt>llvm.stackprotector</tt>' Intrinsic</a></li>
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	257	</ol>
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	258	</li>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	259	</ol>
				260	</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	261	</ol>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	262
				263	<div class="doc_author">
				264	<p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a>
				265	and <a href="mailto:vadve@cs.uiuc.edu">Vikram Adve</a></p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	266	</div>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	267
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	268	<!-- *********************************************************************** -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	269	<div class="doc_section"> <a name="abstract">Abstract </a></div>
				270	<!-- *********************************************************************** -->
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	271
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	272	<div class="doc_text">
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	273	<p>This document is a reference manual for the LLVM assembly language.
Bill Wendling	6e03f9a	2008-08-05 22:29:16 +0000	[diff] [blame]	274	LLVM is a Static Single Assignment (SSA) based representation that provides
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	275	type safety, low-level operations, flexibility, and the capability of
				276	representing 'all' high-level languages cleanly. It is the common code
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	277	representation used throughout all phases of the LLVM compilation
				278	strategy.</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	279	</div>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	280
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	281	<!-- *********************************************************************** -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	282	<div class="doc_section"> <a name="introduction">Introduction</a> </div>
				283	<!-- *********************************************************************** -->
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	284
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	285	<div class="doc_text">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	286
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	287	<p>The LLVM code representation is designed to be used in three
Gabor Greif	a54634a	2007-07-06 22:07:22 +0000	[diff] [blame]	288	different forms: as an in-memory compiler IR, as an on-disk bitcode
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	289	representation (suitable for fast loading by a Just-In-Time compiler),
				290	and as a human readable assembly language representation. This allows
				291	LLVM to provide a powerful intermediate representation for efficient
				292	compiler transformations and analysis, while providing a natural means
				293	to debug and visualize the transformations. The three different forms
				294	of LLVM are all equivalent. This document describes the human readable
				295	representation and notation.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	296
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	297	<p>The LLVM representation aims to be light-weight and low-level
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	298	while being expressive, typed, and extensible at the same time. It
				299	aims to be a "universal IR" of sorts, by being at a low enough level
				300	that high-level ideas may be cleanly mapped to it (similar to how
				301	microprocessors are "universal IR's", allowing many source languages to
				302	be mapped to them). By providing type information, LLVM can be used as
				303	the target of optimizations: for example, through pointer analysis, it
				304	can be proven that a C automatic variable is never accessed outside of
				305	the current function... allowing it to be promoted to a simple SSA
				306	value instead of a memory location.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	307
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	308	</div>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	309
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	310	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	311	<div class="doc_subsubsection"> <a name="wellformed">Well-Formedness</a> </div>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	312
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	313	<div class="doc_text">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	314
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	315	<p>It is important to note that this document describes 'well formed'
				316	LLVM assembly language. There is a difference between what the parser
				317	accepts and what is considered 'well formed'. For example, the
				318	following instruction is syntactically okay, but not well formed:</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	319
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	320	<div class="doc_code">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	321	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	322	%x = <a href="#i_add">add</a> i32 1, %x
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	323	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	324	</div>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	325
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	326	<p>...because the definition of <tt>%x</tt> does not dominate all of
				327	its uses. The LLVM infrastructure provides a verification pass that may
				328	be used to verify that an LLVM module is well formed. This pass is
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	329	automatically run by the parser after parsing input assembly and by
Gabor Greif	a54634a	2007-07-06 22:07:22 +0000	[diff] [blame]	330	the optimizer before it outputs bitcode. The violations pointed out
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	331	by the verifier pass indicate bugs in transformation passes or input to
				332	the parser.</p>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	333	</div>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	334
Chris Lattner	87a3dbe	2007-10-03 17:34:29 +0000	[diff] [blame]	335	<!-- Describe the typesetting conventions here. -->
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	336
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	337	<!-- *********************************************************************** -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	338	<div class="doc_section"> <a name="identifiers">Identifiers</a> </div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	339	<!-- *********************************************************************** -->
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	340
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	341	<div class="doc_text">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	342
Reid Spencer	b23b65f	2007-08-07 14:34:28 +0000	[diff] [blame]	343	<p>LLVM identifiers come in two basic types: global and local. Global
				344	identifiers (functions, global variables) begin with the @ character. Local
				345	identifiers (register names, types) begin with the % character. Additionally,
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	346	there are three different formats for identifiers, for different purposes:</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	347
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	348	<ol>
Reid Spencer	b23b65f	2007-08-07 14:34:28 +0000	[diff] [blame]	349	<li>Named values are represented as a string of characters with their prefix.
				350	For example, %foo, @DivisionByZero, %a.really.long.identifier. The actual
				351	regular expression used is '<tt>[%@][a-zA-Z$._][a-zA-Z$._0-9]*</tt>'.
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	352	Identifiers which require other characters in their names can be surrounded
Daniel Dunbar	0f8155a	2008-10-14 23:51:43 +0000	[diff] [blame]	353	with quotes. Special characters may be escaped using "\xx" where xx is the
				354	ASCII code for the character in hexadecimal. In this way, any character can
				355	be used in a name value, even quotes themselves.
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	356
Reid Spencer	b23b65f	2007-08-07 14:34:28 +0000	[diff] [blame]	357	<li>Unnamed values are represented as an unsigned numeric value with their
				358	prefix. For example, %12, @2, %44.</li>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	359
Reid Spencer	8f08d80	2004-12-09 18:02:53 +0000	[diff] [blame]	360	<li>Constants, which are described in a <a href="#constants">section about
				361	constants</a>, below.</li>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	362	</ol>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	363
Reid Spencer	b23b65f	2007-08-07 14:34:28 +0000	[diff] [blame]	364	<p>LLVM requires that values start with a prefix for two reasons: Compilers
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	365	don't need to worry about name clashes with reserved words, and the set of
				366	reserved words may be expanded in the future without penalty. Additionally,
				367	unnamed identifiers allow a compiler to quickly come up with a temporary
				368	variable without having to avoid symbol table conflicts.</p>
				369
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	370	<p>Reserved words in LLVM are very similar to reserved words in other
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	371	languages. There are keywords for different opcodes
				372	('<tt><a href="#i_add">add</a></tt>',
				373	'<tt><a href="#i_bitcast">bitcast</a></tt>',
				374	'<tt><a href="#i_ret">ret</a></tt>', etc...), for primitive type names ('<tt><a
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	375	href="#t_void">void</a></tt>', '<tt><a href="#t_primitive">i32</a></tt>', etc...),
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	376	and others. These reserved words cannot conflict with variable names, because
Reid Spencer	b23b65f	2007-08-07 14:34:28 +0000	[diff] [blame]	377	none of them start with a prefix character ('%' or '@').</p>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	378
				379	<p>Here is an example of LLVM code to multiply the integer variable
				380	'<tt>%X</tt>' by 8:</p>
				381
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	382	<p>The easy way:</p>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	383
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	384	<div class="doc_code">
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	385	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	386	%result = <a href="#i_mul">mul</a> i32 %X, 8
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	387	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	388	</div>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	389
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	390	<p>After strength reduction:</p>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	391
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	392	<div class="doc_code">
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	393	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	394	%result = <a href="#i_shl">shl</a> i32 %X, i8 3
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	395	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	396	</div>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	397
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	398	<p>And the hard way:</p>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	399
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	400	<div class="doc_code">
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	401	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	402	<a href="#i_add">add</a> i32 %X, %X <i>; yields {i32}:%0</i>
				403	<a href="#i_add">add</a> i32 %0, %0 <i>; yields {i32}:%1</i>
				404	%result = <a href="#i_add">add</a> i32 %1, %1
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	405	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	406	</div>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	407
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	408	<p>This last way of multiplying <tt>%X</tt> by 8 illustrates several
				409	important lexical features of LLVM:</p>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	410
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	411	<ol>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	412
				413	<li>Comments are delimited with a '<tt>;</tt>' and go until the end of
				414	line.</li>
				415
				416	<li>Unnamed temporaries are created when the result of a computation is not
				417	assigned to a named value.</li>
				418
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	419	<li>Unnamed temporaries are numbered sequentially</li>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	420
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	421	</ol>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	422
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	423	<p>...and it also shows a convention that we follow in this document. When
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	424	demonstrating instructions, we will follow an instruction with a comment that
				425	defines the type and name of value produced. Comments are shown in italic
				426	text.</p>
				427
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	428	</div>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	429
				430	<!-- *********************************************************************** -->
				431	<div class="doc_section"> <a name="highlevel">High Level Structure</a> </div>
				432	<!-- *********************************************************************** -->
				433
				434	<!-- ======================================================================= -->
				435	<div class="doc_subsection"> <a name="modulestructure">Module Structure</a>
				436	</div>
				437
				438	<div class="doc_text">
				439
				440	<p>LLVM programs are composed of "Module"s, each of which is a
				441	translation unit of the input programs. Each module consists of
				442	functions, global variables, and symbol table entries. Modules may be
				443	combined together with the LLVM linker, which merges function (and
				444	global variable) definitions, resolves forward declarations, and merges
				445	symbol table entries. Here is an example of the "hello world" module:</p>
				446
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	447	<div class="doc_code">
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	448	<pre><i>; Declare the string constant as a global constant...</i>
Chris Lattner	2b0bf4f	2007-06-12 17:00:26 +0000	[diff] [blame]	449	<a href="#identifiers">@.LC0</a> = <a href="#linkage_internal">internal</a> <a
				450	href="#globalvars">constant</a> <a href="#t_array">[13 x i8]</a> c"hello world\0A\00" <i>; [13 x i8]*</i>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	451
				452	<i>; External declaration of the puts function</i>
Chris Lattner	2b0bf4f	2007-06-12 17:00:26 +0000	[diff] [blame]	453	<a href="#functionstructure">declare</a> i32 @puts(i8 ) <i>; i32(i8 )* </i>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	454
				455	<i>; Definition of main function</i>
Chris Lattner	2b0bf4f	2007-06-12 17:00:26 +0000	[diff] [blame]	456	define i32 @main() { <i>; i32()* </i>
Dan Gohman	623806e	2009-01-04 23:44:43 +0000	[diff] [blame]	457	<i>; Convert [13 x i8]* to i8 *...</i>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	458	%cast210 = <a
Dan Gohman	623806e	2009-01-04 23:44:43 +0000	[diff] [blame]	459	href="#i_getelementptr">getelementptr</a> [13 x i8]* @.LC0, i64 0, i64 0 <i>; i8 *</i>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	460
				461	<i>; Call puts function to write out the string to stdout...</i>
				462	<a
Chris Lattner	2b0bf4f	2007-06-12 17:00:26 +0000	[diff] [blame]	463	href="#i_call">call</a> i32 @puts(i8 * %cast210) <i>; i32</i>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	464	<a
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	465	href="#i_ret">ret</a> i32 0<br>}<br>
				466	</pre>
				467	</div>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	468
				469	<p>This example is made up of a <a href="#globalvars">global variable</a>
				470	named "<tt>.LC0</tt>", an external declaration of the "<tt>puts</tt>"
				471	function, and a <a href="#functionstructure">function definition</a>
				472	for "<tt>main</tt>".</p>
				473
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	474	<p>In general, a module is made up of a list of global values,
				475	where both functions and global variables are global values. Global values are
				476	represented by a pointer to a memory location (in this case, a pointer to an
				477	array of char, and a pointer to a function), and have one of the following <a
				478	href="#linkage">linkage types</a>.</p>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	479
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	480	</div>
				481
				482	<!-- ======================================================================= -->
				483	<div class="doc_subsection">
				484	<a name="linkage">Linkage Types</a>
				485	</div>
				486
				487	<div class="doc_text">
				488
				489	<p>
				490	All Global Variables and Functions have one of the following types of linkage:
				491	</p>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	492
				493	<dl>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	494
Rafael Espindola	6de96a1	2009-01-15 20:18:42 +0000	[diff] [blame]	495	<dt><tt><b><a name="linkage_private">private</a></b></tt>: </dt>
				496
				497	<dd>Global values with private linkage are only directly accessible by
				498	objects in the current module. In particular, linking code into a module with
				499	an private global value may cause the private to be renamed as necessary to
				500	avoid collisions. Because the symbol is private to the module, all
				501	references can be updated. This doesn't show up in any symbol table in the
				502	object file.
				503	</dd>
				504
Dale Johannesen	4188aad	2008-05-23 23:13:41 +0000	[diff] [blame]	505	<dt><tt><b><a name="linkage_internal">internal</a></b></tt>: </dt>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	506
Duncan Sands	35e43c1	2009-01-16 09:29:46 +0000	[diff] [blame]	507	<dd> Similar to private, but the value shows as a local symbol (STB_LOCAL in
Rafael Espindola	6de96a1	2009-01-15 20:18:42 +0000	[diff] [blame]	508	the case of ELF) in the object file. This corresponds to the notion of the
Chris Lattner	e20b470	2007-01-14 06:51:48 +0000	[diff] [blame]	509	'<tt>static</tt>' keyword in C.
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	510	</dd>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	511
Chris Lattner	184f1be	2009-04-13 05:44:34 +0000	[diff] [blame]	512	<dt><tt><b><a name="available_externally">available_externally</a></b></tt>:
				513	</dt>
				514
				515	<dd>Globals with "<tt>available_externally</tt>" linkage are never emitted
				516	into the object file corresponding to the LLVM module. They exist to
				517	allow inlining and other optimizations to take place given knowledge of the
				518	definition of the global, which is known to be somewhere outside the module.
				519	Globals with <tt>available_externally</tt> linkage are allowed to be discarded
				520	at will, and are otherwise the same as <tt>linkonce_odr</tt>. This linkage
				521	type is only allowed on definitions, not declarations.</dd>
				522
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	523	<dt><tt><b><a name="linkage_linkonce">linkonce</a></b></tt>: </dt>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	524
Chris Lattner	e20b470	2007-01-14 06:51:48 +0000	[diff] [blame]	525	<dd>Globals with "<tt>linkonce</tt>" linkage are merged with other globals of
				526	the same name when linkage occurs. This is typically used to implement
				527	inline functions, templates, or other code which must be generated in each
				528	translation unit that uses it. Unreferenced <tt>linkonce</tt> globals are
				529	allowed to be discarded.
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	530	</dd>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	531
Dale Johannesen	4188aad	2008-05-23 23:13:41 +0000	[diff] [blame]	532	<dt><tt><b><a name="linkage_common">common</a></b></tt>: </dt>
				533
				534	<dd>"<tt>common</tt>" linkage is exactly the same as <tt>linkonce</tt>
				535	linkage, except that unreferenced <tt>common</tt> globals may not be
				536	discarded. This is used for globals that may be emitted in multiple
				537	translation units, but that are not guaranteed to be emitted into every
				538	translation unit that uses them. One example of this is tentative
				539	definitions in C, such as "<tt>int X;</tt>" at global scope.
				540	</dd>
				541
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	542	<dt><tt><b><a name="linkage_weak">weak</a></b></tt>: </dt>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	543
Dale Johannesen	4188aad	2008-05-23 23:13:41 +0000	[diff] [blame]	544	<dd>"<tt>weak</tt>" linkage is the same as <tt>common</tt> linkage, except
				545	that some targets may choose to emit different assembly sequences for them
				546	for target-dependent reasons. This is used for globals that are declared
				547	"weak" in C source code.
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	548	</dd>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	549
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	550	<dt><tt><b><a name="linkage_appending">appending</a></b></tt>: </dt>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	551
				552	<dd>"<tt>appending</tt>" linkage may only be applied to global variables of
				553	pointer to array type. When two global variables with appending linkage are
				554	linked together, the two global arrays are appended together. This is the
				555	LLVM, typesafe, equivalent of having the system linker append together
				556	"sections" with identical names when .o files are linked.
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	557	</dd>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	558
Anton Korobeynikov	a0554d9	2007-01-12 19:20:47 +0000	[diff] [blame]	559	<dt><tt><b><a name="linkage_externweak">extern_weak</a></b></tt>: </dt>
Duncan Sands	12da8ce	2009-03-07 15:45:40 +0000	[diff] [blame]	560
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	561	<dd>The semantics of this linkage follow the ELF object file model: the
				562	symbol is weak until linked, if not linked, the symbol becomes null instead
				563	of being an undefined reference.
Anton Korobeynikov	a0554d9	2007-01-12 19:20:47 +0000	[diff] [blame]	564	</dd>
Anton Korobeynikov	a0554d9	2007-01-12 19:20:47 +0000	[diff] [blame]	565
Duncan Sands	12da8ce	2009-03-07 15:45:40 +0000	[diff] [blame]	566	<dt><tt><b><a name="linkage_linkonce">linkonce_odr</a></b></tt>: </dt>
Duncan Sands	12da8ce	2009-03-07 15:45:40 +0000	[diff] [blame]	567	<dt><tt><b><a name="linkage_weak">weak_odr</a></b></tt>: </dt>
Chris Lattner	184f1be	2009-04-13 05:44:34 +0000	[diff] [blame]	568	<dd>Some languages allow differing globals to be merged, such as two
Duncan Sands	12da8ce	2009-03-07 15:45:40 +0000	[diff] [blame]	569	functions with different semantics. Other languages, such as <tt>C++</tt>,
				570	ensure that only equivalent globals are ever merged (the "one definition
Chris Lattner	184f1be	2009-04-13 05:44:34 +0000	[diff] [blame]	571	rule" - "ODR"). Such languages can use the <tt>linkonce_odr</tt>
Duncan Sands	4581beb	2009-03-11 20:14:15 +0000	[diff] [blame]	572	and <tt>weak_odr</tt> linkage types to indicate that the global will only
				573	be merged with equivalent globals. These linkage types are otherwise the
				574	same as their non-<tt>odr</tt> versions.
Duncan Sands	12da8ce	2009-03-07 15:45:40 +0000	[diff] [blame]	575	</dd>
				576
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	577	<dt><tt><b><a name="linkage_external">externally visible</a></b></tt>:</dt>
Chris Lattner	d79749a	2004-12-09 16:36:40 +0000	[diff] [blame]	578
				579	<dd>If none of the above identifiers are used, the global is externally
				580	visible, meaning that it participates in linkage and can be used to resolve
				581	external symbol references.
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	582	</dd>
Reid Spencer	7972c47	2007-04-11 23:49:50 +0000	[diff] [blame]	583	</dl>
Anton Korobeynikov	d61d39e	2006-09-14 18:23:27 +0000	[diff] [blame]	584
Anton Korobeynikov	d61d39e	2006-09-14 18:23:27 +0000	[diff] [blame]	585	<p>
				586	The next two types of linkage are targeted for Microsoft Windows platform
				587	only. They are designed to support importing (exporting) symbols from (to)
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	588	DLLs (Dynamic Link Libraries).
Anton Korobeynikov	d61d39e	2006-09-14 18:23:27 +0000	[diff] [blame]	589	</p>
				590
Anton Korobeynikov	a0554d9	2007-01-12 19:20:47 +0000	[diff] [blame]	591	<dl>
Anton Korobeynikov	d61d39e	2006-09-14 18:23:27 +0000	[diff] [blame]	592	<dt><tt><b><a name="linkage_dllimport">dllimport</a></b></tt>: </dt>
				593
				594	<dd>"<tt>dllimport</tt>" linkage causes the compiler to reference a function
				595	or variable via a global pointer to a pointer that is set up by the DLL
				596	exporting the symbol. On Microsoft Windows targets, the pointer name is
Dan Gohman	33a9cef	2009-01-12 21:35:55 +0000	[diff] [blame]	597	formed by combining <code>__imp_</code> and the function or variable name.
Anton Korobeynikov	d61d39e	2006-09-14 18:23:27 +0000	[diff] [blame]	598	</dd>
				599
				600	<dt><tt><b><a name="linkage_dllexport">dllexport</a></b></tt>: </dt>
				601
				602	<dd>"<tt>dllexport</tt>" linkage causes the compiler to provide a global
				603	pointer to a pointer in a DLL, so that it can be referenced with the
				604	<tt>dllimport</tt> attribute. On Microsoft Windows targets, the pointer
Dan Gohman	33a9cef	2009-01-12 21:35:55 +0000	[diff] [blame]	605	name is formed by combining <code>__imp_</code> and the function or variable
Anton Korobeynikov	d61d39e	2006-09-14 18:23:27 +0000	[diff] [blame]	606	name.
				607	</dd>
				608
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	609	</dl>
				610
Dan Gohman	8ef4498	2008-11-24 17:18:39 +0000	[diff] [blame]	611	<p>For example, since the "<tt>.LC0</tt>"
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	612	variable is defined to be internal, if another module defined a "<tt>.LC0</tt>"
				613	variable and was linked with this one, one of the two would be renamed,
				614	preventing a collision. Since "<tt>main</tt>" and "<tt>puts</tt>" are
				615	external (i.e., lacking any linkage declarations), they are accessible
Reid Spencer	92c671e	2007-01-05 00:59:10 +0000	[diff] [blame]	616	outside of the current module.</p>
				617	<p>It is illegal for a function <i>declaration</i>
Duncan Sands	e288105	2009-03-11 08:08:06 +0000	[diff] [blame]	618	to have any linkage type other than "externally visible", <tt>dllimport</tt>
				619	or <tt>extern_weak</tt>.</p>
Duncan Sands	12da8ce	2009-03-07 15:45:40 +0000	[diff] [blame]	620	<p>Aliases can have only <tt>external</tt>, <tt>internal</tt>, <tt>weak</tt>
				621	or <tt>weak_odr</tt> linkages.</p>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	622	</div>
				623
				624	<!-- ======================================================================= -->
				625	<div class="doc_subsection">
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	626	<a name="callingconv">Calling Conventions</a>
				627	</div>
				628
				629	<div class="doc_text">
				630
				631	<p>LLVM <a href="#functionstructure">functions</a>, <a href="#i_call">calls</a>
				632	and <a href="#i_invoke">invokes</a> can all have an optional calling convention
				633	specified for the call. The calling convention of any pair of dynamic
				634	caller/callee must match, or the behavior of the program is undefined. The
				635	following calling conventions are supported by LLVM, and more may be added in
				636	the future:</p>
				637
				638	<dl>
				639	<dt><b>"<tt>ccc</tt>" - The C calling convention</b>:</dt>
				640
				641	<dd>This calling convention (the default if no other calling convention is
				642	specified) matches the target C calling conventions. This calling convention
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	643	supports varargs function calls and tolerates some mismatch in the declared
Reid Spencer	72ba499	2006-12-31 21:30:18 +0000	[diff] [blame]	644	prototype and implemented declaration of the function (as does normal C).
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	645	</dd>
				646
				647	<dt><b>"<tt>fastcc</tt>" - The fast calling convention</b>:</dt>
				648
				649	<dd>This calling convention attempts to make calls as fast as possible
				650	(e.g. by passing things in registers). This calling convention allows the
				651	target to use whatever tricks it wants to produce fast code for the target,
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	652	without having to conform to an externally specified ABI (Application Binary
				653	Interface). Implementations of this convention should allow arbitrary
Arnold Schwaighofer	2c6b888	2008-05-14 09:17:12 +0000	[diff] [blame]	654	<a href="CodeGenerator.html#tailcallopt">tail call optimization</a> to be
				655	supported. This calling convention does not support varargs and requires the
				656	prototype of all callees to exactly match the prototype of the function
				657	definition.
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	658	</dd>
				659
				660	<dt><b>"<tt>coldcc</tt>" - The cold calling convention</b>:</dt>
				661
				662	<dd>This calling convention attempts to make code in the caller as efficient
				663	as possible under the assumption that the call is not commonly executed. As
				664	such, these calls often preserve all registers so that the call does not break
				665	any live ranges in the caller side. This calling convention does not support
				666	varargs and requires the prototype of all callees to exactly match the
				667	prototype of the function definition.
				668	</dd>
				669
Chris Lattner	573f64e	2005-05-07 01:46:40 +0000	[diff] [blame]	670	<dt><b>"<tt>cc <<em>n</em>></tt>" - Numbered convention</b>:</dt>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	671
				672	<dd>Any calling convention may be specified by number, allowing
				673	target-specific calling conventions to be used. Target specific calling
				674	conventions start at 64.
				675	</dd>
Chris Lattner	573f64e	2005-05-07 01:46:40 +0000	[diff] [blame]	676	</dl>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	677
				678	<p>More calling conventions can be added/defined on an as-needed basis, to
				679	support pascal conventions or any other well-known target-independent
				680	convention.</p>
				681
				682	</div>
				683
				684	<!-- ======================================================================= -->
				685	<div class="doc_subsection">
Anton Korobeynikov	c7f9f3d	2007-01-23 12:35:46 +0000	[diff] [blame]	686	<a name="visibility">Visibility Styles</a>
				687	</div>
				688
				689	<div class="doc_text">
				690
				691	<p>
				692	All Global Variables and Functions have one of the following visibility styles:
				693	</p>
				694
				695	<dl>
				696	<dt><b>"<tt>default</tt>" - Default style</b>:</dt>
				697
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	698	<dd>On targets that use the ELF object file format, default visibility means
				699	that the declaration is visible to other
Anton Korobeynikov	c7f9f3d	2007-01-23 12:35:46 +0000	[diff] [blame]	700	modules and, in shared libraries, means that the declared entity may be
				701	overridden. On Darwin, default visibility means that the declaration is
				702	visible to other modules. Default visibility corresponds to "external
				703	linkage" in the language.
				704	</dd>
				705
				706	<dt><b>"<tt>hidden</tt>" - Hidden style</b>:</dt>
				707
				708	<dd>Two declarations of an object with hidden visibility refer to the same
				709	object if they are in the same shared object. Usually, hidden visibility
				710	indicates that the symbol will not be placed into the dynamic symbol table,
				711	so no other module (executable or shared library) can reference it
				712	directly.
				713	</dd>
				714
Anton Korobeynikov	39f3cff	2007-04-29 18:35:00 +0000	[diff] [blame]	715	<dt><b>"<tt>protected</tt>" - Protected style</b>:</dt>
				716
				717	<dd>On ELF, protected visibility indicates that the symbol will be placed in
				718	the dynamic symbol table, but that references within the defining module will
				719	bind to the local symbol. That is, the symbol cannot be overridden by another
				720	module.
				721	</dd>
Anton Korobeynikov	c7f9f3d	2007-01-23 12:35:46 +0000	[diff] [blame]	722	</dl>
				723
				724	</div>
				725
				726	<!-- ======================================================================= -->
				727	<div class="doc_subsection">
Chris Lattner	bc08821	2009-01-11 20:53:49 +0000	[diff] [blame]	728	<a name="namedtypes">Named Types</a>
				729	</div>
				730
				731	<div class="doc_text">
				732
				733	<p>LLVM IR allows you to specify name aliases for certain types. This can make
				734	it easier to read the IR and make the IR more condensed (particularly when
				735	recursive types are involved). An example of a name specification is:
				736	</p>
				737
				738	<div class="doc_code">
				739	<pre>
				740	%mytype = type { %mytype*, i32 }
				741	</pre>
				742	</div>
				743
				744	<p>You may give a name to any <a href="#typesystem">type</a> except "<a
				745	href="t_void">void</a>". Type name aliases may be used anywhere a type is
				746	expected with the syntax "%mytype".</p>
				747
				748	<p>Note that type names are aliases for the structural type that they indicate,
				749	and that you can therefore specify multiple names for the same type. This often
				750	leads to confusing behavior when dumping out a .ll file. Since LLVM IR uses
				751	structural typing, the name is not part of the type. When printing out LLVM IR,
				752	the printer will pick <em>one name</em> to render all types of a particular
				753	shape. This means that if you have code where two different source types end up
				754	having the same LLVM type, that the dumper will sometimes print the "wrong" or
				755	unexpected type. This is an important design point and isn't going to
				756	change.</p>
				757
				758	</div>
				759
Chris Lattner	bc08821	2009-01-11 20:53:49 +0000	[diff] [blame]	760	<!-- ======================================================================= -->
				761	<div class="doc_subsection">
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	762	<a name="globalvars">Global Variables</a>
				763	</div>
				764
				765	<div class="doc_text">
				766
Chris Lattner	5d5aede	2005-02-12 19:30:21 +0000	[diff] [blame]	767	<p>Global variables define regions of memory allocated at compilation time
Chris Lattner	662c872	2005-11-12 00:45:07 +0000	[diff] [blame]	768	instead of run-time. Global variables may optionally be initialized, may have
Anton Korobeynikov	a97b694	2007-04-25 14:27:10 +0000	[diff] [blame]	769	an explicit section to be placed in, and may have an optional explicit alignment
				770	specified. A variable may be defined as "thread_local", which means that it
				771	will not be shared by threads (each thread will have a separated copy of the
				772	variable). A variable may be defined as a global "constant," which indicates
				773	that the contents of the variable will <b>never</b> be modified (enabling better
Chris Lattner	5d5aede	2005-02-12 19:30:21 +0000	[diff] [blame]	774	optimization, allowing the global data to be placed in the read-only section of
				775	an executable, etc). Note that variables that need runtime initialization
John Criswell	4c0cf7f	2005-10-24 16:17:18 +0000	[diff] [blame]	776	cannot be marked "constant" as there is a store to the variable.</p>
Chris Lattner	5d5aede	2005-02-12 19:30:21 +0000	[diff] [blame]	777
				778	<p>
				779	LLVM explicitly allows <em>declarations</em> of global variables to be marked
				780	constant, even if the final definition of the global is not. This capability
				781	can be used to enable slightly better optimization of the program, but requires
				782	the language definition to guarantee that optimizations based on the
				783	'constantness' are valid for the translation units that do not include the
				784	definition.
				785	</p>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	786
				787	<p>As SSA values, global variables define pointer values that are in
				788	scope (i.e. they dominate) all basic blocks in the program. Global
				789	variables always define a pointer to their "content" type because they
				790	describe a region of memory, and all memory objects in LLVM are
				791	accessed through pointers.</p>
				792
Christopher Lamb	308121c	2007-12-11 09:31:00 +0000	[diff] [blame]	793	<p>A global variable may be declared to reside in a target-specifc numbered
				794	address space. For targets that support them, address spaces may affect how
				795	optimizations are performed and/or what target instructions are used to access
Christopher Lamb	25f5076	2007-12-12 08:44:39 +0000	[diff] [blame]	796	the variable. The default address space is zero. The address space qualifier
				797	must precede any other attributes.</p>
Christopher Lamb	308121c	2007-12-11 09:31:00 +0000	[diff] [blame]	798
Chris Lattner	662c872	2005-11-12 00:45:07 +0000	[diff] [blame]	799	<p>LLVM allows an explicit section to be specified for globals. If the target
				800	supports it, it will emit globals to the section specified.</p>
				801
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	802	<p>An explicit alignment may be specified for a global. If not present, or if
				803	the alignment is set to zero, the alignment of the global is set by the target
				804	to whatever it feels convenient. If an explicit alignment is specified, the
				805	global is forced to have at least that much alignment. All alignments must be
				806	a power of 2.</p>
				807
Christopher Lamb	308121c	2007-12-11 09:31:00 +0000	[diff] [blame]	808	<p>For example, the following defines a global in a numbered address space with
				809	an initializer, section, and alignment:</p>
Chris Lattner	5760c50	2007-01-14 00:27:09 +0000	[diff] [blame]	810
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	811	<div class="doc_code">
Chris Lattner	5760c50	2007-01-14 00:27:09 +0000	[diff] [blame]	812	<pre>
Dan Gohman	aaa679b	2009-01-11 00:40:00 +0000	[diff] [blame]	813	@G = addrspace(5) constant float 1.0, section "foo", align 4
Chris Lattner	5760c50	2007-01-14 00:27:09 +0000	[diff] [blame]	814	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	815	</div>
Chris Lattner	5760c50	2007-01-14 00:27:09 +0000	[diff] [blame]	816
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	817	</div>
				818
				819
				820	<!-- ======================================================================= -->
				821	<div class="doc_subsection">
				822	<a name="functionstructure">Functions</a>
				823	</div>
				824
				825	<div class="doc_text">
				826
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	827	<p>LLVM function definitions consist of the "<tt>define</tt>" keyord,
				828	an optional <a href="#linkage">linkage type</a>, an optional
Anton Korobeynikov	c7f9f3d	2007-01-23 12:35:46 +0000	[diff] [blame]	829	<a href="#visibility">visibility style</a>, an optional
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	830	<a href="#callingconv">calling convention</a>, a return type, an optional
				831	<a href="#paramattrs">parameter attribute</a> for the return type, a function
				832	name, a (possibly empty) argument list (each with optional
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	833	<a href="#paramattrs">parameter attributes</a>), optional
				834	<a href="#fnattrs">function attributes</a>, an optional section,
				835	an optional alignment, an optional <a href="#gc">garbage collector name</a>,
Chris Lattner	cbc4d2a	2008-10-04 18:10:21 +0000	[diff] [blame]	836	an opening curly brace, a list of basic blocks, and a closing curly brace.
Anton Korobeynikov	c7f9f3d	2007-01-23 12:35:46 +0000	[diff] [blame]	837
				838	LLVM function declarations consist of the "<tt>declare</tt>" keyword, an
				839	optional <a href="#linkage">linkage type</a>, an optional
				840	<a href="#visibility">visibility style</a>, an optional
				841	<a href="#callingconv">calling convention</a>, a return type, an optional
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	842	<a href="#paramattrs">parameter attribute</a> for the return type, a function
Gordon Henriksen	71183b6	2007-12-10 03:18:06 +0000	[diff] [blame]	843	name, a possibly empty list of arguments, an optional alignment, and an optional
Gordon Henriksen	dc5cafb	2007-12-10 03:30:21 +0000	[diff] [blame]	844	<a href="#gc">garbage collector name</a>.</p>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	845
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	846	<p>A function definition contains a list of basic blocks, forming the CFG
				847	(Control Flow Graph) for
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	848	the function. Each basic block may optionally start with a label (giving the
				849	basic block a symbol table entry), contains a list of instructions, and ends
				850	with a <a href="#terminators">terminator</a> instruction (such as a branch or
				851	function return).</p>
				852
Chris Lattner	a59fb10	2007-06-08 16:52:14 +0000	[diff] [blame]	853	<p>The first basic block in a function is special in two ways: it is immediately
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	854	executed on entrance to the function, and it is not allowed to have predecessor
				855	basic blocks (i.e. there can not be any branches to the entry block of a
				856	function). Because the block can have no predecessors, it also cannot have any
				857	<a href="#i_phi">PHI nodes</a>.</p>
				858
Chris Lattner	662c872	2005-11-12 00:45:07 +0000	[diff] [blame]	859	<p>LLVM allows an explicit section to be specified for functions. If the target
				860	supports it, it will emit functions to the section specified.</p>
				861
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	862	<p>An explicit alignment may be specified for a function. If not present, or if
				863	the alignment is set to zero, the alignment of the function is set by the target
				864	to whatever it feels convenient. If an explicit alignment is specified, the
				865	function is forced to have at least that much alignment. All alignments must be
				866	a power of 2.</p>
				867
Devang Patel	0225623	2008-10-07 17:48:33 +0000	[diff] [blame]	868	<h5>Syntax:</h5>
				869
				870	<div class="doc_code">
Chris Lattner	0ae0209	2008-10-13 16:55:18 +0000	[diff] [blame]	871	<tt>
				872	define [<a href="#linkage">linkage</a>] [<a href="#visibility">visibility</a>]
				873	[<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>]
				874	<ResultType> @<FunctionName> ([argument list])
				875	[<a href="#fnattrs">fn Attrs</a>] [section "name"] [align N]
				876	[<a href="#gc">gc</a>] { ... }
				877	</tt>
Devang Patel	0225623	2008-10-07 17:48:33 +0000	[diff] [blame]	878	</div>
				879
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	880	</div>
				881
Anton Korobeynikov	a97b694	2007-04-25 14:27:10 +0000	[diff] [blame]	882
				883	<!-- ======================================================================= -->
				884	<div class="doc_subsection">
				885	<a name="aliasstructure">Aliases</a>
				886	</div>
				887	<div class="doc_text">
				888	<p>Aliases act as "second name" for the aliasee value (which can be either
Anton Korobeynikov	25b2e82	2008-03-22 08:36:14 +0000	[diff] [blame]	889	function, global variable, another alias or bitcast of global value). Aliases
				890	may have an optional <a href="#linkage">linkage type</a>, and an
Anton Korobeynikov	a97b694	2007-04-25 14:27:10 +0000	[diff] [blame]	891	optional <a href="#visibility">visibility style</a>.</p>
				892
				893	<h5>Syntax:</h5>
				894
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	895	<div class="doc_code">
Bill Wendling	2d8b9a8	2007-05-29 09:42:13 +0000	[diff] [blame]	896	<pre>
Duncan Sands	7e99a94	2008-09-12 20:48:21 +0000	[diff] [blame]	897	@<Name> = alias [Linkage] [Visibility] <AliaseeTy> @<Aliasee>
Bill Wendling	2d8b9a8	2007-05-29 09:42:13 +0000	[diff] [blame]	898	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	899	</div>
Anton Korobeynikov	a97b694	2007-04-25 14:27:10 +0000	[diff] [blame]	900
				901	</div>
				902
				903
				904
Chris Lattner	91c15c4	2006-01-23 23:23:47 +0000	[diff] [blame]	905	<!-- ======================================================================= -->
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	906	<div class="doc_subsection"><a name="paramattrs">Parameter Attributes</a></div>
				907	<div class="doc_text">
				908	<p>The return type and each parameter of a function type may have a set of
				909	<i>parameter attributes</i> associated with them. Parameter attributes are
				910	used to communicate additional information about the result or parameters of
Duncan Sands	ad0ea2d	2007-11-27 13:23:08 +0000	[diff] [blame]	911	a function. Parameter attributes are considered to be part of the function,
				912	not of the function type, so functions with different parameter attributes
				913	can have the same function type.</p>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	914
Reid Spencer	cf7ebf5	2007-01-15 18:27:39 +0000	[diff] [blame]	915	<p>Parameter attributes are simple keywords that follow the type specified. If
				916	multiple parameter attributes are needed, they are space separated. For
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	917	example:</p>
				918
				919	<div class="doc_code">
				920	<pre>
Nick Lewycky	dac78d8	2009-02-15 23:06:14 +0000	[diff] [blame]	921	declare i32 @printf(i8* noalias nocapture, ...)
Chris Lattner	d2597d7	2008-10-04 18:33:34 +0000	[diff] [blame]	922	declare i32 @atoi(i8 zeroext)
				923	declare signext i8 @returns_signed_char()
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	924	</pre>
				925	</div>
				926
Duncan Sands	ad0ea2d	2007-11-27 13:23:08 +0000	[diff] [blame]	927	<p>Note that any attributes for the function result (<tt>nounwind</tt>,
				928	<tt>readonly</tt>) come immediately after the argument list.</p>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	929
Anton Korobeynikov	a0554d9	2007-01-12 19:20:47 +0000	[diff] [blame]	930	<p>Currently, only the following parameter attributes are defined:</p>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	931	<dl>
Reid Spencer	314e1cb	2007-07-19 23:13:04 +0000	[diff] [blame]	932	<dt><tt>zeroext</tt></dt>
Chris Lattner	d2597d7	2008-10-04 18:33:34 +0000	[diff] [blame]	933	<dd>This indicates to the code generator that the parameter or return value
				934	should be zero-extended to a 32-bit value by the caller (for a parameter)
				935	or the callee (for a return value).</dd>
Chris Lattner	5cee13f	2008-01-11 06:20:47 +0000	[diff] [blame]	936
Reid Spencer	314e1cb	2007-07-19 23:13:04 +0000	[diff] [blame]	937	<dt><tt>signext</tt></dt>
Chris Lattner	d2597d7	2008-10-04 18:33:34 +0000	[diff] [blame]	938	<dd>This indicates to the code generator that the parameter or return value
				939	should be sign-extended to a 32-bit value by the caller (for a parameter)
				940	or the callee (for a return value).</dd>
Chris Lattner	5cee13f	2008-01-11 06:20:47 +0000	[diff] [blame]	941
Anton Korobeynikov	e816685	2007-01-28 14:30:45 +0000	[diff] [blame]	942	<dt><tt>inreg</tt></dt>
Dale Johannesen	c50ada2	2008-09-25 20:47:45 +0000	[diff] [blame]	943	<dd>This indicates that this parameter or return value should be treated
				944	in a special target-dependent fashion during while emitting code for a
				945	function call or return (usually, by putting it in a register as opposed
Chris Lattner	d2597d7	2008-10-04 18:33:34 +0000	[diff] [blame]	946	to memory, though some targets use it to distinguish between two different
				947	kinds of registers). Use of this attribute is target-specific.</dd>
Chris Lattner	5cee13f	2008-01-11 06:20:47 +0000	[diff] [blame]	948
Duncan Sands	2a1d8ba	2008-10-06 08:14:18 +0000	[diff] [blame]	949	<dt><tt><a name="byval">byval</a></tt></dt>
Chris Lattner	352ab9b	2008-01-15 04:34:22 +0000	[diff] [blame]	950	<dd>This indicates that the pointer parameter should really be passed by
				951	value to the function. The attribute implies that a hidden copy of the
				952	pointee is made between the caller and the callee, so the callee is unable
Chris Lattner	1ca5c64	2008-08-05 18:21:08 +0000	[diff] [blame]	953	to modify the value in the callee. This attribute is only valid on LLVM
Chris Lattner	352ab9b	2008-01-15 04:34:22 +0000	[diff] [blame]	954	pointer arguments. It is generally used to pass structs and arrays by
Duncan Sands	2a1d8ba	2008-10-06 08:14:18 +0000	[diff] [blame]	955	value, but is also valid on pointers to scalars. The copy is considered to
				956	belong to the caller not the callee (for example,
				957	<tt><a href="#readonly">readonly</a></tt> functions should not write to
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	958	<tt>byval</tt> parameters). This is not a valid attribute for return
Chris Lattner	08aa906	2009-02-05 05:42:28 +0000	[diff] [blame]	959	values. The byval attribute also supports specifying an alignment with the
				960	align attribute. This has a target-specific effect on the code generator
				961	that usually indicates a desired alignment for the synthesized stack
				962	slot.</dd>
Chris Lattner	5cee13f	2008-01-11 06:20:47 +0000	[diff] [blame]	963
Anton Korobeynikov	e816685	2007-01-28 14:30:45 +0000	[diff] [blame]	964	<dt><tt>sret</tt></dt>
Duncan Sands	fa4b673	2008-02-18 04:19:38 +0000	[diff] [blame]	965	<dd>This indicates that the pointer parameter specifies the address of a
				966	structure that is the return value of the function in the source program.
Chris Lattner	d2597d7	2008-10-04 18:33:34 +0000	[diff] [blame]	967	This pointer must be guaranteed by the caller to be valid: loads and stores
				968	to the structure may be assumed by the callee to not to trap. This may only
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	969	be applied to the first parameter. This is not a valid attribute for
				970	return values. </dd>
Chris Lattner	5cee13f	2008-01-11 06:20:47 +0000	[diff] [blame]	971
Zhou Sheng	2444a9a	2007-06-05 05:28:26 +0000	[diff] [blame]	972	<dt><tt>noalias</tt></dt>
Nick Lewycky	f5ffcbc	2008-11-24 03:41:24 +0000	[diff] [blame]	973	<dd>This indicates that the pointer does not alias any global or any other
				974	parameter. The caller is responsible for ensuring that this is the
Nick Lewycky	d59572c	2008-11-24 05:00:44 +0000	[diff] [blame]	975	case. On a function return value, <tt>noalias</tt> additionally indicates
				976	that the pointer does not alias any other pointers visible to the
Nick Lewycky	2abb108	2008-12-19 06:39:12 +0000	[diff] [blame]	977	caller. For further details, please see the discussion of the NoAlias
				978	response in
				979	<a href="http://llvm.org/docs/AliasAnalysis.html#MustMayNo">alias
				980	analysis</a>.</dd>
				981
				982	<dt><tt>nocapture</tt></dt>
				983	<dd>This indicates that the callee does not make any copies of the pointer
				984	that outlive the callee itself. This is not a valid attribute for return
				985	values.</dd>
Chris Lattner	5cee13f	2008-01-11 06:20:47 +0000	[diff] [blame]	986
Duncan Sands	27e9159	2007-07-27 19:57:41 +0000	[diff] [blame]	987	<dt><tt>nest</tt></dt>
Duncan Sands	825bde4	2008-07-08 09:27:25 +0000	[diff] [blame]	988	<dd>This indicates that the pointer parameter can be excised using the
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	989	<a href="#int_trampoline">trampoline intrinsics</a>. This is not a valid
				990	attribute for return values.</dd>
Anton Korobeynikov	a0554d9	2007-01-12 19:20:47 +0000	[diff] [blame]	991	</dl>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	992
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	993	</div>
				994
				995	<!-- ======================================================================= -->
Chris Lattner	91c15c4	2006-01-23 23:23:47 +0000	[diff] [blame]	996	<div class="doc_subsection">
Gordon Henriksen	71183b6	2007-12-10 03:18:06 +0000	[diff] [blame]	997	<a name="gc">Garbage Collector Names</a>
				998	</div>
				999
				1000	<div class="doc_text">
				1001	<p>Each function may specify a garbage collector name, which is simply a
				1002	string.</p>
				1003
				1004	<div class="doc_code"><pre
				1005	>define void @f() gc "name" { ...</pre></div>
				1006
				1007	<p>The compiler declares the supported values of <i>name</i>. Specifying a
				1008	collector which will cause the compiler to alter its output in order to support
				1009	the named garbage collection algorithm.</p>
				1010	</div>
				1011
				1012	<!-- ======================================================================= -->
				1013	<div class="doc_subsection">
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1014	<a name="fnattrs">Function Attributes</a>
Devang Patel	caacdba	2008-09-04 23:05:13 +0000	[diff] [blame]	1015	</div>
				1016
				1017	<div class="doc_text">
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1018
				1019	<p>Function attributes are set to communicate additional information about
				1020	a function. Function attributes are considered to be part of the function,
				1021	not of the function type, so functions with different parameter attributes
				1022	can have the same function type.</p>
				1023
				1024	<p>Function attributes are simple keywords that follow the type specified. If
				1025	multiple attributes are needed, they are space separated. For
				1026	example:</p>
Devang Patel	caacdba	2008-09-04 23:05:13 +0000	[diff] [blame]	1027
				1028	<div class="doc_code">
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1029	<pre>
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1030	define void @f() noinline { ... }
				1031	define void @f() alwaysinline { ... }
				1032	define void @f() alwaysinline optsize { ... }
				1033	define void @f() optsize
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1034	</pre>
Devang Patel	caacdba	2008-09-04 23:05:13 +0000	[diff] [blame]	1035	</div>
				1036
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1037	<dl>
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1038	<dt><tt>alwaysinline</tt></dt>
Chris Lattner	fbf60a4	2008-10-04 18:23:17 +0000	[diff] [blame]	1039	<dd>This attribute indicates that the inliner should attempt to inline this
				1040	function into callers whenever possible, ignoring any active inlining size
				1041	threshold for this caller.</dd>
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1042
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1043	<dt><tt>noinline</tt></dt>
Chris Lattner	fbf60a4	2008-10-04 18:23:17 +0000	[diff] [blame]	1044	<dd>This attribute indicates that the inliner should never inline this function
Chris Lattner	0625c28	2008-10-05 17:14:59 +0000	[diff] [blame]	1045	in any situation. This attribute may not be used together with the
Chris Lattner	fbf60a4	2008-10-04 18:23:17 +0000	[diff] [blame]	1046	<tt>alwaysinline</tt> attribute.</dd>
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1047
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1048	<dt><tt>optsize</tt></dt>
Devang Patel	e974390	2008-09-29 18:34:44 +0000	[diff] [blame]	1049	<dd>This attribute suggests that optimization passes and code generator passes
Chris Lattner	fbf60a4	2008-10-04 18:23:17 +0000	[diff] [blame]	1050	make choices that keep the code size of this function low, and otherwise do
				1051	optimizations specifically to reduce code size.</dd>
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1052
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1053	<dt><tt>noreturn</tt></dt>
Chris Lattner	fbf60a4	2008-10-04 18:23:17 +0000	[diff] [blame]	1054	<dd>This function attribute indicates that the function never returns normally.
				1055	This produces undefined behavior at runtime if the function ever does
				1056	dynamically return.</dd>
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1057
				1058	<dt><tt>nounwind</tt></dt>
Chris Lattner	fbf60a4	2008-10-04 18:23:17 +0000	[diff] [blame]	1059	<dd>This function attribute indicates that the function never returns with an
				1060	unwind or exceptional control flow. If the function does unwind, its runtime
				1061	behavior is undefined.</dd>
				1062
				1063	<dt><tt>readnone</tt></dt>
Duncan Sands	1efabaa	2009-05-06 06:49:50 +0000	[diff] [blame]	1064	<dd>This attribute indicates that the function computes its result (or decides to
				1065	unwind an exception) based strictly on its arguments, without dereferencing any
Duncan Sands	2a1d8ba	2008-10-06 08:14:18 +0000	[diff] [blame]	1066	pointer arguments or otherwise accessing any mutable state (e.g. memory, control
				1067	registers, etc) visible to caller functions. It does not write through any
				1068	pointer arguments (including <tt><a href="#byval">byval</a></tt> arguments) and
Duncan Sands	1efabaa	2009-05-06 06:49:50 +0000	[diff] [blame]	1069	never changes any state visible to callers. This means that it cannot unwind
				1070	exceptions by calling the <tt>C++</tt> exception throwing methods, but could
				1071	use the <tt>unwind</tt> instruction.</dd>
Devang Patel	9eb525d	2008-09-26 23:51:19 +0000	[diff] [blame]	1072
Duncan Sands	2a1d8ba	2008-10-06 08:14:18 +0000	[diff] [blame]	1073	<dt><tt><a name="readonly">readonly</a></tt></dt>
				1074	<dd>This attribute indicates that the function does not write through any
				1075	pointer arguments (including <tt><a href="#byval">byval</a></tt> arguments)
				1076	or otherwise modify any state (e.g. memory, control registers, etc) visible to
				1077	caller functions. It may dereference pointer arguments and read state that may
Duncan Sands	1efabaa	2009-05-06 06:49:50 +0000	[diff] [blame]	1078	be set in the caller. A readonly function always returns the same value (or
				1079	unwinds an exception identically) when called with the same set of arguments
				1080	and global state. It cannot unwind an exception by calling the <tt>C++</tt>
				1081	exception throwing methods, but may use the <tt>unwind</tt> instruction.</dd>
Bill Wendling	a813017	2008-11-13 01:02:51 +0000	[diff] [blame]	1082
				1083	<dt><tt><a name="ssp">ssp</a></tt></dt>
Bill Wendling	6e41add	2008-11-26 19:19:05 +0000	[diff] [blame]	1084	<dd>This attribute indicates that the function should emit a stack smashing
Bill Wendling	a813017	2008-11-13 01:02:51 +0000	[diff] [blame]	1085	protector. It is in the form of a "canary"—a random value placed on the
				1086	stack before the local variables that's checked upon return from the function to
				1087	see if it has been overwritten. A heuristic is used to determine if a function
Bill Wendling	6e41add	2008-11-26 19:19:05 +0000	[diff] [blame]	1088	needs stack protectors or not.
Bill Wendling	a813017	2008-11-13 01:02:51 +0000	[diff] [blame]	1089
Bill Wendling	0f5541e	2008-11-26 19:07:40 +0000	[diff] [blame]	1090	<p>If a function that has an <tt>ssp</tt> attribute is inlined into a function
				1091	that doesn't have an <tt>ssp</tt> attribute, then the resulting function will
				1092	have an <tt>ssp</tt> attribute.</p></dd>
				1093
				1094	<dt><tt>sspreq</tt></dt>
Bill Wendling	6e41add	2008-11-26 19:19:05 +0000	[diff] [blame]	1095	<dd>This attribute indicates that the function should <em>always</em> emit a
Bill Wendling	a813017	2008-11-13 01:02:51 +0000	[diff] [blame]	1096	stack smashing protector. This overrides the <tt><a href="#ssp">ssp</a></tt>
Bill Wendling	6e41add	2008-11-26 19:19:05 +0000	[diff] [blame]	1097	function attribute.
Bill Wendling	0f5541e	2008-11-26 19:07:40 +0000	[diff] [blame]	1098
				1099	<p>If a function that has an <tt>sspreq</tt> attribute is inlined into a
				1100	function that doesn't have an <tt>sspreq</tt> attribute or which has
				1101	an <tt>ssp</tt> attribute, then the resulting function will have
				1102	an <tt>sspreq</tt> attribute.</p></dd>
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1103	</dl>
				1104
Devang Patel	caacdba	2008-09-04 23:05:13 +0000	[diff] [blame]	1105	</div>
				1106
				1107	<!-- ======================================================================= -->
				1108	<div class="doc_subsection">
Chris Lattner	9356489	2006-04-08 04:40:53 +0000	[diff] [blame]	1109	<a name="moduleasm">Module-Level Inline Assembly</a>
Chris Lattner	91c15c4	2006-01-23 23:23:47 +0000	[diff] [blame]	1110	</div>
				1111
				1112	<div class="doc_text">
				1113	<p>
				1114	Modules may contain "module-level inline asm" blocks, which corresponds to the
				1115	GCC "file scope inline asm" blocks. These blocks are internally concatenated by
				1116	LLVM and treated as a single unit, but may be separated in the .ll file if
				1117	desired. The syntax is very simple:
				1118	</p>
				1119
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	1120	<div class="doc_code">
				1121	<pre>
				1122	module asm "inline asm code goes here"
				1123	module asm "more can go here"
				1124	</pre>
				1125	</div>
Chris Lattner	91c15c4	2006-01-23 23:23:47 +0000	[diff] [blame]	1126
				1127	<p>The strings can contain any character by escaping non-printable characters.
				1128	The escape sequence used is simply "\xx" where "xx" is the two digit hex code
				1129	for the number.
				1130	</p>
				1131
				1132	<p>
				1133	The inline asm code is simply printed to the machine code .s file when
				1134	assembly code is generated.
				1135	</p>
				1136	</div>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	1137
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	1138	<!-- ======================================================================= -->
				1139	<div class="doc_subsection">
				1140	<a name="datalayout">Data Layout</a>
				1141	</div>
				1142
				1143	<div class="doc_text">
				1144	<p>A module may specify a target specific data layout string that specifies how
Reid Spencer	7972c47	2007-04-11 23:49:50 +0000	[diff] [blame]	1145	data is to be laid out in memory. The syntax for the data layout is simply:</p>
				1146	<pre> target datalayout = "<i>layout specification</i>"</pre>
				1147	<p>The <i>layout specification</i> consists of a list of specifications
				1148	separated by the minus sign character ('-'). Each specification starts with a
				1149	letter and may include other information after the letter to define some
				1150	aspect of the data layout. The specifications accepted are as follows: </p>
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	1151	<dl>
				1152	<dt><tt>E</tt></dt>
				1153	<dd>Specifies that the target lays out data in big-endian form. That is, the
				1154	bits with the most significance have the lowest address location.</dd>
				1155	<dt><tt>e</tt></dt>
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	1156	<dd>Specifies that the target lays out data in little-endian form. That is,
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	1157	the bits with the least significance have the lowest address location.</dd>
				1158	<dt><tt>p:<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
				1159	<dd>This specifies the <i>size</i> of a pointer and its <i>abi</i> and
				1160	<i>preferred</i> alignments. All sizes are in bits. Specifying the <i>pref</i>
				1161	alignment is optional. If omitted, the preceding <tt>:</tt> should be omitted
				1162	too.</dd>
				1163	<dt><tt>i<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
				1164	<dd>This specifies the alignment for an integer type of a given bit
				1165	<i>size</i>. The value of <i>size</i> must be in the range [1,2^23).</dd>
				1166	<dt><tt>v<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
				1167	<dd>This specifies the alignment for a vector type of a given bit
				1168	<i>size</i>.</dd>
				1169	<dt><tt>f<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
				1170	<dd>This specifies the alignment for a floating point type of a given bit
				1171	<i>size</i>. The value of <i>size</i> must be either 32 (float) or 64
				1172	(double).</dd>
				1173	<dt><tt>a<i>size</i>:<i>abi</i>:<i>pref</i></tt></dt>
				1174	<dd>This specifies the alignment for an aggregate type of a given bit
				1175	<i>size</i>.</dd>
				1176	</dl>
				1177	<p>When constructing the data layout for a given target, LLVM starts with a
				1178	default set of specifications which are then (possibly) overriden by the
				1179	specifications in the <tt>datalayout</tt> keyword. The default specifications
				1180	are given in this list:</p>
				1181	<ul>
				1182	<li><tt>E</tt> - big endian</li>
				1183	<li><tt>p:32:64:64</tt> - 32-bit pointers with 64-bit alignment</li>
				1184	<li><tt>i1:8:8</tt> - i1 is 8-bit (byte) aligned</li>
				1185	<li><tt>i8:8:8</tt> - i8 is 8-bit (byte) aligned</li>
				1186	<li><tt>i16:16:16</tt> - i16 is 16-bit aligned</li>
				1187	<li><tt>i32:32:32</tt> - i32 is 32-bit aligned</li>
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	1188	<li><tt>i64:32:64</tt> - i64 has ABI alignment of 32-bits but preferred
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	1189	alignment of 64-bits</li>
				1190	<li><tt>f32:32:32</tt> - float is 32-bit aligned</li>
				1191	<li><tt>f64:64:64</tt> - double is 64-bit aligned</li>
				1192	<li><tt>v64:64:64</tt> - 64-bit vector is 64-bit aligned</li>
				1193	<li><tt>v128:128:128</tt> - 128-bit vector is 128-bit aligned</li>
				1194	<li><tt>a0:0:1</tt> - aggregates are 8-bit aligned</li>
				1195	</ul>
Chris Lattner	1ca5c64	2008-08-05 18:21:08 +0000	[diff] [blame]	1196	<p>When LLVM is determining the alignment for a given type, it uses the
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	1197	following rules:</p>
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	1198	<ol>
				1199	<li>If the type sought is an exact match for one of the specifications, that
				1200	specification is used.</li>
				1201	<li>If no match is found, and the type sought is an integer type, then the
				1202	smallest integer type that is larger than the bitwidth of the sought type is
				1203	used. If none of the specifications are larger than the bitwidth then the the
				1204	largest integer type is used. For example, given the default specifications
				1205	above, the i7 type will use the alignment of i8 (next largest) while both
				1206	i65 and i256 will use the alignment of i64 (largest specified).</li>
				1207	<li>If no match is found, and the type sought is a vector type, then the
				1208	largest vector type that is smaller than the sought vector type will be used
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	1209	as a fall back. This happens because <128 x double> can be implemented
				1210	in terms of 64 <2 x double>, for example.</li>
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	1211	</ol>
				1212	</div>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	1213
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1214	<!-- *********************************************************************** -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1215	<div class="doc_section"> <a name="typesystem">Type System</a> </div>
				1216	<!-- *********************************************************************** -->
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	1217
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1218	<div class="doc_text">
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	1219
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1220	<p>The LLVM type system is one of the most important features of the
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1221	intermediate representation. Being typed enables a number of
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	1222	optimizations to be performed on the intermediate representation directly,
				1223	without having to do
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1224	extra analyses on the side before the transformation. A strong type
				1225	system makes it easier to read the generated code and enables novel
				1226	analyses and transformations that are not feasible to perform on normal
				1227	three address code representations.</p>
Chris Lattner	6af02f3	2004-12-09 16:11:40 +0000	[diff] [blame]	1228
				1229	</div>
				1230
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1231	<!-- ======================================================================= -->
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1232	<div class="doc_subsection"> <a name="t_classifications">Type
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1233	Classifications</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1234	<div class="doc_text">
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1235	<p>The types fall into a few useful
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1236	classifications:</p>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	1237
				1238	<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1239	<tbody>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1240	<tr><th>Classification</th><th>Types</th></tr>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1241	<tr>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1242	<td><a href="#t_integer">integer</a></td>
Reid Spencer	138249b	2007-05-16 18:44:01 +0000	[diff] [blame]	1243	<td><tt>i1, i2, i3, ... i8, ... i16, ... i32, ... i64, ... </tt></td>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1244	</tr>
				1245	<tr>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1246	<td><a href="#t_floating">floating point</a></td>
				1247	<td><tt>float, double, x86_fp80, fp128, ppc_fp128</tt></td>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1248	</tr>
				1249	<tr>
				1250	<td><a name="t_firstclass">first class</a></td>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1251	<td><a href="#t_integer">integer</a>,
				1252	<a href="#t_floating">floating point</a>,
				1253	<a href="#t_pointer">pointer</a>,
Dan Gohman	08783a88	2008-06-18 18:42:13 +0000	[diff] [blame]	1254	<a href="#t_vector">vector</a>,
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	1255	<a href="#t_struct">structure</a>,
				1256	<a href="#t_array">array</a>,
Dan Gohman	da52d21	2008-05-23 22:50:26 +0000	[diff] [blame]	1257	<a href="#t_label">label</a>.
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	1258	</td>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1259	</tr>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1260	<tr>
				1261	<td><a href="#t_primitive">primitive</a></td>
				1262	<td><a href="#t_label">label</a>,
				1263	<a href="#t_void">void</a>,
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1264	<a href="#t_floating">floating point</a>.</td>
				1265	</tr>
				1266	<tr>
				1267	<td><a href="#t_derived">derived</a></td>
				1268	<td><a href="#t_integer">integer</a>,
				1269	<a href="#t_array">array</a>,
				1270	<a href="#t_function">function</a>,
				1271	<a href="#t_pointer">pointer</a>,
				1272	<a href="#t_struct">structure</a>,
				1273	<a href="#t_pstruct">packed structure</a>,
				1274	<a href="#t_vector">vector</a>,
				1275	<a href="#t_opaque">opaque</a>.
Dan Gohman	93bf60d	2008-10-14 16:32:04 +0000	[diff] [blame]	1276	</td>
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1277	</tr>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1278	</tbody>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1279	</table>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	1280
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1281	<p>The <a href="#t_firstclass">first class</a> types are perhaps the
				1282	most important. Values of these types are the only ones which can be
				1283	produced by instructions, passed as arguments, or used as operands to
Dan Gohman	34d1c0d	2008-05-23 21:53:15 +0000	[diff] [blame]	1284	instructions.</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1285	</div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1286
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1287	<!-- ======================================================================= -->
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1288	<div class="doc_subsection"> <a name="t_primitive">Primitive Types</a> </div>
Chris Lattner	43542b3	2008-01-04 04:34:14 +0000	[diff] [blame]	1289
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1290	<div class="doc_text">
				1291	<p>The primitive types are the fundamental building blocks of the LLVM
				1292	system.</p>
				1293
Chris Lattner	43542b3	2008-01-04 04:34:14 +0000	[diff] [blame]	1294	</div>
				1295
Chris Lattner	7824d18	2008-01-04 04:32:38 +0000	[diff] [blame]	1296	<!-- _______________________________________________________________________ -->
				1297	<div class="doc_subsubsection"> <a name="t_floating">Floating Point Types</a> </div>
				1298
				1299	<div class="doc_text">
				1300	<table>
				1301	<tbody>
				1302	<tr><th>Type</th><th>Description</th></tr>
				1303	<tr><td><tt>float</tt></td><td>32-bit floating point value</td></tr>
				1304	<tr><td><tt>double</tt></td><td>64-bit floating point value</td></tr>
				1305	<tr><td><tt>fp128</tt></td><td>128-bit floating point value (112-bit mantissa)</td></tr>
				1306	<tr><td><tt>x86_fp80</tt></td><td>80-bit floating point value (X87)</td></tr>
				1307	<tr><td><tt>ppc_fp128</tt></td><td>128-bit floating point value (two 64-bits)</td></tr>
				1308	</tbody>
				1309	</table>
				1310	</div>
				1311
				1312	<!-- _______________________________________________________________________ -->
				1313	<div class="doc_subsubsection"> <a name="t_void">Void Type</a> </div>
				1314
				1315	<div class="doc_text">
				1316	<h5>Overview:</h5>
				1317	<p>The void type does not represent any value and has no size.</p>
				1318
				1319	<h5>Syntax:</h5>
				1320
				1321	<pre>
				1322	void
				1323	</pre>
				1324	</div>
				1325
				1326	<!-- _______________________________________________________________________ -->
				1327	<div class="doc_subsubsection"> <a name="t_label">Label Type</a> </div>
				1328
				1329	<div class="doc_text">
				1330	<h5>Overview:</h5>
				1331	<p>The label type represents code labels.</p>
				1332
				1333	<h5>Syntax:</h5>
				1334
				1335	<pre>
				1336	label
				1337	</pre>
				1338	</div>
				1339
				1340
				1341	<!-- ======================================================================= -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1342	<div class="doc_subsection"> <a name="t_derived">Derived Types</a> </div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1343
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1344	<div class="doc_text">
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1345
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1346	<p>The real power in LLVM comes from the derived types in the system.
				1347	This is what allows a programmer to represent arrays, functions,
				1348	pointers, and other useful types. Note that these derived types may be
				1349	recursive: For example, it is possible to have a two dimensional array.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1350
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1351	</div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1352
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1353	<!-- _______________________________________________________________________ -->
Reid Spencer	138249b	2007-05-16 18:44:01 +0000	[diff] [blame]	1354	<div class="doc_subsubsection"> <a name="t_integer">Integer Type</a> </div>
				1355
				1356	<div class="doc_text">
				1357
				1358	<h5>Overview:</h5>
				1359	<p>The integer type is a very simple derived type that simply specifies an
				1360	arbitrary bit width for the integer type desired. Any bit width from 1 bit to
				1361	2^23-1 (about 8 million) can be specified.</p>
				1362
				1363	<h5>Syntax:</h5>
				1364
				1365	<pre>
				1366	iN
				1367	</pre>
				1368
				1369	<p>The number of bits the integer will occupy is specified by the <tt>N</tt>
				1370	value.</p>
				1371
				1372	<h5>Examples:</h5>
				1373	<table class="layout">
Nick Lewycky	aab930a	2009-05-24 02:46:06 +0000	[diff] [blame^]	1374	<tr class="layout">
				1375	<td class="left"><tt>i1</tt></td>
				1376	<td class="left">a single-bit integer.</td>
Reid Spencer	138249b	2007-05-16 18:44:01 +0000	[diff] [blame]	1377	</tr>
Nick Lewycky	aab930a	2009-05-24 02:46:06 +0000	[diff] [blame^]	1378	<tr class="layout">
				1379	<td class="left"><tt>i32</tt></td>
				1380	<td class="left">a 32-bit integer.</td>
				1381	</tr>
				1382	<tr class="layout">
				1383	<td class="left"><tt>i1942652</tt></td>
				1384	<td class="left">a really big integer of over 1 million bits.</td>
				1385	</tr>
Reid Spencer	138249b	2007-05-16 18:44:01 +0000	[diff] [blame]	1386	</table>
Dan Gohman	142ccc0	2009-01-24 15:58:40 +0000	[diff] [blame]	1387
				1388	<p>Note that the code generator does not yet support large integer types
				1389	to be used as function return types. The specific limit on how large a
				1390	return type the code generator can currently handle is target-dependent;
				1391	currently it's often 64 bits for 32-bit targets and 128 bits for 64-bit
				1392	targets.</p>
				1393
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	1394	</div>
Reid Spencer	138249b	2007-05-16 18:44:01 +0000	[diff] [blame]	1395
				1396	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1397	<div class="doc_subsubsection"> <a name="t_array">Array Type</a> </div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1398
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1399	<div class="doc_text">
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1400
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1401	<h5>Overview:</h5>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1402
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1403	<p>The array type is a very simple derived type that arranges elements
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1404	sequentially in memory. The array type requires a size (number of
				1405	elements) and an underlying data type.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1406
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	1407	<h5>Syntax:</h5>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1408
				1409	<pre>
				1410	[<# elements> x <elementtype>]
				1411	</pre>
				1412
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	1413	<p>The number of elements is a constant integer value; elementtype may
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1414	be any type with a size.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1415
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	1416	<h5>Examples:</h5>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1417	<table class="layout">
				1418	<tr class="layout">
Chris Lattner	747359f	2007-12-19 05:04:11 +0000	[diff] [blame]	1419	<td class="left"><tt>[40 x i32]</tt></td>
				1420	<td class="left">Array of 40 32-bit integer values.</td>
				1421	</tr>
				1422	<tr class="layout">
				1423	<td class="left"><tt>[41 x i32]</tt></td>
				1424	<td class="left">Array of 41 32-bit integer values.</td>
				1425	</tr>
				1426	<tr class="layout">
				1427	<td class="left"><tt>[4 x i8]</tt></td>
				1428	<td class="left">Array of 4 8-bit integer values.</td>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1429	</tr>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1430	</table>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1431	<p>Here are some examples of multidimensional arrays:</p>
				1432	<table class="layout">
				1433	<tr class="layout">
Chris Lattner	747359f	2007-12-19 05:04:11 +0000	[diff] [blame]	1434	<td class="left"><tt>[3 x [4 x i32]]</tt></td>
				1435	<td class="left">3x4 array of 32-bit integer values.</td>
				1436	</tr>
				1437	<tr class="layout">
				1438	<td class="left"><tt>[12 x [10 x float]]</tt></td>
				1439	<td class="left">12x10 array of single precision floating point values.</td>
				1440	</tr>
				1441	<tr class="layout">
				1442	<td class="left"><tt>[2 x [3 x [4 x i16]]]</tt></td>
				1443	<td class="left">2x3x4 array of 16-bit integer values.</td>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1444	</tr>
				1445	</table>
Chris Lattner	c0ad71e	2005-06-24 17:22:57 +0000	[diff] [blame]	1446
John Criswell	4c0cf7f	2005-10-24 16:17:18 +0000	[diff] [blame]	1447	<p>Note that 'variable sized arrays' can be implemented in LLVM with a zero
				1448	length array. Normally, accesses past the end of an array are undefined in
Chris Lattner	c0ad71e	2005-06-24 17:22:57 +0000	[diff] [blame]	1449	LLVM (e.g. it is illegal to access the 5th element of a 3 element array).
				1450	As a special case, however, zero length arrays are recognized to be variable
				1451	length. This allows implementation of 'pascal style arrays' with the LLVM
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	1452	type "{ i32, [0 x float]}", for example.</p>
Chris Lattner	c0ad71e	2005-06-24 17:22:57 +0000	[diff] [blame]	1453
Dan Gohman	142ccc0	2009-01-24 15:58:40 +0000	[diff] [blame]	1454	<p>Note that the code generator does not yet support large aggregate types
				1455	to be used as function return types. The specific limit on how large an
				1456	aggregate return type the code generator can currently handle is
				1457	target-dependent, and also dependent on the aggregate element types.</p>
				1458
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1459	</div>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1460
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1461	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1462	<div class="doc_subsubsection"> <a name="t_function">Function Type</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1463	<div class="doc_text">
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	1464
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1465	<h5>Overview:</h5>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	1466
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1467	<p>The function type can be thought of as a function signature. It
Devang Patel	e3dfc1c	2008-03-24 05:35:41 +0000	[diff] [blame]	1468	consists of a return type and a list of formal parameter types. The
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	1469	return type of a function type is a scalar type, a void type, or a struct type.
Devang Patel	9c1f8b1	2008-03-24 20:52:42 +0000	[diff] [blame]	1470	If the return type is a struct type then all struct elements must be of first
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	1471	class types, and the struct must have at least one element.</p>
Devang Patel	d6cff51	2008-03-10 20:49:15 +0000	[diff] [blame]	1472
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1473	<h5>Syntax:</h5>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	1474
				1475	<pre>
				1476	<returntype list> (<parameter list>)
				1477	</pre>
				1478
John Criswell	4c0cf7f	2005-10-24 16:17:18 +0000	[diff] [blame]	1479	<p>...where '<tt><parameter list></tt>' is a comma-separated list of type
Misha Brukman	20f9a62	2004-08-12 20:16:08 +0000	[diff] [blame]	1480	specifiers. Optionally, the parameter list may include a type <tt>...</tt>,
Chris Lattner	5ed6061	2003-09-03 00:41:47 +0000	[diff] [blame]	1481	which indicates that the function takes a variable number of arguments.
				1482	Variable argument functions can access their arguments with the <a
Devang Patel	d6cff51	2008-03-10 20:49:15 +0000	[diff] [blame]	1483	href="#int_varargs">variable argument handling intrinsic</a> functions.
				1484	'<tt><returntype list></tt>' is a comma-separated list of
				1485	<a href="#t_firstclass">first class</a> type specifiers.</p>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	1486
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1487	<h5>Examples:</h5>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1488	<table class="layout">
				1489	<tr class="layout">
Reid Spencer	58c0871	2006-12-31 07:18:34 +0000	[diff] [blame]	1490	<td class="left"><tt>i32 (i32)</tt></td>
				1491	<td class="left">function taking an <tt>i32</tt>, returning an <tt>i32</tt>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1492	</td>
Reid Spencer	58c0871	2006-12-31 07:18:34 +0000	[diff] [blame]	1493	</tr><tr class="layout">
Reid Spencer	314e1cb	2007-07-19 23:13:04 +0000	[diff] [blame]	1494	<td class="left"><tt>float (i16 signext, i32 )
Reid Spencer	655dcc6	2006-12-31 07:20:23 +0000	[diff] [blame]	1495	</tt></td>
Reid Spencer	58c0871	2006-12-31 07:18:34 +0000	[diff] [blame]	1496	<td class="left"><a href="#t_pointer">Pointer</a> to a function that takes
				1497	an <tt>i16</tt> that should be sign extended and a
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	1498	<a href="#t_pointer">pointer</a> to <tt>i32</tt>, returning
Reid Spencer	58c0871	2006-12-31 07:18:34 +0000	[diff] [blame]	1499	<tt>float</tt>.
				1500	</td>
				1501	</tr><tr class="layout">
				1502	<td class="left"><tt>i32 (i8*, ...)</tt></td>
				1503	<td class="left">A vararg function that takes at least one
Reid Spencer	3e628eb9	2007-01-04 16:43:23 +0000	[diff] [blame]	1504	<a href="#t_pointer">pointer</a> to <tt>i8 </tt> (char in C),
Reid Spencer	58c0871	2006-12-31 07:18:34 +0000	[diff] [blame]	1505	which returns an integer. This is the signature for <tt>printf</tt> in
				1506	LLVM.
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1507	</td>
Devang Patel	e3dfc1c	2008-03-24 05:35:41 +0000	[diff] [blame]	1508	</tr><tr class="layout">
				1509	<td class="left"><tt>{i32, i32} (i32)</tt></td>
Misha Brukman	c9813bd	2008-11-27 06:41:20 +0000	[diff] [blame]	1510	<td class="left">A function taking an <tt>i32</tt>, returning two
				1511	<tt>i32</tt> values as an aggregate of type <tt>{ i32, i32 }</tt>
Devang Patel	e3dfc1c	2008-03-24 05:35:41 +0000	[diff] [blame]	1512	</td>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1513	</tr>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1514	</table>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	1515
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1516	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1517	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1518	<div class="doc_subsubsection"> <a name="t_struct">Structure Type</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1519	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1520	<h5>Overview:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1521	<p>The structure type is used to represent a collection of data members
				1522	together in memory. The packing of the field types is defined to match
				1523	the ABI of the underlying processor. The elements of a structure may
				1524	be any type that has a size.</p>
				1525	<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
				1526	and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
				1527	field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
				1528	instruction.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1529	<h5>Syntax:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1530	<pre> { <type list> }<br></pre>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1531	<h5>Examples:</h5>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1532	<table class="layout">
				1533	<tr class="layout">
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	1534	<td class="left"><tt>{ i32, i32, i32 }</tt></td>
				1535	<td class="left">A triple of three <tt>i32</tt> values</td>
				1536	</tr><tr class="layout">
				1537	<td class="left"><tt>{ float, i32 (i32) * }</tt></td>
				1538	<td class="left">A pair, where the first element is a <tt>float</tt> and the
				1539	second element is a <a href="#t_pointer">pointer</a> to a
				1540	<a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
				1541	an <tt>i32</tt>.</td>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1542	</tr>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1543	</table>
Dan Gohman	142ccc0	2009-01-24 15:58:40 +0000	[diff] [blame]	1544
				1545	<p>Note that the code generator does not yet support large aggregate types
				1546	to be used as function return types. The specific limit on how large an
				1547	aggregate return type the code generator can currently handle is
				1548	target-dependent, and also dependent on the aggregate element types.</p>
				1549
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1550	</div>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1551
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	1552	<!-- _______________________________________________________________________ -->
Andrew Lenharth	8df88e2	2006-12-08 17:13:00 +0000	[diff] [blame]	1553	<div class="doc_subsubsection"> <a name="t_pstruct">Packed Structure Type</a>
				1554	</div>
				1555	<div class="doc_text">
				1556	<h5>Overview:</h5>
				1557	<p>The packed structure type is used to represent a collection of data members
				1558	together in memory. There is no padding between fields. Further, the alignment
				1559	of a packed structure is 1 byte. The elements of a packed structure may
				1560	be any type that has a size.</p>
				1561	<p>Structures are accessed using '<tt><a href="#i_load">load</a></tt>
				1562	and '<tt><a href="#i_store">store</a></tt>' by getting a pointer to a
				1563	field with the '<tt><a href="#i_getelementptr">getelementptr</a></tt>'
				1564	instruction.</p>
				1565	<h5>Syntax:</h5>
				1566	<pre> < { <type list> } > <br></pre>
				1567	<h5>Examples:</h5>
				1568	<table class="layout">
				1569	<tr class="layout">
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	1570	<td class="left"><tt>< { i32, i32, i32 } ></tt></td>
				1571	<td class="left">A triple of three <tt>i32</tt> values</td>
				1572	</tr><tr class="layout">
Bill Wendling	b175fa4	2008-09-07 10:26:33 +0000	[diff] [blame]	1573	<td class="left">
				1574	<tt>< { float, i32 (i32)* } ></tt></td>
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	1575	<td class="left">A pair, where the first element is a <tt>float</tt> and the
				1576	second element is a <a href="#t_pointer">pointer</a> to a
				1577	<a href="#t_function">function</a> that takes an <tt>i32</tt>, returning
				1578	an <tt>i32</tt>.</td>
Andrew Lenharth	8df88e2	2006-12-08 17:13:00 +0000	[diff] [blame]	1579	</tr>
				1580	</table>
				1581	</div>
				1582
				1583	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1584	<div class="doc_subsubsection"> <a name="t_pointer">Pointer Type</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1585	<div class="doc_text">
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	1586	<h5>Overview:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1587	<p>As in many languages, the pointer type represents a pointer or
Christopher Lamb	308121c	2007-12-11 09:31:00 +0000	[diff] [blame]	1588	reference to another object, which must live in memory. Pointer types may have
				1589	an optional address space attribute defining the target-specific numbered
				1590	address space where the pointed-to object resides. The default address space is
				1591	zero.</p>
Chris Lattner	4a67c91	2009-02-08 19:53:29 +0000	[diff] [blame]	1592
				1593	<p>Note that LLVM does not permit pointers to void (<tt>void*</tt>) nor does
Chris Lattner	d1d4cff	2009-02-08 22:21:28 +0000	[diff] [blame]	1594	it permit pointers to labels (<tt>label</tt>). Use <tt>i8</tt> instead.</p>
Chris Lattner	4a67c91	2009-02-08 19:53:29 +0000	[diff] [blame]	1595
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	1596	<h5>Syntax:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	1597	<pre> <type> *<br></pre>
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	1598	<h5>Examples:</h5>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1599	<table class="layout">
				1600	<tr class="layout">
Dan Gohman	623806e	2009-01-04 23:44:43 +0000	[diff] [blame]	1601	<td class="left"><tt>[4 x i32]*</tt></td>
Chris Lattner	747359f	2007-12-19 05:04:11 +0000	[diff] [blame]	1602	<td class="left">A <a href="#t_pointer">pointer</a> to <a
				1603	href="#t_array">array</a> of four <tt>i32</tt> values.</td>
				1604	</tr>
				1605	<tr class="layout">
				1606	<td class="left"><tt>i32 (i32 ) </tt></td>
				1607	<td class="left"> A <a href="#t_pointer">pointer</a> to a <a
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	1608	href="#t_function">function</a> that takes an <tt>i32*</tt>, returning an
Chris Lattner	747359f	2007-12-19 05:04:11 +0000	[diff] [blame]	1609	<tt>i32</tt>.</td>
				1610	</tr>
				1611	<tr class="layout">
				1612	<td class="left"><tt>i32 addrspace(5)*</tt></td>
				1613	<td class="left">A <a href="#t_pointer">pointer</a> to an <tt>i32</tt> value
				1614	that resides in address space #5.</td>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1615	</tr>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1616	</table>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1617	</div>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1618
Chris Lattner	c8cb695	2004-08-12 19:12:28 +0000	[diff] [blame]	1619	<!-- _______________________________________________________________________ -->
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1620	<div class="doc_subsubsection"> <a name="t_vector">Vector Type</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1621	<div class="doc_text">
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1622
Chris Lattner	c8cb695	2004-08-12 19:12:28 +0000	[diff] [blame]	1623	<h5>Overview:</h5>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1624
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1625	<p>A vector type is a simple derived type that represents a vector
				1626	of elements. Vector types are used when multiple primitive data
Chris Lattner	c8cb695	2004-08-12 19:12:28 +0000	[diff] [blame]	1627	are operated in parallel using a single instruction (SIMD).
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1628	A vector type requires a size (number of
Chris Lattner	330ce69	2005-11-10 01:44:22 +0000	[diff] [blame]	1629	elements) and an underlying primitive data type. Vectors must have a power
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1630	of two length (1, 2, 4, 8, 16 ...). Vector types are
Chris Lattner	c8cb695	2004-08-12 19:12:28 +0000	[diff] [blame]	1631	considered <a href="#t_firstclass">first class</a>.</p>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1632
Chris Lattner	c8cb695	2004-08-12 19:12:28 +0000	[diff] [blame]	1633	<h5>Syntax:</h5>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1634
				1635	<pre>
				1636	< <# elements> x <elementtype> >
				1637	</pre>
				1638
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	1639	<p>The number of elements is a constant integer value; elementtype may
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	1640	be any integer or floating point type.</p>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1641
Chris Lattner	c8cb695	2004-08-12 19:12:28 +0000	[diff] [blame]	1642	<h5>Examples:</h5>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1643
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1644	<table class="layout">
				1645	<tr class="layout">
Chris Lattner	747359f	2007-12-19 05:04:11 +0000	[diff] [blame]	1646	<td class="left"><tt><4 x i32></tt></td>
				1647	<td class="left">Vector of 4 32-bit integer values.</td>
				1648	</tr>
				1649	<tr class="layout">
				1650	<td class="left"><tt><8 x float></tt></td>
				1651	<td class="left">Vector of 8 32-bit floating-point values.</td>
				1652	</tr>
				1653	<tr class="layout">
				1654	<td class="left"><tt><2 x i64></tt></td>
				1655	<td class="left">Vector of 2 64-bit integer values.</td>
Reid Spencer	c3c4c4f	2004-11-01 08:19:36 +0000	[diff] [blame]	1656	</tr>
				1657	</table>
Dan Gohman	142ccc0	2009-01-24 15:58:40 +0000	[diff] [blame]	1658
				1659	<p>Note that the code generator does not yet support large vector types
				1660	to be used as function return types. The specific limit on how large a
				1661	vector return type codegen can currently handle is target-dependent;
				1662	currently it's often a few times longer than a hardware vector register.</p>
				1663
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	1664	</div>
				1665
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1666	<!-- _______________________________________________________________________ -->
				1667	<div class="doc_subsubsection"> <a name="t_opaque">Opaque Type</a> </div>
				1668	<div class="doc_text">
				1669
				1670	<h5>Overview:</h5>
				1671
				1672	<p>Opaque types are used to represent unknown types in the system. This
Gordon Henriksen	a699c4d	2007-10-14 00:34:53 +0000	[diff] [blame]	1673	corresponds (for example) to the C notion of a forward declared structure type.
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1674	In LLVM, opaque types can eventually be resolved to any type (not just a
				1675	structure type).</p>
				1676
				1677	<h5>Syntax:</h5>
				1678
				1679	<pre>
				1680	opaque
				1681	</pre>
				1682
				1683	<h5>Examples:</h5>
				1684
				1685	<table class="layout">
				1686	<tr class="layout">
Chris Lattner	747359f	2007-12-19 05:04:11 +0000	[diff] [blame]	1687	<td class="left"><tt>opaque</tt></td>
				1688	<td class="left">An opaque type.</td>
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1689	</tr>
				1690	</table>
				1691	</div>
				1692
Chris Lattner	cf7a584	2009-02-02 07:32:36 +0000	[diff] [blame]	1693	<!-- ======================================================================= -->
				1694	<div class="doc_subsection">
				1695	<a name="t_uprefs">Type Up-references</a>
				1696	</div>
				1697
				1698	<div class="doc_text">
				1699	<h5>Overview:</h5>
				1700	<p>
				1701	An "up reference" allows you to refer to a lexically enclosing type without
				1702	requiring it to have a name. For instance, a structure declaration may contain a
				1703	pointer to any of the types it is lexically a member of. Example of up
				1704	references (with their equivalent as named type declarations) include:</p>
				1705
				1706	<pre>
Chris Lattner	bf1d545	2009-02-09 10:00:56 +0000	[diff] [blame]	1707	{ \2 * } %x = type { %x* }
Chris Lattner	cf7a584	2009-02-02 07:32:36 +0000	[diff] [blame]	1708	{ \2 }* %y = type { %y }*
				1709	\1* %z = type %z*
				1710	</pre>
				1711
				1712	<p>
				1713	An up reference is needed by the asmprinter for printing out cyclic types when
				1714	there is no declared name for a type in the cycle. Because the asmprinter does
				1715	not want to print out an infinite type string, it needs a syntax to handle
				1716	recursive types that have no names (all names are optional in llvm IR).
				1717	</p>
				1718
				1719	<h5>Syntax:</h5>
				1720	<pre>
				1721	\<level>
				1722	</pre>
				1723
				1724	<p>
				1725	The level is the count of the lexical type that is being referred to.
				1726	</p>
				1727
				1728	<h5>Examples:</h5>
				1729
				1730	<table class="layout">
				1731	<tr class="layout">
				1732	<td class="left"><tt>\1*</tt></td>
				1733	<td class="left">Self-referential pointer.</td>
				1734	</tr>
				1735	<tr class="layout">
				1736	<td class="left"><tt>{ { \3*, i8 }, i32 }</tt></td>
				1737	<td class="left">Recursive structure where the upref refers to the out-most
				1738	structure.</td>
				1739	</tr>
				1740	</table>
				1741	</div>
				1742
Chris Lattner	37b6b09	2005-04-25 17:34:15 +0000	[diff] [blame]	1743
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1744	<!-- *********************************************************************** -->
				1745	<div class="doc_section"> <a name="constants">Constants</a> </div>
				1746	<!-- *********************************************************************** -->
				1747
				1748	<div class="doc_text">
				1749
				1750	<p>LLVM has several different basic types of constants. This section describes
				1751	them all and their syntax.</p>
				1752
				1753	</div>
				1754
				1755	<!-- ======================================================================= -->
Reid Spencer	8f08d80	2004-12-09 18:02:53 +0000	[diff] [blame]	1756	<div class="doc_subsection"><a name="simpleconstants">Simple Constants</a></div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1757
				1758	<div class="doc_text">
				1759
				1760	<dl>
				1761	<dt><b>Boolean constants</b></dt>
				1762
				1763	<dd>The two strings '<tt>true</tt>' and '<tt>false</tt>' are both valid
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	1764	constants of the <tt><a href="#t_primitive">i1</a></tt> type.
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1765	</dd>
				1766
				1767	<dt><b>Integer constants</b></dt>
				1768
Reid Spencer	8f08d80	2004-12-09 18:02:53 +0000	[diff] [blame]	1769	<dd>Standard integers (such as '4') are constants of the <a
Reid Spencer	3e628eb9	2007-01-04 16:43:23 +0000	[diff] [blame]	1770	href="#t_integer">integer</a> type. Negative numbers may be used with
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1771	integer types.
				1772	</dd>
				1773
				1774	<dt><b>Floating point constants</b></dt>
				1775
				1776	<dd>Floating point constants use standard decimal notation (e.g. 123.421),
				1777	exponential notation (e.g. 1.23421e+2), or a more precise hexadecimal
Chris Lattner	1429e6f	2008-04-01 18:45:27 +0000	[diff] [blame]	1778	notation (see below). The assembler requires the exact decimal value of
				1779	a floating-point constant. For example, the assembler accepts 1.25 but
				1780	rejects 1.3 because 1.3 is a repeating decimal in binary. Floating point
				1781	constants must have a <a href="#t_floating">floating point</a> type. </dd>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1782
				1783	<dt><b>Null pointer constants</b></dt>
				1784
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	1785	<dd>The identifier '<tt>null</tt>' is recognized as a null pointer constant
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1786	and must be of <a href="#t_pointer">pointer type</a>.</dd>
				1787
				1788	</dl>
				1789
Dale Johannesen	cd4a301	2009-02-11 22:14:51 +0000	[diff] [blame]	1790	<p>The one non-intuitive notation for constants is the hexadecimal form
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1791	of floating point constants. For example, the form '<tt>double
				1792	0x432ff973cafa8000</tt>' is equivalent to (but harder to read than) '<tt>double
				1793	4.5e+15</tt>'. The only time hexadecimal floating point constants are required
Reid Spencer	8f08d80	2004-12-09 18:02:53 +0000	[diff] [blame]	1794	(and the only time that they are generated by the disassembler) is when a
				1795	floating point constant must be emitted but it cannot be represented as a
Dale Johannesen	cd4a301	2009-02-11 22:14:51 +0000	[diff] [blame]	1796	decimal floating point number in a reasonable number of digits. For example,
				1797	NaN's, infinities, and other
Reid Spencer	8f08d80	2004-12-09 18:02:53 +0000	[diff] [blame]	1798	special values are represented in their IEEE hexadecimal format so that
				1799	assembly and disassembly do not cause any bits to change in the constants.</p>
Dale Johannesen	cd4a301	2009-02-11 22:14:51 +0000	[diff] [blame]	1800	<p>When using the hexadecimal form, constants of types float and double are
				1801	represented using the 16-digit form shown above (which matches the IEEE754
				1802	representation for double); float values must, however, be exactly representable
				1803	as IEE754 single precision.
				1804	Hexadecimal format is always used for long
				1805	double, and there are three forms of long double. The 80-bit
				1806	format used by x86 is represented as <tt>0xK</tt>
				1807	followed by 20 hexadecimal digits.
				1808	The 128-bit format used by PowerPC (two adjacent doubles) is represented
				1809	by <tt>0xM</tt> followed by 32 hexadecimal digits. The IEEE 128-bit
				1810	format is represented
				1811	by <tt>0xL</tt> followed by 32 hexadecimal digits; no currently supported
				1812	target uses this format. Long doubles will only work if they match
				1813	the long double format on your target. All hexadecimal formats are big-endian
				1814	(sign bit at the left).</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1815	</div>
				1816
				1817	<!-- ======================================================================= -->
Chris Lattner	361bfcd	2009-02-28 18:32:25 +0000	[diff] [blame]	1818	<div class="doc_subsection">
				1819	<a name="aggregateconstants"> <!-- old anchor -->
				1820	<a name="complexconstants">Complex Constants</a></a>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1821	</div>
				1822
				1823	<div class="doc_text">
Chris Lattner	361bfcd	2009-02-28 18:32:25 +0000	[diff] [blame]	1824	<p>Complex constants are a (potentially recursive) combination of simple
				1825	constants and smaller complex constants.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1826
				1827	<dl>
				1828	<dt><b>Structure constants</b></dt>
				1829
				1830	<dd>Structure constants are represented with notation similar to structure
				1831	type definitions (a comma separated list of elements, surrounded by braces
Chris Lattner	bea1117	2007-12-25 20:34:52 +0000	[diff] [blame]	1832	(<tt>{}</tt>)). For example: "<tt>{ i32 4, float 17.0, i32* @G }</tt>",
				1833	where "<tt>@G</tt>" is declared as "<tt>@G = external global i32</tt>". Structure constants
Chris Lattner	455fc8c	2005-03-07 22:13:59 +0000	[diff] [blame]	1834	must have <a href="#t_struct">structure type</a>, and the number and
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1835	types of elements must match those specified by the type.
				1836	</dd>
				1837
				1838	<dt><b>Array constants</b></dt>
				1839
				1840	<dd>Array constants are represented with notation similar to array type
				1841	definitions (a comma separated list of elements, surrounded by square brackets
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	1842	(<tt>[]</tt>)). For example: "<tt>[ i32 42, i32 11, i32 74 ]</tt>". Array
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1843	constants must have <a href="#t_array">array type</a>, and the number and
				1844	types of elements must match those specified by the type.
				1845	</dd>
				1846
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1847	<dt><b>Vector constants</b></dt>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1848
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1849	<dd>Vector constants are represented with notation similar to vector type
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1850	definitions (a comma separated list of elements, surrounded by
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	1851	less-than/greater-than's (<tt><></tt>)). For example: "<tt>< i32 42,
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	1852	i32 11, i32 74, i32 100 ></tt>". Vector constants must have <a
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	1853	href="#t_vector">vector type</a>, and the number and types of elements must
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1854	match those specified by the type.
				1855	</dd>
				1856
				1857	<dt><b>Zero initialization</b></dt>
				1858
				1859	<dd>The string '<tt>zeroinitializer</tt>' can be used to zero initialize a
				1860	value to zero of <em>any</em> type, including scalar and aggregate types.
				1861	This is often used to avoid having to print large zero initializers (e.g. for
John Criswell	4c0cf7f	2005-10-24 16:17:18 +0000	[diff] [blame]	1862	large arrays) and is always exactly equivalent to using explicit zero
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1863	initializers.
				1864	</dd>
Nick Lewycky	49f8919	2009-04-04 07:22:01 +0000	[diff] [blame]	1865
				1866	<dt><b>Metadata node</b></dt>
				1867
				1868	<dd>A metadata node is a structure-like constant with the type of an empty
				1869	struct. For example: "<tt>{ } !{ i32 0, { } !"test" }</tt>". Unlike other
				1870	constants that are meant to be interpreted as part of the instruction stream,
				1871	metadata is a place to attach additional information such as debug info.
				1872	</dd>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1873	</dl>
				1874
				1875	</div>
				1876
				1877	<!-- ======================================================================= -->
				1878	<div class="doc_subsection">
				1879	<a name="globalconstants">Global Variable and Function Addresses</a>
				1880	</div>
				1881
				1882	<div class="doc_text">
				1883
				1884	<p>The addresses of <a href="#globalvars">global variables</a> and <a
				1885	href="#functionstructure">functions</a> are always implicitly valid (link-time)
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	1886	constants. These constants are explicitly referenced when the <a
				1887	href="#identifiers">identifier for the global</a> is used and always have <a
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1888	href="#t_pointer">pointer</a> type. For example, the following is a legal LLVM
				1889	file:</p>
				1890
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	1891	<div class="doc_code">
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1892	<pre>
Chris Lattner	00538a1	2007-06-06 18:28:13 +0000	[diff] [blame]	1893	@X = global i32 17
				1894	@Y = global i32 42
				1895	@Z = global [2 x i32] [ i32 @X, i32* @Y ]
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1896	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	1897	</div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1898
				1899	</div>
				1900
				1901	<!-- ======================================================================= -->
Reid Spencer	641f5c9	2004-12-09 18:13:12 +0000	[diff] [blame]	1902	<div class="doc_subsection"><a name="undefvalues">Undefined Values</a></div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1903	<div class="doc_text">
Reid Spencer	641f5c9	2004-12-09 18:13:12 +0000	[diff] [blame]	1904	<p>The string '<tt>undef</tt>' is recognized as a type-less constant that has
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	1905	no specific value. Undefined values may be of any type and be used anywhere
Reid Spencer	641f5c9	2004-12-09 18:13:12 +0000	[diff] [blame]	1906	a constant is permitted.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1907
Reid Spencer	641f5c9	2004-12-09 18:13:12 +0000	[diff] [blame]	1908	<p>Undefined values indicate to the compiler that the program is well defined
				1909	no matter what value is used, giving the compiler more freedom to optimize.
				1910	</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1911	</div>
				1912
				1913	<!-- ======================================================================= -->
				1914	<div class="doc_subsection"><a name="constantexprs">Constant Expressions</a>
				1915	</div>
				1916
				1917	<div class="doc_text">
				1918
				1919	<p>Constant expressions are used to allow expressions involving other constants
				1920	to be used as constants. Constant expressions may be of any <a
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	1921	href="#t_firstclass">first class</a> type and may involve any LLVM operation
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1922	that does not have side effects (e.g. load and call are not supported). The
				1923	following is the syntax for constant expressions:</p>
				1924
				1925	<dl>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1926	<dt><b><tt>trunc ( CST to TYPE )</tt></b></dt>
				1927	<dd>Truncate a constant to another type. The bit size of CST must be larger
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	1928	than the bit size of TYPE. Both types must be integers.</dd>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1929
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1930	<dt><b><tt>zext ( CST to TYPE )</tt></b></dt>
				1931	<dd>Zero extend a constant to another type. The bit size of CST must be
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	1932	smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1933
				1934	<dt><b><tt>sext ( CST to TYPE )</tt></b></dt>
				1935	<dd>Sign extend a constant to another type. The bit size of CST must be
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	1936	smaller or equal to the bit size of TYPE. Both types must be integers.</dd>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1937
				1938	<dt><b><tt>fptrunc ( CST to TYPE )</tt></b></dt>
				1939	<dd>Truncate a floating point constant to another floating point type. The
				1940	size of CST must be larger than the size of TYPE. Both types must be
				1941	floating point.</dd>
				1942
				1943	<dt><b><tt>fpext ( CST to TYPE )</tt></b></dt>
				1944	<dd>Floating point extend a constant to another type. The size of CST must be
				1945	smaller or equal to the size of TYPE. Both types must be floating point.</dd>
				1946
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	1947	<dt><b><tt>fptoui ( CST to TYPE )</tt></b></dt>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1948	<dd>Convert a floating point constant to the corresponding unsigned integer
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	1949	constant. TYPE must be a scalar or vector integer type. CST must be of scalar
				1950	or vector floating point type. Both CST and TYPE must be scalars, or vectors
				1951	of the same number of elements. If the value won't fit in the integer type,
				1952	the results are undefined.</dd>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1953
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	1954	<dt><b><tt>fptosi ( CST to TYPE )</tt></b></dt>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1955	<dd>Convert a floating point constant to the corresponding signed integer
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	1956	constant. TYPE must be a scalar or vector integer type. CST must be of scalar
				1957	or vector floating point type. Both CST and TYPE must be scalars, or vectors
				1958	of the same number of elements. If the value won't fit in the integer type,
				1959	the results are undefined.</dd>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1960
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	1961	<dt><b><tt>uitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1962	<dd>Convert an unsigned integer constant to the corresponding floating point
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	1963	constant. TYPE must be a scalar or vector floating point type. CST must be of
				1964	scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
				1965	of the same number of elements. If the value won't fit in the floating point
				1966	type, the results are undefined.</dd>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1967
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	1968	<dt><b><tt>sitofp ( CST to TYPE )</tt></b></dt>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1969	<dd>Convert a signed integer constant to the corresponding floating point
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	1970	constant. TYPE must be a scalar or vector floating point type. CST must be of
				1971	scalar or vector integer type. Both CST and TYPE must be scalars, or vectors
				1972	of the same number of elements. If the value won't fit in the floating point
				1973	type, the results are undefined.</dd>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	1974
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	1975	<dt><b><tt>ptrtoint ( CST to TYPE )</tt></b></dt>
				1976	<dd>Convert a pointer typed constant to the corresponding integer constant
				1977	TYPE must be an integer type. CST must be of pointer type. The CST value is
				1978	zero extended, truncated, or unchanged to make it fit in TYPE.</dd>
				1979
				1980	<dt><b><tt>inttoptr ( CST to TYPE )</tt></b></dt>
				1981	<dd>Convert a integer constant to a pointer constant. TYPE must be a
				1982	pointer type. CST must be of integer type. The CST value is zero extended,
				1983	truncated, or unchanged to make it fit in a pointer size. This one is
				1984	<i>really</i> dangerous!</dd>
				1985
				1986	<dt><b><tt>bitcast ( CST to TYPE )</tt></b></dt>
Chris Lattner	789dee3	2009-02-28 18:27:03 +0000	[diff] [blame]	1987	<dd>Convert a constant, CST, to another TYPE. The constraints of the operands
				1988	are the same as those for the <a href="#i_bitcast">bitcast
				1989	instruction</a>.</dd>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	1990
				1991	<dt><b><tt>getelementptr ( CSTPTR, IDX0, IDX1, ... )</tt></b></dt>
				1992
				1993	<dd>Perform the <a href="#i_getelementptr">getelementptr operation</a> on
				1994	constants. As with the <a href="#i_getelementptr">getelementptr</a>
				1995	instruction, the index list may have zero or more indexes, which are required
				1996	to make sense for the type of "CSTPTR".</dd>
				1997
Robert Bocchino	7e97a6d	2006-01-10 19:31:34 +0000	[diff] [blame]	1998	<dt><b><tt>select ( COND, VAL1, VAL2 )</tt></b></dt>
				1999
				2000	<dd>Perform the <a href="#i_select">select operation</a> on
Reid Spencer	9965ee7	2006-12-04 19:23:19 +0000	[diff] [blame]	2001	constants.</dd>
				2002
				2003	<dt><b><tt>icmp COND ( VAL1, VAL2 )</tt></b></dt>
				2004	<dd>Performs the <a href="#i_icmp">icmp operation</a> on constants.</dd>
				2005
				2006	<dt><b><tt>fcmp COND ( VAL1, VAL2 )</tt></b></dt>
				2007	<dd>Performs the <a href="#i_fcmp">fcmp operation</a> on constants.</dd>
Robert Bocchino	7e97a6d	2006-01-10 19:31:34 +0000	[diff] [blame]	2008
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	2009	<dt><b><tt>vicmp COND ( VAL1, VAL2 )</tt></b></dt>
				2010	<dd>Performs the <a href="#i_vicmp">vicmp operation</a> on constants.</dd>
				2011
				2012	<dt><b><tt>vfcmp COND ( VAL1, VAL2 )</tt></b></dt>
				2013	<dd>Performs the <a href="#i_vfcmp">vfcmp operation</a> on constants.</dd>
				2014
Robert Bocchino	7e97a6d	2006-01-10 19:31:34 +0000	[diff] [blame]	2015	<dt><b><tt>extractelement ( VAL, IDX )</tt></b></dt>
				2016
				2017	<dd>Perform the <a href="#i_extractelement">extractelement
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	2018	operation</a> on constants.</dd>
Robert Bocchino	7e97a6d	2006-01-10 19:31:34 +0000	[diff] [blame]	2019
Robert Bocchino	f72fdfe	2006-01-15 20:48:27 +0000	[diff] [blame]	2020	<dt><b><tt>insertelement ( VAL, ELT, IDX )</tt></b></dt>
				2021
				2022	<dd>Perform the <a href="#i_insertelement">insertelement
Reid Spencer	9965ee7	2006-12-04 19:23:19 +0000	[diff] [blame]	2023	operation</a> on constants.</dd>
Robert Bocchino	f72fdfe	2006-01-15 20:48:27 +0000	[diff] [blame]	2024
Chris Lattner	016a0e5	2006-04-08 00:13:41 +0000	[diff] [blame]	2025
				2026	<dt><b><tt>shufflevector ( VEC1, VEC2, IDXMASK )</tt></b></dt>
				2027
				2028	<dd>Perform the <a href="#i_shufflevector">shufflevector
Reid Spencer	9965ee7	2006-12-04 19:23:19 +0000	[diff] [blame]	2029	operation</a> on constants.</dd>
Chris Lattner	016a0e5	2006-04-08 00:13:41 +0000	[diff] [blame]	2030
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2031	<dt><b><tt>OPCODE ( LHS, RHS )</tt></b></dt>
				2032
Reid Spencer	641f5c9	2004-12-09 18:13:12 +0000	[diff] [blame]	2033	<dd>Perform the specified operation of the LHS and RHS constants. OPCODE may
				2034	be any of the <a href="#binaryops">binary</a> or <a href="#bitwiseops">bitwise
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2035	binary</a> operations. The constraints on operands are the same as those for
				2036	the corresponding instruction (e.g. no bitwise operations on floating point
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	2037	values are allowed).</dd>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2038	</dl>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2039	</div>
Chris Lattner	b165261	2004-03-08 16:49:10 +0000	[diff] [blame]	2040
Nick Lewycky	49f8919	2009-04-04 07:22:01 +0000	[diff] [blame]	2041	<!-- ======================================================================= -->
				2042	<div class="doc_subsection"><a name="metadata">Embedded Metadata</a>
				2043	</div>
				2044
				2045	<div class="doc_text">
				2046
				2047	<p>Embedded metadata provides a way to attach arbitrary data to the
				2048	instruction stream without affecting the behaviour of the program. There are
				2049	two metadata primitives, strings and nodes. All metadata has the type of an
				2050	empty struct and is identified in syntax by a preceding exclamation point
				2051	('<tt>!</tt>').
				2052	</p>
				2053
				2054	<p>A metadata string is a string surrounded by double quotes. It can contain
				2055	any character by escaping non-printable characters with "\xx" where "xx" is
				2056	the two digit hex code. For example: "<tt>!"test\00"</tt>".
				2057	</p>
				2058
				2059	<p>Metadata nodes are represented with notation similar to structure constants
				2060	(a comma separated list of elements, surrounded by braces and preceeded by an
				2061	exclamation point). For example: "<tt>!{ { } !"test\00", i32 10}</tt>".
				2062	</p>
				2063
Nick Lewycky	b8f9b7a	2009-05-10 20:57:05 +0000	[diff] [blame]	2064	<p>A metadata node will attempt to track changes to the values it holds. In
				2065	the event that a value is deleted, it will be replaced with a typeless
				2066	"<tt>null</tt>", such as "<tt>{ } !{null, i32 0}</tt>".</p>
				2067
Nick Lewycky	49f8919	2009-04-04 07:22:01 +0000	[diff] [blame]	2068	<p>Optimizations may rely on metadata to provide additional information about
				2069	the program that isn't available in the instructions, or that isn't easily
				2070	computable. Similarly, the code generator may expect a certain metadata format
				2071	to be used to express debugging information.</p>
				2072	</div>
				2073
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2074	<!-- *********************************************************************** -->
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2075	<div class="doc_section"> <a name="othervalues">Other Values</a> </div>
				2076	<!-- *********************************************************************** -->
				2077
				2078	<!-- ======================================================================= -->
				2079	<div class="doc_subsection">
				2080	<a name="inlineasm">Inline Assembler Expressions</a>
				2081	</div>
				2082
				2083	<div class="doc_text">
				2084
				2085	<p>
				2086	LLVM supports inline assembler expressions (as opposed to <a href="#moduleasm">
				2087	Module-Level Inline Assembly</a>) through the use of a special value. This
				2088	value represents the inline assembler as a string (containing the instructions
				2089	to emit), a list of operand constraints (stored as a string), and a flag that
				2090	indicates whether or not the inline asm expression has side effects. An example
				2091	inline assembler expression is:
				2092	</p>
				2093
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2094	<div class="doc_code">
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2095	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2096	i32 (i32) asm "bswap $0", "=r,r"
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2097	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2098	</div>
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2099
				2100	<p>
				2101	Inline assembler expressions may <b>only</b> be used as the callee operand of
				2102	a <a href="#i_call"><tt>call</tt> instruction</a>. Thus, typically we have:
				2103	</p>
				2104
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2105	<div class="doc_code">
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2106	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2107	%X = call i32 asm "<a href="#int_bswap">bswap</a> $0", "=r,r"(i32 %Y)
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2108	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2109	</div>
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2110
				2111	<p>
				2112	Inline asms with side effects not visible in the constraint list must be marked
				2113	as having side effects. This is done through the use of the
				2114	'<tt>sideeffect</tt>' keyword, like so:
				2115	</p>
				2116
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2117	<div class="doc_code">
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2118	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2119	call void asm sideeffect "eieio", ""()
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2120	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	2121	</div>
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2122
				2123	<p>TODO: The format of the asm and constraints string still need to be
				2124	documented here. Constraints on what can be done (e.g. duplication, moving, etc
Chris Lattner	d552826	2008-10-04 18:36:02 +0000	[diff] [blame]	2125	need to be documented). This is probably best done by reference to another
				2126	document that covers inline asm from a holistic perspective.
Chris Lattner	98f013c	2006-01-25 23:47:57 +0000	[diff] [blame]	2127	</p>
				2128
				2129	</div>
				2130
				2131	<!-- *********************************************************************** -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2132	<div class="doc_section"> <a name="instref">Instruction Reference</a> </div>
				2133	<!-- *********************************************************************** -->
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2134
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2135	<div class="doc_text">
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2136
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2137	<p>The LLVM instruction set consists of several different
				2138	classifications of instructions: <a href="#terminators">terminator
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	2139	instructions</a>, <a href="#binaryops">binary instructions</a>,
				2140	<a href="#bitwiseops">bitwise binary instructions</a>, <a
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2141	href="#memoryops">memory instructions</a>, and <a href="#otherops">other
				2142	instructions</a>.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2143
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2144	</div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2145
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2146	<!-- ======================================================================= -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2147	<div class="doc_subsection"> <a name="terminators">Terminator
				2148	Instructions</a> </div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2149
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2150	<div class="doc_text">
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2151
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2152	<p>As mentioned <a href="#functionstructure">previously</a>, every
				2153	basic block in a program ends with a "Terminator" instruction, which
				2154	indicates which block should be executed after the current block is
				2155	finished. These terminator instructions typically yield a '<tt>void</tt>'
				2156	value: they produce control flow, not values (the one exception being
				2157	the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction).</p>
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	2158	<p>There are six different terminator instructions: the '<a
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2159	href="#i_ret"><tt>ret</tt></a>' instruction, the '<a href="#i_br"><tt>br</tt></a>'
				2160	instruction, the '<a href="#i_switch"><tt>switch</tt></a>' instruction,
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2161	the '<a href="#i_invoke"><tt>invoke</tt></a>' instruction, the '<a
				2162	href="#i_unwind"><tt>unwind</tt></a>' instruction, and the '<a
				2163	href="#i_unreachable"><tt>unreachable</tt></a>' instruction.</p>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2164
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2165	</div>
Chris Lattner	74d3f82	2004-12-09 17:30:23 +0000	[diff] [blame]	2166
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2167	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2168	<div class="doc_subsubsection"> <a name="i_ret">'<tt>ret</tt>'
				2169	Instruction</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2170	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2171	<h5>Syntax:</h5>
Dan Gohman	cc3132e	2008-10-04 19:00:07 +0000	[diff] [blame]	2172	<pre>
				2173	ret <type> <value> <i>; Return a value from a non-void function</i>
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	2174	ret void <i>; Return from void function</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2175	</pre>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2176
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2177	<h5>Overview:</h5>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2178
Dan Gohman	cc3132e	2008-10-04 19:00:07 +0000	[diff] [blame]	2179	<p>The '<tt>ret</tt>' instruction is used to return control flow (and
				2180	optionally a value) from a function back to the caller.</p>
John Criswell	417228d	2004-04-09 16:48:45 +0000	[diff] [blame]	2181	<p>There are two forms of the '<tt>ret</tt>' instruction: one that
Dan Gohman	cc3132e	2008-10-04 19:00:07 +0000	[diff] [blame]	2182	returns a value and then causes control flow, and one that just causes
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2183	control flow to occur.</p>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2184
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2185	<h5>Arguments:</h5>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2186
Dan Gohman	cc3132e	2008-10-04 19:00:07 +0000	[diff] [blame]	2187	<p>The '<tt>ret</tt>' instruction optionally accepts a single argument,
				2188	the return value. The type of the return value must be a
				2189	'<a href="#t_firstclass">first class</a>' type.</p>
				2190
				2191	<p>A function is not <a href="#wellformed">well formed</a> if
				2192	it it has a non-void return type and contains a '<tt>ret</tt>'
				2193	instruction with no return value or a return value with a type that
				2194	does not match its type, or if it has a void return type and contains
				2195	a '<tt>ret</tt>' instruction with a return value.</p>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2196
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2197	<h5>Semantics:</h5>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2198
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2199	<p>When the '<tt>ret</tt>' instruction is executed, control flow
				2200	returns back to the calling function's context. If the caller is a "<a
John Criswell	40db33f	2004-06-25 15:16:57 +0000	[diff] [blame]	2201	href="#i_call"><tt>call</tt></a>" instruction, execution continues at
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2202	the instruction after the call. If the caller was an "<a
				2203	href="#i_invoke"><tt>invoke</tt></a>" instruction, execution continues
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	2204	at the beginning of the "normal" destination block. If the instruction
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2205	returns a value, that value shall set the call or invoke instruction's
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	2206	return value.</p>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2207
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2208	<h5>Example:</h5>
Chris Lattner	da508ac	2008-04-23 04:59:35 +0000	[diff] [blame]	2209
				2210	<pre>
				2211	ret i32 5 <i>; Return an integer value of 5</i>
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	2212	ret void <i>; Return from a void function</i>
Bill Wendling	050ee8f	2009-02-28 22:12:54 +0000	[diff] [blame]	2213	ret { i32, i8 } { i32 4, i8 2 } <i>; Return a struct of values 4 and 2</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2214	</pre>
Dan Gohman	3065b61	2009-01-12 23:12:39 +0000	[diff] [blame]	2215
Dan Gohman	142ccc0	2009-01-24 15:58:40 +0000	[diff] [blame]	2216	<p>Note that the code generator does not yet fully support large
				2217	return values. The specific sizes that are currently supported are
				2218	dependent on the target. For integers, on 32-bit targets the limit
				2219	is often 64 bits, and on 64-bit targets the limit is often 128 bits.
				2220	For aggregate types, the current limits are dependent on the element
				2221	types; for example targets are often limited to 2 total integer
				2222	elements and 2 total floating-point elements.</p>
Dan Gohman	3065b61	2009-01-12 23:12:39 +0000	[diff] [blame]	2223
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2224	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2225	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2226	<div class="doc_subsubsection"> <a name="i_br">'<tt>br</tt>' Instruction</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2227	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2228	<h5>Syntax:</h5>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	2229	<pre> br i1 <cond>, label <iftrue>, label <iffalse><br> br label <dest> <i>; Unconditional branch</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2230	</pre>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2231	<h5>Overview:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2232	<p>The '<tt>br</tt>' instruction is used to cause control flow to
				2233	transfer to a different basic block in the current function. There are
				2234	two forms of this instruction, corresponding to a conditional branch
				2235	and an unconditional branch.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2236	<h5>Arguments:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2237	<p>The conditional branch form of the '<tt>br</tt>' instruction takes a
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	2238	single '<tt>i1</tt>' value and two '<tt>label</tt>' values. The
Reid Spencer	50c723a	2007-02-19 23:54:10 +0000	[diff] [blame]	2239	unconditional form of the '<tt>br</tt>' instruction takes a single
				2240	'<tt>label</tt>' value as a target.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2241	<h5>Semantics:</h5>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	2242	<p>Upon execution of a conditional '<tt>br</tt>' instruction, the '<tt>i1</tt>'
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2243	argument is evaluated. If the value is <tt>true</tt>, control flows
				2244	to the '<tt>iftrue</tt>' <tt>label</tt> argument. If "cond" is <tt>false</tt>,
				2245	control flows to the '<tt>iffalse</tt>' <tt>label</tt> argument.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2246	<h5>Example:</h5>
Chris Lattner	e648c28	2009-05-09 18:11:50 +0000	[diff] [blame]	2247	<pre>Test:<br> %cond = <a href="#i_icmp">icmp</a> eq i32 %a, %b<br> br i1 %cond, label %IfEqual, label %IfUnequal<br>IfEqual:<br> <a
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2248	href="#i_ret">ret</a> i32 1<br>IfUnequal:<br> <a href="#i_ret">ret</a> i32 0<br></pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2249	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2250	<!-- _______________________________________________________________________ -->
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2251	<div class="doc_subsubsection">
				2252	<a name="i_switch">'<tt>switch</tt>' Instruction</a>
				2253	</div>
				2254
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2255	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2256	<h5>Syntax:</h5>
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2257
				2258	<pre>
				2259	switch <intty> <value>, label <defaultdest> [ <intty> <val>, label <dest> ... ]
				2260	</pre>
				2261
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2262	<h5>Overview:</h5>
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2263
				2264	<p>The '<tt>switch</tt>' instruction is used to transfer control flow to one of
				2265	several different places. It is a generalization of the '<tt>br</tt>'
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2266	instruction, allowing a branch to occur to one of many possible
				2267	destinations.</p>
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2268
				2269
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2270	<h5>Arguments:</h5>
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2271
				2272	<p>The '<tt>switch</tt>' instruction uses three parameters: an integer
				2273	comparison value '<tt>value</tt>', a default '<tt>label</tt>' destination, and
				2274	an array of pairs of comparison value constants and '<tt>label</tt>'s. The
				2275	table is not allowed to contain duplicate constant entries.</p>
				2276
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2277	<h5>Semantics:</h5>
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2278
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2279	<p>The <tt>switch</tt> instruction specifies a table of values and
				2280	destinations. When the '<tt>switch</tt>' instruction is executed, this
John Criswell	bcbb18c	2004-06-25 16:05:06 +0000	[diff] [blame]	2281	table is searched for the given value. If the value is found, control flow is
				2282	transfered to the corresponding destination; otherwise, control flow is
				2283	transfered to the default destination.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2284
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2285	<h5>Implementation:</h5>
				2286
				2287	<p>Depending on properties of the target machine and the particular
				2288	<tt>switch</tt> instruction, this instruction may be code generated in different
John Criswell	bcbb18c	2004-06-25 16:05:06 +0000	[diff] [blame]	2289	ways. For example, it could be generated as a series of chained conditional
				2290	branches or with a lookup table.</p>
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2291
				2292	<h5>Example:</h5>
				2293
				2294	<pre>
				2295	<i>; Emulate a conditional br instruction</i>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	2296	%Val = <a href="#i_zext">zext</a> i1 %value to i32
Dan Gohman	623806e	2009-01-04 23:44:43 +0000	[diff] [blame]	2297	switch i32 %Val, label %truedest [ i32 0, label %falsedest ]
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2298
				2299	<i>; Emulate an unconditional br instruction</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2300	switch i32 0, label %dest [ ]
Chris Lattner	cf96c6c	2004-02-24 04:54:45 +0000	[diff] [blame]	2301
				2302	<i>; Implement a jump table:</i>
Dan Gohman	623806e	2009-01-04 23:44:43 +0000	[diff] [blame]	2303	switch i32 %val, label %otherwise [ i32 0, label %onzero
				2304	i32 1, label %onone
				2305	i32 2, label %ontwo ]
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2306	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2307	</div>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2308
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2309	<!-- _______________________________________________________________________ -->
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2310	<div class="doc_subsubsection">
				2311	<a name="i_invoke">'<tt>invoke</tt>' Instruction</a>
				2312	</div>
				2313
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2314	<div class="doc_text">
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2315
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2316	<h5>Syntax:</h5>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2317
				2318	<pre>
Devang Patel	0225623	2008-10-07 17:48:33 +0000	[diff] [blame]	2319	<result> = invoke [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>] <ptr to function ty> <function ptr val>(<function args>) [<a href="#fnattrs">fn attrs</a>]
Chris Lattner	6b7a008	2006-05-14 18:23:06 +0000	[diff] [blame]	2320	to label <normal label> unwind label <exception label>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2321	</pre>
				2322
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	2323	<h5>Overview:</h5>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2324
				2325	<p>The '<tt>invoke</tt>' instruction causes control to transfer to a specified
				2326	function, with the possibility of control flow transfer to either the
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	2327	'<tt>normal</tt>' label or the
				2328	'<tt>exception</tt>' label. If the callee function returns with the
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2329	"<tt><a href="#i_ret">ret</a></tt>" instruction, control flow will return to the
				2330	"normal" label. If the callee (or any indirect callees) returns with the "<a
John Criswell	02fdc6f	2005-05-12 16:52:32 +0000	[diff] [blame]	2331	href="#i_unwind"><tt>unwind</tt></a>" instruction, control is interrupted and
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	2332	continued at the dynamically nearest "exception" label.</p>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2333
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2334	<h5>Arguments:</h5>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2335
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2336	<p>This instruction requires several arguments:</p>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2337
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2338	<ol>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2339	<li>
Duncan Sands	16f122e	2007-03-30 12:22:09 +0000	[diff] [blame]	2340	The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2341	convention</a> the call should use. If none is specified, the call defaults
				2342	to using C calling conventions.
				2343	</li>
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	2344
				2345	<li>The optional <a href="#paramattrs">Parameter Attributes</a> list for
				2346	return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>',
				2347	and '<tt>inreg</tt>' attributes are valid here.</li>
				2348
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2349	<li>'<tt>ptr to function ty</tt>': shall be the signature of the pointer to
				2350	function value being invoked. In most cases, this is a direct function
				2351	invocation, but indirect <tt>invoke</tt>s are just as possible, branching off
				2352	an arbitrary pointer to function value.
				2353	</li>
				2354
				2355	<li>'<tt>function ptr val</tt>': An LLVM value containing a pointer to a
				2356	function to be invoked. </li>
				2357
				2358	<li>'<tt>function args</tt>': argument list whose types match the function
				2359	signature argument types. If the function signature indicates the function
				2360	accepts a variable number of arguments, the extra arguments can be
				2361	specified. </li>
				2362
				2363	<li>'<tt>normal label</tt>': the label reached when the called function
				2364	executes a '<tt><a href="#i_ret">ret</a></tt>' instruction. </li>
				2365
				2366	<li>'<tt>exception label</tt>': the label reached when a callee returns with
				2367	the <a href="#i_unwind"><tt>unwind</tt></a> instruction. </li>
				2368
Devang Patel	0225623	2008-10-07 17:48:33 +0000	[diff] [blame]	2369	<li>The optional <a href="#fnattrs">function attributes</a> list. Only
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	2370	'<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
				2371	'<tt>readnone</tt>' attributes are valid here.</li>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2372	</ol>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2373
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2374	<h5>Semantics:</h5>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2375
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2376	<p>This instruction is designed to operate as a standard '<tt><a
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2377	href="#i_call">call</a></tt>' instruction in most regards. The primary
				2378	difference is that it establishes an association with a label, which is used by
				2379	the runtime library to unwind the stack.</p>
				2380
				2381	<p>This instruction is used in languages with destructors to ensure that proper
				2382	cleanup is performed in the case of either a <tt>longjmp</tt> or a thrown
				2383	exception. Additionally, this is important for implementation of
				2384	'<tt>catch</tt>' clauses in high-level languages that support them.</p>
				2385
Dan Gohman	9069d89	2009-05-22 21:47:08 +0000	[diff] [blame]	2386	<p>It is not valid to reference the return value of an invoke call from
				2387	anywhere not dominated by the normal label, since an unwind does not
				2388	provide a return value.</p>
				2389
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2390	<h5>Example:</h5>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	2391	<pre>
Nick Lewycky	084ab47	2008-03-16 07:18:12 +0000	[diff] [blame]	2392	%retval = invoke i32 @Test(i32 15) to label %Continue
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	2393	unwind label %TestCleanup <i>; {i32}:retval set</i>
Nick Lewycky	084ab47	2008-03-16 07:18:12 +0000	[diff] [blame]	2394	%retval = invoke <a href="#callingconv">coldcc</a> i32 %Testfnptr(i32 15) to label %Continue
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	2395	unwind label %TestCleanup <i>; {i32}:retval set</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2396	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2397	</div>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2398
				2399
Chris Lattner	5ed6061	2003-09-03 00:41:47 +0000	[diff] [blame]	2400	<!-- _______________________________________________________________________ -->
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2401
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2402	<div class="doc_subsubsection"> <a name="i_unwind">'<tt>unwind</tt>'
				2403	Instruction</a> </div>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2404
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2405	<div class="doc_text">
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2406
Chris Lattner	5ed6061	2003-09-03 00:41:47 +0000	[diff] [blame]	2407	<h5>Syntax:</h5>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2408	<pre>
				2409	unwind
				2410	</pre>
				2411
Chris Lattner	5ed6061	2003-09-03 00:41:47 +0000	[diff] [blame]	2412	<h5>Overview:</h5>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2413
				2414	<p>The '<tt>unwind</tt>' instruction unwinds the stack, continuing control flow
				2415	at the first callee in the dynamic call stack which used an <a
				2416	href="#i_invoke"><tt>invoke</tt></a> instruction to perform the call. This is
				2417	primarily used to implement exception handling.</p>
				2418
Chris Lattner	5ed6061	2003-09-03 00:41:47 +0000	[diff] [blame]	2419	<h5>Semantics:</h5>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2420
Chris Lattner	fe8519c	2008-04-19 21:01:16 +0000	[diff] [blame]	2421	<p>The '<tt>unwind</tt>' instruction causes execution of the current function to
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2422	immediately halt. The dynamic call stack is then searched for the first <a
				2423	href="#i_invoke"><tt>invoke</tt></a> instruction on the call stack. Once found,
				2424	execution continues at the "exceptional" destination block specified by the
				2425	<tt>invoke</tt> instruction. If there is no <tt>invoke</tt> instruction in the
				2426	dynamic call chain, undefined behavior results.</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2427	</div>
Chris Lattner	08b7d5b	2004-10-16 18:04:13 +0000	[diff] [blame]	2428
				2429	<!-- _______________________________________________________________________ -->
				2430
				2431	<div class="doc_subsubsection"> <a name="i_unreachable">'<tt>unreachable</tt>'
				2432	Instruction</a> </div>
				2433
				2434	<div class="doc_text">
				2435
				2436	<h5>Syntax:</h5>
				2437	<pre>
				2438	unreachable
				2439	</pre>
				2440
				2441	<h5>Overview:</h5>
				2442
				2443	<p>The '<tt>unreachable</tt>' instruction has no defined semantics. This
				2444	instruction is used to inform the optimizer that a particular portion of the
				2445	code is not reachable. This can be used to indicate that the code after a
				2446	no-return function cannot be reached, and other facts.</p>
				2447
				2448	<h5>Semantics:</h5>
				2449
				2450	<p>The '<tt>unreachable</tt>' instruction has no defined semantics.</p>
				2451	</div>
				2452
				2453
				2454
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2455	<!-- ======================================================================= -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2456	<div class="doc_subsection"> <a name="binaryops">Binary Operations</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2457	<div class="doc_text">
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2458	<p>Binary operators are used to do most of the computation in a
Chris Lattner	81f9297	2008-04-01 18:47:32 +0000	[diff] [blame]	2459	program. They require two operands of the same type, execute an operation on them, and
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	2460	produce a single value. The operands might represent
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	2461	multiple data, as is the case with the <a href="#t_vector">vector</a> data type.
Chris Lattner	81f9297	2008-04-01 18:47:32 +0000	[diff] [blame]	2462	The result value has the same type as its operands.</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2463	<p>There are several different binary operators:</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2464	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2465	<!-- _______________________________________________________________________ -->
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2466	<div class="doc_subsubsection">
				2467	<a name="i_add">'<tt>add</tt>' Instruction</a>
				2468	</div>
				2469
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2470	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2471
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2472	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2473
				2474	<pre>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2475	<result> = add <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2476	</pre>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2477
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2478	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2479
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2480	<p>The '<tt>add</tt>' instruction returns the sum of its two operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2481
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2482	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2483
				2484	<p>The two arguments to the '<tt>add</tt>' instruction must be <a
				2485	href="#t_integer">integer</a>, <a href="#t_floating">floating point</a>, or
				2486	<a href="#t_vector">vector</a> values. Both arguments must have identical
				2487	types.</p>
				2488
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2489	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2490
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2491	<p>The value produced is the integer or floating point sum of the two
				2492	operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2493
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2494	<p>If an integer sum has unsigned overflow, the result returned is the
				2495	mathematical result modulo 2<sup>n</sup>, where n is the bit width of
				2496	the result.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2497
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2498	<p>Because LLVM integers use a two's complement representation, this
				2499	instruction is appropriate for both signed and unsigned integers.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2500
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2501	<h5>Example:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2502
				2503	<pre>
				2504	<result> = add i32 4, %var <i>; yields {i32}:result = 4 + %var</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2505	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2506	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2507	<!-- _______________________________________________________________________ -->
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2508	<div class="doc_subsubsection">
				2509	<a name="i_sub">'<tt>sub</tt>' Instruction</a>
				2510	</div>
				2511
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2512	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2513
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2514	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2515
				2516	<pre>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2517	<result> = sub <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2518	</pre>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2519
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2520	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2521
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2522	<p>The '<tt>sub</tt>' instruction returns the difference of its two
				2523	operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2524
				2525	<p>Note that the '<tt>sub</tt>' instruction is used to represent the
				2526	'<tt>neg</tt>' instruction present in most other intermediate
				2527	representations.</p>
				2528
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2529	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2530
				2531	<p>The two arguments to the '<tt>sub</tt>' instruction must be <a
				2532	href="#t_integer">integer</a>, <a href="#t_floating">floating point</a>,
				2533	or <a href="#t_vector">vector</a> values. Both arguments must have identical
				2534	types.</p>
				2535
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2536	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2537
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2538	<p>The value produced is the integer or floating point difference of
				2539	the two operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2540
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2541	<p>If an integer difference has unsigned overflow, the result returned is the
				2542	mathematical result modulo 2<sup>n</sup>, where n is the bit width of
				2543	the result.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2544
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2545	<p>Because LLVM integers use a two's complement representation, this
				2546	instruction is appropriate for both signed and unsigned integers.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2547
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2548	<h5>Example:</h5>
Bill Wendling	2d8b9a8	2007-05-29 09:42:13 +0000	[diff] [blame]	2549	<pre>
				2550	<result> = sub i32 4, %var <i>; yields {i32}:result = 4 - %var</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2551	<result> = sub i32 0, %val <i>; yields {i32}:result = -%var</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2552	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2553	</div>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2554
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2555	<!-- _______________________________________________________________________ -->
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2556	<div class="doc_subsubsection">
				2557	<a name="i_mul">'<tt>mul</tt>' Instruction</a>
				2558	</div>
				2559
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2560	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2561
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2562	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2563	<pre> <result> = mul <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2564	</pre>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2565	<h5>Overview:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2566	<p>The '<tt>mul</tt>' instruction returns the product of its two
				2567	operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2568
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2569	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2570
				2571	<p>The two arguments to the '<tt>mul</tt>' instruction must be <a
				2572	href="#t_integer">integer</a>, <a href="#t_floating">floating point</a>,
				2573	or <a href="#t_vector">vector</a> values. Both arguments must have identical
				2574	types.</p>
				2575
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2576	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2577
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2578	<p>The value produced is the integer or floating point product of the
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2579	two operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2580
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2581	<p>If the result of an integer multiplication has unsigned overflow,
				2582	the result returned is the mathematical result modulo
				2583	2<sup>n</sup>, where n is the bit width of the result.</p>
				2584	<p>Because LLVM integers use a two's complement representation, and the
				2585	result is the same width as the operands, this instruction returns the
				2586	correct result for both signed and unsigned integers. If a full product
				2587	(e.g. <tt>i32</tt>x<tt>i32</tt>-><tt>i64</tt>) is needed, the operands
				2588	should be sign-extended or zero-extended as appropriate to the
				2589	width of the full product.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2590	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2591	<pre> <result> = mul i32 4, %var <i>; yields {i32}:result = 4 * %var</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2592	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2593	</div>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2594
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2595	<!-- _______________________________________________________________________ -->
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2596	<div class="doc_subsubsection"> <a name="i_udiv">'<tt>udiv</tt>' Instruction
				2597	</a></div>
				2598	<div class="doc_text">
				2599	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2600	<pre> <result> = udiv <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2601	</pre>
				2602	<h5>Overview:</h5>
				2603	<p>The '<tt>udiv</tt>' instruction returns the quotient of its two
				2604	operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2605
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2606	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2607
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2608	<p>The two arguments to the '<tt>udiv</tt>' instruction must be
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2609	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				2610	values. Both arguments must have identical types.</p>
				2611
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2612	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2613
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2614	<p>The value produced is the unsigned integer quotient of the two operands.</p>
				2615	<p>Note that unsigned integer division and signed integer division are distinct
				2616	operations; for signed integer division, use '<tt>sdiv</tt>'.</p>
				2617	<p>Division by zero leads to undefined behavior.</p>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2618	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2619	<pre> <result> = udiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2620	</pre>
				2621	</div>
				2622	<!-- _______________________________________________________________________ -->
				2623	<div class="doc_subsubsection"> <a name="i_sdiv">'<tt>sdiv</tt>' Instruction
				2624	</a> </div>
				2625	<div class="doc_text">
				2626	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2627	<pre>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2628	<result> = sdiv <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2629	</pre>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2630
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2631	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2632
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2633	<p>The '<tt>sdiv</tt>' instruction returns the quotient of its two
				2634	operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2635
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2636	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2637
				2638	<p>The two arguments to the '<tt>sdiv</tt>' instruction must be
				2639	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				2640	values. Both arguments must have identical types.</p>
				2641
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2642	<h5>Semantics:</h5>
Chris Lattner	1429e6f	2008-04-01 18:45:27 +0000	[diff] [blame]	2643	<p>The value produced is the signed integer quotient of the two operands rounded towards zero.</p>
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2644	<p>Note that signed integer division and unsigned integer division are distinct
				2645	operations; for unsigned integer division, use '<tt>udiv</tt>'.</p>
				2646	<p>Division by zero leads to undefined behavior. Overflow also leads to
				2647	undefined behavior; this is a rare case, but can occur, for example,
				2648	by doing a 32-bit division of -2147483648 by -1.</p>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2649	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2650	<pre> <result> = sdiv i32 4, %var <i>; yields {i32}:result = 4 / %var</i>
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2651	</pre>
				2652	</div>
				2653	<!-- _______________________________________________________________________ -->
				2654	<div class="doc_subsubsection"> <a name="i_fdiv">'<tt>fdiv</tt>'
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2655	Instruction</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2656	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2657	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2658	<pre>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2659	<result> = fdiv <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2660	</pre>
				2661	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2662
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2663	<p>The '<tt>fdiv</tt>' instruction returns the quotient of its two
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2664	operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2665
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2666	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2667
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	2668	<p>The two arguments to the '<tt>fdiv</tt>' instruction must be
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2669	<a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a>
				2670	of floating point values. Both arguments must have identical types.</p>
				2671
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2672	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2673
Reid Spencer	7e80b0b	2006-10-26 06:15:43 +0000	[diff] [blame]	2674	<p>The value produced is the floating point quotient of the two operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2675
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2676	<h5>Example:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2677
				2678	<pre>
				2679	<result> = fdiv float 4.0, %var <i>; yields {float}:result = 4.0 / %var</i>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2680	</pre>
				2681	</div>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2682
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2683	<!-- _______________________________________________________________________ -->
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2684	<div class="doc_subsubsection"> <a name="i_urem">'<tt>urem</tt>' Instruction</a>
				2685	</div>
				2686	<div class="doc_text">
				2687	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2688	<pre> <result> = urem <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2689	</pre>
				2690	<h5>Overview:</h5>
				2691	<p>The '<tt>urem</tt>' instruction returns the remainder from the
				2692	unsigned division of its two arguments.</p>
				2693	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2694	<p>The two arguments to the '<tt>urem</tt>' instruction must be
				2695	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				2696	values. Both arguments must have identical types.</p>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2697	<h5>Semantics:</h5>
				2698	<p>This instruction returns the unsigned integer <i>remainder</i> of a division.
Chris Lattner	1429e6f	2008-04-01 18:45:27 +0000	[diff] [blame]	2699	This instruction always performs an unsigned division to get the remainder.</p>
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2700	<p>Note that unsigned integer remainder and signed integer remainder are
				2701	distinct operations; for signed integer remainder, use '<tt>srem</tt>'.</p>
				2702	<p>Taking the remainder of a division by zero leads to undefined behavior.</p>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2703	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2704	<pre> <result> = urem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2705	</pre>
				2706
				2707	</div>
				2708	<!-- _______________________________________________________________________ -->
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2709	<div class="doc_subsubsection">
				2710	<a name="i_srem">'<tt>srem</tt>' Instruction</a>
				2711	</div>
				2712
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2713	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2714
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2715	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2716
				2717	<pre>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2718	<result> = srem <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2719	</pre>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2720
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2721	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2722
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2723	<p>The '<tt>srem</tt>' instruction returns the remainder from the
Dan Gohman	08143e3	2007-11-05 23:35:22 +0000	[diff] [blame]	2724	signed division of its two operands. This instruction can also take
				2725	<a href="#t_vector">vector</a> versions of the values in which case
				2726	the elements must be integers.</p>
Chris Lattner	b8f816e	2008-01-04 04:33:49 +0000	[diff] [blame]	2727
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2728	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2729
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2730	<p>The two arguments to the '<tt>srem</tt>' instruction must be
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2731	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				2732	values. Both arguments must have identical types.</p>
				2733
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2734	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2735
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2736	<p>This instruction returns the <i>remainder</i> of a division (where the result
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2737	has the same sign as the dividend, <tt>op1</tt>), not the <i>modulo</i>
				2738	operator (where the result has the same sign as the divisor, <tt>op2</tt>) of
Reid Spencer	806ad6a	2007-03-24 22:23:39 +0000	[diff] [blame]	2739	a value. For more information about the difference, see <a
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2740	href="http://mathforum.org/dr.math/problems/anne.4.28.99.html">The
Reid Spencer	806ad6a	2007-03-24 22:23:39 +0000	[diff] [blame]	2741	Math Forum</a>. For a table of how this is implemented in various languages,
Reid Spencer	db3b93b	2007-03-24 22:40:44 +0000	[diff] [blame]	2742	please see <a href="http://en.wikipedia.org/wiki/Modulo_operation">
Reid Spencer	806ad6a	2007-03-24 22:23:39 +0000	[diff] [blame]	2743	Wikipedia: modulo operation</a>.</p>
Chris Lattner	2f2427e	2008-01-28 00:36:27 +0000	[diff] [blame]	2744	<p>Note that signed integer remainder and unsigned integer remainder are
				2745	distinct operations; for unsigned integer remainder, use '<tt>urem</tt>'.</p>
				2746	<p>Taking the remainder of a division by zero leads to undefined behavior.
				2747	Overflow also leads to undefined behavior; this is a rare case, but can occur,
				2748	for example, by taking the remainder of a 32-bit division of -2147483648 by -1.
				2749	(The remainder doesn't actually overflow, but this rule lets srem be
				2750	implemented using instructions that return both the result of the division
				2751	and the remainder.)</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2752	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2753	<pre> <result> = srem i32 4, %var <i>; yields {i32}:result = 4 % %var</i>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2754	</pre>
				2755
				2756	</div>
				2757	<!-- _______________________________________________________________________ -->
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2758	<div class="doc_subsubsection">
				2759	<a name="i_frem">'<tt>frem</tt>' Instruction</a> </div>
				2760
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2761	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2762
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2763	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2764	<pre> <result> = frem <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2765	</pre>
				2766	<h5>Overview:</h5>
				2767	<p>The '<tt>frem</tt>' instruction returns the remainder from the
				2768	division of its two operands.</p>
				2769	<h5>Arguments:</h5>
				2770	<p>The two arguments to the '<tt>frem</tt>' instruction must be
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2771	<a href="#t_floating">floating point</a> or <a href="#t_vector">vector</a>
				2772	of floating point values. Both arguments must have identical types.</p>
				2773
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2774	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2775
Chris Lattner	1429e6f	2008-04-01 18:45:27 +0000	[diff] [blame]	2776	<p>This instruction returns the <i>remainder</i> of a division.
				2777	The remainder has the same sign as the dividend.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2778
Reid Spencer	7eb55b3	2006-11-02 01:53:59 +0000	[diff] [blame]	2779	<h5>Example:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2780
				2781	<pre>
				2782	<result> = frem float 4.0, %var <i>; yields {float}:result = 4.0 % %var</i>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2783	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2784	</div>
Robert Bocchino	820bc75b	2006-02-17 21:18:08 +0000	[diff] [blame]	2785
Reid Spencer	2ab0193	2007-02-02 13:57:07 +0000	[diff] [blame]	2786	<!-- ======================================================================= -->
				2787	<div class="doc_subsection"> <a name="bitwiseops">Bitwise Binary
				2788	Operations</a> </div>
				2789	<div class="doc_text">
				2790	<p>Bitwise binary operators are used to do various forms of
				2791	bit-twiddling in a program. They are generally very efficient
				2792	instructions and can commonly be strength reduced from other
Chris Lattner	1429e6f	2008-04-01 18:45:27 +0000	[diff] [blame]	2793	instructions. They require two operands of the same type, execute an operation on them,
				2794	and produce a single value. The resulting value is the same type as its operands.</p>
Reid Spencer	2ab0193	2007-02-02 13:57:07 +0000	[diff] [blame]	2795	</div>
				2796
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2797	<!-- _______________________________________________________________________ -->
				2798	<div class="doc_subsubsection"> <a name="i_shl">'<tt>shl</tt>'
				2799	Instruction</a> </div>
				2800	<div class="doc_text">
				2801	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2802	<pre> <result> = shl <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2803	</pre>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2804
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2805	<h5>Overview:</h5>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2806
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2807	<p>The '<tt>shl</tt>' instruction returns the first operand shifted to
				2808	the left a specified number of bits.</p>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2809
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2810	<h5>Arguments:</h5>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2811
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2812	<p>Both arguments to the '<tt>shl</tt>' instruction must be the same <a
Nate Begeman	fecbc8c	2008-07-29 15:49:41 +0000	[diff] [blame]	2813	href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2814	type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2815
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2816	<h5>Semantics:</h5>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2817
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2818	<p>The value produced is <tt>op1</tt> * 2<sup><tt>op2</tt></sup> mod 2<sup>n</sup>,
				2819	where n is the width of the result. If <tt>op2</tt> is (statically or dynamically) negative or
Mon P Wang	68d4eee	2008-12-10 08:55:09 +0000	[diff] [blame]	2820	equal to or larger than the number of bits in <tt>op1</tt>, the result is undefined.
				2821	If the arguments are vectors, each vector element of <tt>op1</tt> is shifted by the
				2822	corresponding shift amount in <tt>op2</tt>.</p>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2823
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2824	<h5>Example:</h5><pre>
				2825	<result> = shl i32 4, %var <i>; yields {i32}: 4 << %var</i>
				2826	<result> = shl i32 4, 2 <i>; yields {i32}: 16</i>
				2827	<result> = shl i32 1, 10 <i>; yields {i32}: 1024</i>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2828	<result> = shl i32 1, 32 <i>; undefined</i>
Mon P Wang	4dd832d	2008-12-09 05:46:39 +0000	[diff] [blame]	2829	<result> = shl <2 x i32> < i32 1, i32 1>, < i32 1, i32 2> <i>; yields: result=<2 x i32> < i32 2, i32 4></i>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2830	</pre>
				2831	</div>
				2832	<!-- _______________________________________________________________________ -->
				2833	<div class="doc_subsubsection"> <a name="i_lshr">'<tt>lshr</tt>'
				2834	Instruction</a> </div>
				2835	<div class="doc_text">
				2836	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2837	<pre> <result> = lshr <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2838	</pre>
				2839
				2840	<h5>Overview:</h5>
				2841	<p>The '<tt>lshr</tt>' instruction (logical shift right) returns the first
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	2842	operand shifted to the right a specified number of bits with zero fill.</p>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2843
				2844	<h5>Arguments:</h5>
				2845	<p>Both arguments to the '<tt>lshr</tt>' instruction must be the same
Nate Begeman	fecbc8c	2008-07-29 15:49:41 +0000	[diff] [blame]	2846	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2847	type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2848
				2849	<h5>Semantics:</h5>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2850
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2851	<p>This instruction always performs a logical shift right operation. The most
				2852	significant bits of the result will be filled with zero bits after the
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2853	shift. If <tt>op2</tt> is (statically or dynamically) equal to or larger than
Mon P Wang	68d4eee	2008-12-10 08:55:09 +0000	[diff] [blame]	2854	the number of bits in <tt>op1</tt>, the result is undefined. If the arguments are
				2855	vectors, each vector element of <tt>op1</tt> is shifted by the corresponding shift
				2856	amount in <tt>op2</tt>.</p>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2857
				2858	<h5>Example:</h5>
				2859	<pre>
				2860	<result> = lshr i32 4, 1 <i>; yields {i32}:result = 2</i>
				2861	<result> = lshr i32 4, 2 <i>; yields {i32}:result = 1</i>
				2862	<result> = lshr i8 4, 3 <i>; yields {i8}:result = 0</i>
				2863	<result> = lshr i8 -2, 1 <i>; yields {i8}:result = 0x7FFFFFFF </i>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2864	<result> = lshr i32 1, 32 <i>; undefined</i>
Mon P Wang	4dd832d	2008-12-09 05:46:39 +0000	[diff] [blame]	2865	<result> = lshr <2 x i32> < i32 -2, i32 4>, < i32 1, i32 2> <i>; yields: result=<2 x i32> < i32 0x7FFFFFFF, i32 1></i>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2866	</pre>
				2867	</div>
				2868
Reid Spencer	2ab0193	2007-02-02 13:57:07 +0000	[diff] [blame]	2869	<!-- _______________________________________________________________________ -->
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2870	<div class="doc_subsubsection"> <a name="i_ashr">'<tt>ashr</tt>'
				2871	Instruction</a> </div>
				2872	<div class="doc_text">
				2873
				2874	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2875	<pre> <result> = ashr <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2876	</pre>
				2877
				2878	<h5>Overview:</h5>
				2879	<p>The '<tt>ashr</tt>' instruction (arithmetic shift right) returns the first
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	2880	operand shifted to the right a specified number of bits with sign extension.</p>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2881
				2882	<h5>Arguments:</h5>
				2883	<p>Both arguments to the '<tt>ashr</tt>' instruction must be the same
Nate Begeman	fecbc8c	2008-07-29 15:49:41 +0000	[diff] [blame]	2884	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2885	type. '<tt>op2</tt>' is treated as an unsigned value.</p>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2886
				2887	<h5>Semantics:</h5>
				2888	<p>This instruction always performs an arithmetic shift right operation,
				2889	The most significant bits of the result will be filled with the sign bit
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2890	of <tt>op1</tt>. If <tt>op2</tt> is (statically or dynamically) equal to or
Mon P Wang	68d4eee	2008-12-10 08:55:09 +0000	[diff] [blame]	2891	larger than the number of bits in <tt>op1</tt>, the result is undefined. If the
				2892	arguments are vectors, each vector element of <tt>op1</tt> is shifted by the
				2893	corresponding shift amount in <tt>op2</tt>.</p>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2894
				2895	<h5>Example:</h5>
				2896	<pre>
				2897	<result> = ashr i32 4, 1 <i>; yields {i32}:result = 2</i>
				2898	<result> = ashr i32 4, 2 <i>; yields {i32}:result = 1</i>
				2899	<result> = ashr i8 4, 3 <i>; yields {i8}:result = 0</i>
				2900	<result> = ashr i8 -2, 1 <i>; yields {i8}:result = -1</i>
Chris Lattner	f0e5011	2007-10-03 21:01:14 +0000	[diff] [blame]	2901	<result> = ashr i32 1, 32 <i>; undefined</i>
Mon P Wang	4dd832d	2008-12-09 05:46:39 +0000	[diff] [blame]	2902	<result> = ashr <2 x i32> < i32 -2, i32 4>, < i32 1, i32 3> <i>; yields: result=<2 x i32> < i32 -1, i32 0></i>
Reid Spencer	04e259b	2007-01-31 21:39:12 +0000	[diff] [blame]	2903	</pre>
				2904	</div>
				2905
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2906	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2907	<div class="doc_subsubsection"> <a name="i_and">'<tt>and</tt>'
				2908	Instruction</a> </div>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2909
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2910	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2911
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2912	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2913
				2914	<pre>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2915	<result> = and <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2916	</pre>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2917
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2918	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2919
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2920	<p>The '<tt>and</tt>' instruction returns the bitwise logical and of
				2921	its two operands.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2922
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2923	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2924
				2925	<p>The two arguments to the '<tt>and</tt>' instruction must be
				2926	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				2927	values. Both arguments must have identical types.</p>
				2928
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2929	<h5>Semantics:</h5>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2930	<p>The truth table used for the '<tt>and</tt>' instruction is:</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2931	<p> </p>
Bill Wendling	5703c6e	2008-09-07 10:29:20 +0000	[diff] [blame]	2932	<div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2933	<table border="1" cellspacing="0" cellpadding="4">
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2934	<tbody>
				2935	<tr>
				2936	<td>In0</td>
				2937	<td>In1</td>
				2938	<td>Out</td>
				2939	</tr>
				2940	<tr>
				2941	<td>0</td>
				2942	<td>0</td>
				2943	<td>0</td>
				2944	</tr>
				2945	<tr>
				2946	<td>0</td>
				2947	<td>1</td>
				2948	<td>0</td>
				2949	</tr>
				2950	<tr>
				2951	<td>1</td>
				2952	<td>0</td>
				2953	<td>0</td>
				2954	</tr>
				2955	<tr>
				2956	<td>1</td>
				2957	<td>1</td>
				2958	<td>1</td>
				2959	</tr>
				2960	</tbody>
				2961	</table>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	2962	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2963	<h5>Example:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2964	<pre>
				2965	<result> = and i32 4, %var <i>; yields {i32}:result = 4 & %var</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	2966	<result> = and i32 15, 40 <i>; yields {i32}:result = 8</i>
				2967	<result> = and i32 4, 8 <i>; yields {i32}:result = 0</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2968	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2969	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2970	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2971	<div class="doc_subsubsection"> <a name="i_or">'<tt>or</tt>' Instruction</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2972	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2973	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	2974	<pre> <result> = or <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2975	</pre>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2976	<h5>Overview:</h5>
				2977	<p>The '<tt>or</tt>' instruction returns the bitwise logical inclusive
				2978	or of its two operands.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2979	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	2980
				2981	<p>The two arguments to the '<tt>or</tt>' instruction must be
				2982	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				2983	values. Both arguments must have identical types.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	2984	<h5>Semantics:</h5>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	2985	<p>The truth table used for the '<tt>or</tt>' instruction is:</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2986	<p> </p>
Bill Wendling	5703c6e	2008-09-07 10:29:20 +0000	[diff] [blame]	2987	<div>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	2988	<table border="1" cellspacing="0" cellpadding="4">
				2989	<tbody>
				2990	<tr>
				2991	<td>In0</td>
				2992	<td>In1</td>
				2993	<td>Out</td>
				2994	</tr>
				2995	<tr>
				2996	<td>0</td>
				2997	<td>0</td>
				2998	<td>0</td>
				2999	</tr>
				3000	<tr>
				3001	<td>0</td>
				3002	<td>1</td>
				3003	<td>1</td>
				3004	</tr>
				3005	<tr>
				3006	<td>1</td>
				3007	<td>0</td>
				3008	<td>1</td>
				3009	</tr>
				3010	<tr>
				3011	<td>1</td>
				3012	<td>1</td>
				3013	<td>1</td>
				3014	</tr>
				3015	</tbody>
				3016	</table>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	3017	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3018	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3019	<pre> <result> = or i32 4, %var <i>; yields {i32}:result = 4 \| %var</i>
				3020	<result> = or i32 15, 40 <i>; yields {i32}:result = 47</i>
				3021	<result> = or i32 4, 8 <i>; yields {i32}:result = 12</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3022	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3023	</div>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3024	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3025	<div class="doc_subsubsection"> <a name="i_xor">'<tt>xor</tt>'
				3026	Instruction</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3027	<div class="doc_text">
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3028	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	3029	<pre> <result> = xor <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3030	</pre>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3031	<h5>Overview:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3032	<p>The '<tt>xor</tt>' instruction returns the bitwise logical exclusive
				3033	or of its two operands. The <tt>xor</tt> is used to implement the
				3034	"one's complement" operation, which is the "~" operator in C.</p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3035	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	3036	<p>The two arguments to the '<tt>xor</tt>' instruction must be
				3037	<a href="#t_integer">integer</a> or <a href="#t_vector">vector</a> of integer
				3038	values. Both arguments must have identical types.</p>
				3039
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3040	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	3041
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3042	<p>The truth table used for the '<tt>xor</tt>' instruction is:</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3043	<p> </p>
Bill Wendling	5703c6e	2008-09-07 10:29:20 +0000	[diff] [blame]	3044	<div>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3045	<table border="1" cellspacing="0" cellpadding="4">
				3046	<tbody>
				3047	<tr>
				3048	<td>In0</td>
				3049	<td>In1</td>
				3050	<td>Out</td>
				3051	</tr>
				3052	<tr>
				3053	<td>0</td>
				3054	<td>0</td>
				3055	<td>0</td>
				3056	</tr>
				3057	<tr>
				3058	<td>0</td>
				3059	<td>1</td>
				3060	<td>1</td>
				3061	</tr>
				3062	<tr>
				3063	<td>1</td>
				3064	<td>0</td>
				3065	<td>1</td>
				3066	</tr>
				3067	<tr>
				3068	<td>1</td>
				3069	<td>1</td>
				3070	<td>0</td>
				3071	</tr>
				3072	</tbody>
				3073	</table>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	3074	</div>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3075	<p> </p>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3076	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3077	<pre> <result> = xor i32 4, %var <i>; yields {i32}:result = 4 ^ %var</i>
				3078	<result> = xor i32 15, 40 <i>; yields {i32}:result = 39</i>
				3079	<result> = xor i32 4, 8 <i>; yields {i32}:result = 12</i>
				3080	<result> = xor i32 %V, -1 <i>; yields {i32}:result = ~%V</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3081	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3082	</div>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3083
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3084	<!-- ======================================================================= -->
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3085	<div class="doc_subsection">
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3086	<a name="vectorops">Vector Operations</a>
				3087	</div>
				3088
				3089	<div class="doc_text">
				3090
				3091	<p>LLVM supports several instructions to represent vector operations in a
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	3092	target-independent manner. These instructions cover the element-access and
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3093	vector-specific operations needed to process vectors effectively. While LLVM
				3094	does directly support these vector operations, many sophisticated algorithms
				3095	will want to use target-specific intrinsics to take full advantage of a specific
				3096	target.</p>
				3097
				3098	</div>
				3099
				3100	<!-- _______________________________________________________________________ -->
				3101	<div class="doc_subsubsection">
				3102	<a name="i_extractelement">'<tt>extractelement</tt>' Instruction</a>
				3103	</div>
				3104
				3105	<div class="doc_text">
				3106
				3107	<h5>Syntax:</h5>
				3108
				3109	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3110	<result> = extractelement <n x <ty>> <val>, i32 <idx> <i>; yields <ty></i>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3111	</pre>
				3112
				3113	<h5>Overview:</h5>
				3114
				3115	<p>
				3116	The '<tt>extractelement</tt>' instruction extracts a single scalar
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	3117	element from a vector at a specified index.
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3118	</p>
				3119
				3120
				3121	<h5>Arguments:</h5>
				3122
				3123	<p>
				3124	The first operand of an '<tt>extractelement</tt>' instruction is a
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	3125	value of <a href="#t_vector">vector</a> type. The second operand is
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3126	an index indicating the position from which to extract the element.
				3127	The index may be a variable.</p>
				3128
				3129	<h5>Semantics:</h5>
				3130
				3131	<p>
				3132	The result is a scalar of the same type as the element type of
				3133	<tt>val</tt>. Its value is the value at position <tt>idx</tt> of
				3134	<tt>val</tt>. If <tt>idx</tt> exceeds the length of <tt>val</tt>, the
				3135	results are undefined.
				3136	</p>
				3137
				3138	<h5>Example:</h5>
				3139
				3140	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3141	%result = extractelement <4 x i32> %vec, i32 0 <i>; yields i32</i>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3142	</pre>
				3143	</div>
				3144
				3145
				3146	<!-- _______________________________________________________________________ -->
				3147	<div class="doc_subsubsection">
				3148	<a name="i_insertelement">'<tt>insertelement</tt>' Instruction</a>
				3149	</div>
				3150
				3151	<div class="doc_text">
				3152
				3153	<h5>Syntax:</h5>
				3154
				3155	<pre>
Dan Gohman	43ba067	2008-05-12 23:38:42 +0000	[diff] [blame]	3156	<result> = insertelement <n x <ty>> <val>, <ty> <elt>, i32 <idx> <i>; yields <n x <ty>></i>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3157	</pre>
				3158
				3159	<h5>Overview:</h5>
				3160
				3161	<p>
				3162	The '<tt>insertelement</tt>' instruction inserts a scalar
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	3163	element into a vector at a specified index.
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3164	</p>
				3165
				3166
				3167	<h5>Arguments:</h5>
				3168
				3169	<p>
				3170	The first operand of an '<tt>insertelement</tt>' instruction is a
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	3171	value of <a href="#t_vector">vector</a> type. The second operand is a
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3172	scalar value whose type must equal the element type of the first
				3173	operand. The third operand is an index indicating the position at
				3174	which to insert the value. The index may be a variable.</p>
				3175
				3176	<h5>Semantics:</h5>
				3177
				3178	<p>
Reid Spencer	404a325	2007-02-15 03:07:05 +0000	[diff] [blame]	3179	The result is a vector of the same type as <tt>val</tt>. Its
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3180	element values are those of <tt>val</tt> except at position
				3181	<tt>idx</tt>, where it gets the value <tt>elt</tt>. If <tt>idx</tt>
				3182	exceeds the length of <tt>val</tt>, the results are undefined.
				3183	</p>
				3184
				3185	<h5>Example:</h5>
				3186
				3187	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3188	%result = insertelement <4 x i32> %vec, i32 1, i32 0 <i>; yields <4 x i32></i>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3189	</pre>
				3190	</div>
				3191
				3192	<!-- _______________________________________________________________________ -->
				3193	<div class="doc_subsubsection">
				3194	<a name="i_shufflevector">'<tt>shufflevector</tt>' Instruction</a>
				3195	</div>
				3196
				3197	<div class="doc_text">
				3198
				3199	<h5>Syntax:</h5>
				3200
				3201	<pre>
Mon P Wang	25f0106	2008-11-10 04:46:22 +0000	[diff] [blame]	3202	<result> = shufflevector <n x <ty>> <v1>, <n x <ty>> <v2>, <m x i32> <mask> <i>; yields <m x <ty>></i>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3203	</pre>
				3204
				3205	<h5>Overview:</h5>
				3206
				3207	<p>
				3208	The '<tt>shufflevector</tt>' instruction constructs a permutation of elements
Mon P Wang	25f0106	2008-11-10 04:46:22 +0000	[diff] [blame]	3209	from two input vectors, returning a vector with the same element type as
				3210	the input and length that is the same as the shuffle mask.
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3211	</p>
				3212
				3213	<h5>Arguments:</h5>
				3214
				3215	<p>
Mon P Wang	25f0106	2008-11-10 04:46:22 +0000	[diff] [blame]	3216	The first two operands of a '<tt>shufflevector</tt>' instruction are vectors
				3217	with types that match each other. The third argument is a shuffle mask whose
				3218	element type is always 'i32'. The result of the instruction is a vector whose
				3219	length is the same as the shuffle mask and whose element type is the same as
				3220	the element type of the first two operands.
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3221	</p>
				3222
				3223	<p>
				3224	The shuffle mask operand is required to be a constant vector with either
				3225	constant integer or undef values.
				3226	</p>
				3227
				3228	<h5>Semantics:</h5>
				3229
				3230	<p>
				3231	The elements of the two input vectors are numbered from left to right across
				3232	both of the vectors. The shuffle mask operand specifies, for each element of
Mon P Wang	25f0106	2008-11-10 04:46:22 +0000	[diff] [blame]	3233	the result vector, which element of the two input vectors the result element
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3234	gets. The element selector may be undef (meaning "don't care") and the second
				3235	operand may be undef if performing a shuffle from only one vector.
				3236	</p>
				3237
				3238	<h5>Example:</h5>
				3239
				3240	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3241	%result = shufflevector <4 x i32> %v1, <4 x i32> %v2,
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	3242	<4 x i32> <i32 0, i32 4, i32 1, i32 5> <i>; yields <4 x i32></i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3243	%result = shufflevector <4 x i32> %v1, <4 x i32> undef,
				3244	<4 x i32> <i32 0, i32 1, i32 2, i32 3> <i>; yields <4 x i32></i> - Identity shuffle.
Mon P Wang	25f0106	2008-11-10 04:46:22 +0000	[diff] [blame]	3245	%result = shufflevector <8 x i32> %v1, <8 x i32> undef,
				3246	<4 x i32> <i32 0, i32 1, i32 2, i32 3> <i>; yields <4 x i32></i>
				3247	%result = shufflevector <4 x i32> %v1, <4 x i32> %v2,
				3248	<8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7 > <i>; yields <8 x i32></i>
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3249	</pre>
				3250	</div>
				3251
Tanya Lattner	b138bbe	2006-04-14 19:24:33 +0000	[diff] [blame]	3252
Chris Lattner	ce83bff	2006-04-08 23:07:04 +0000	[diff] [blame]	3253	<!-- ======================================================================= -->
				3254	<div class="doc_subsection">
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3255	<a name="aggregateops">Aggregate Operations</a>
				3256	</div>
				3257
				3258	<div class="doc_text">
				3259
				3260	<p>LLVM supports several instructions for working with aggregate values.
				3261	</p>
				3262
				3263	</div>
				3264
				3265	<!-- _______________________________________________________________________ -->
				3266	<div class="doc_subsubsection">
				3267	<a name="i_extractvalue">'<tt>extractvalue</tt>' Instruction</a>
				3268	</div>
				3269
				3270	<div class="doc_text">
				3271
				3272	<h5>Syntax:</h5>
				3273
				3274	<pre>
				3275	<result> = extractvalue <aggregate type> <val>, <idx>{, <idx>}*
				3276	</pre>
				3277
				3278	<h5>Overview:</h5>
				3279
				3280	<p>
Dan Gohman	35a835c	2008-05-13 18:16:06 +0000	[diff] [blame]	3281	The '<tt>extractvalue</tt>' instruction extracts the value of a struct field
				3282	or array element from an aggregate value.
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3283	</p>
				3284
				3285
				3286	<h5>Arguments:</h5>
				3287
				3288	<p>
				3289	The first operand of an '<tt>extractvalue</tt>' instruction is a
				3290	value of <a href="#t_struct">struct</a> or <a href="#t_array">array</a>
Dan Gohman	35a835c	2008-05-13 18:16:06 +0000	[diff] [blame]	3291	type. The operands are constant indices to specify which value to extract
Dan Gohman	1ecaf45	2008-05-31 00:58:22 +0000	[diff] [blame]	3292	in a similar manner as indices in a
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3293	'<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.
				3294	</p>
				3295
				3296	<h5>Semantics:</h5>
				3297
				3298	<p>
				3299	The result is the value at the position in the aggregate specified by
				3300	the index operands.
				3301	</p>
				3302
				3303	<h5>Example:</h5>
				3304
				3305	<pre>
Dan Gohman	1ecaf45	2008-05-31 00:58:22 +0000	[diff] [blame]	3306	%result = extractvalue {i32, float} %agg, 0 <i>; yields i32</i>
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3307	</pre>
				3308	</div>
				3309
				3310
				3311	<!-- _______________________________________________________________________ -->
				3312	<div class="doc_subsubsection">
				3313	<a name="i_insertvalue">'<tt>insertvalue</tt>' Instruction</a>
				3314	</div>
				3315
				3316	<div class="doc_text">
				3317
				3318	<h5>Syntax:</h5>
				3319
				3320	<pre>
Dan Gohman	1ecaf45	2008-05-31 00:58:22 +0000	[diff] [blame]	3321	<result> = insertvalue <aggregate type> <val>, <ty> <val>, <idx> <i>; yields <n x <ty>></i>
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3322	</pre>
				3323
				3324	<h5>Overview:</h5>
				3325
				3326	<p>
				3327	The '<tt>insertvalue</tt>' instruction inserts a value
Dan Gohman	35a835c	2008-05-13 18:16:06 +0000	[diff] [blame]	3328	into a struct field or array element in an aggregate.
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3329	</p>
				3330
				3331
				3332	<h5>Arguments:</h5>
				3333
				3334	<p>
				3335	The first operand of an '<tt>insertvalue</tt>' instruction is a
				3336	value of <a href="#t_struct">struct</a> or <a href="#t_array">array</a> type.
				3337	The second operand is a first-class value to insert.
Dan Gohman	34d1c0d	2008-05-23 21:53:15 +0000	[diff] [blame]	3338	The following operands are constant indices
Dan Gohman	1ecaf45	2008-05-31 00:58:22 +0000	[diff] [blame]	3339	indicating the position at which to insert the value in a similar manner as
Dan Gohman	35a835c	2008-05-13 18:16:06 +0000	[diff] [blame]	3340	indices in a
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3341	'<tt><a href="#i_getelementptr">getelementptr</a></tt>' instruction.
				3342	The value to insert must have the same type as the value identified
Dan Gohman	35a835c	2008-05-13 18:16:06 +0000	[diff] [blame]	3343	by the indices.
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	3344	</p>
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3345
				3346	<h5>Semantics:</h5>
				3347
				3348	<p>
				3349	The result is an aggregate of the same type as <tt>val</tt>. Its
				3350	value is that of <tt>val</tt> except that the value at the position
Dan Gohman	35a835c	2008-05-13 18:16:06 +0000	[diff] [blame]	3351	specified by the indices is that of <tt>elt</tt>.
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3352	</p>
				3353
				3354	<h5>Example:</h5>
				3355
				3356	<pre>
Dan Gohman	88ce1a5	2008-06-23 15:26:37 +0000	[diff] [blame]	3357	%result = insertvalue {i32, float} %agg, i32 1, 0 <i>; yields {i32, float}</i>
Dan Gohman	b9d6660	2008-05-12 23:51:09 +0000	[diff] [blame]	3358	</pre>
				3359	</div>
				3360
				3361
				3362	<!-- ======================================================================= -->
				3363	<div class="doc_subsection">
Chris Lattner	6ab6672	2006-08-15 00:45:58 +0000	[diff] [blame]	3364	<a name="memoryops">Memory Access and Addressing Operations</a>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3365	</div>
				3366
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3367	<div class="doc_text">
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3368
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3369	<p>A key design point of an SSA-based representation is how it
				3370	represents memory. In LLVM, no memory locations are in SSA form, which
				3371	makes things very simple. This section describes how to read, write,
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	3372	allocate, and free memory in LLVM.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3373
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3374	</div>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3375
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3376	<!-- _______________________________________________________________________ -->
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3377	<div class="doc_subsubsection">
				3378	<a name="i_malloc">'<tt>malloc</tt>' Instruction</a>
				3379	</div>
				3380
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3381	<div class="doc_text">
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3382
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3383	<h5>Syntax:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3384
				3385	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3386	<result> = malloc <type>[, i32 <NumElements>][, align <alignment>] <i>; yields {type*}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3387	</pre>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3388
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3389	<h5>Overview:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3390
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3391	<p>The '<tt>malloc</tt>' instruction allocates memory from the system
Christopher Lamb	55c6d4f	2007-12-17 01:00:21 +0000	[diff] [blame]	3392	heap and returns a pointer to it. The object is always allocated in the generic
				3393	address space (address space zero).</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3394
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3395	<h5>Arguments:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3396
				3397	<p>The '<tt>malloc</tt>' instruction allocates
				3398	<tt>sizeof(<type>)*NumElements</tt>
John Criswell	a92e586	2004-02-24 16:13:56 +0000	[diff] [blame]	3399	bytes of memory from the operating system and returns a pointer of the
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3400	appropriate type to the program. If "NumElements" is specified, it is the
Gabor Greif	dd1fc98	2008-02-09 22:24:34 +0000	[diff] [blame]	3401	number of elements allocated, otherwise "NumElements" is defaulted to be one.
Chris Lattner	1f17cce	2008-04-02 00:38:26 +0000	[diff] [blame]	3402	If a constant alignment is specified, the value result of the allocation is guaranteed to
Gabor Greif	dd1fc98	2008-02-09 22:24:34 +0000	[diff] [blame]	3403	be aligned to at least that boundary. If not specified, or if zero, the target can
				3404	choose to align the allocation on any convenient boundary.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3405
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3406	<p>'<tt>type</tt>' must be a sized type.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3407
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3408	<h5>Semantics:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3409
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3410	<p>Memory is allocated using the system "<tt>malloc</tt>" function, and
Nick Lewycky	f5ffcbc	2008-11-24 03:41:24 +0000	[diff] [blame]	3411	a pointer is returned. The result of a zero byte allocation is undefined. The
Chris Lattner	fe8519c	2008-04-19 21:01:16 +0000	[diff] [blame]	3412	result is null if there is insufficient memory available.</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3413
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3414	<h5>Example:</h5>
				3415
				3416	<pre>
Dan Gohman	7a5acb5	2009-01-04 23:49:44 +0000	[diff] [blame]	3417	%array = malloc [4 x i8] <i>; yields {[%4 x i8]*}:array</i>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3418
Bill Wendling	2d8b9a8	2007-05-29 09:42:13 +0000	[diff] [blame]	3419	%size = <a href="#i_add">add</a> i32 2, 2 <i>; yields {i32}:size = i32 4</i>
				3420	%array1 = malloc i8, i32 4 <i>; yields {i8*}:array1</i>
				3421	%array2 = malloc [12 x i8], i32 %size <i>; yields {[12 x i8]*}:array2</i>
				3422	%array3 = malloc i32, i32 4, align 1024 <i>; yields {i32*}:array3</i>
				3423	%array4 = malloc i32, align 1024 <i>; yields {i32*}:array4</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3424	</pre>
Dan Gohman	3065b61	2009-01-12 23:12:39 +0000	[diff] [blame]	3425
				3426	<p>Note that the code generator does not yet respect the
				3427	alignment value.</p>
				3428
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3429	</div>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3430
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3431	<!-- _______________________________________________________________________ -->
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3432	<div class="doc_subsubsection">
				3433	<a name="i_free">'<tt>free</tt>' Instruction</a>
				3434	</div>
				3435
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3436	<div class="doc_text">
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3437
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3438	<h5>Syntax:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3439
				3440	<pre>
Dan Gohman	7a5acb5	2009-01-04 23:49:44 +0000	[diff] [blame]	3441	free <type> <value> <i>; yields {void}</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3442	</pre>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3443
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3444	<h5>Overview:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3445
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3446	<p>The '<tt>free</tt>' instruction returns memory back to the unused
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	3447	memory heap to be reallocated in the future.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3448
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3449	<h5>Arguments:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3450
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3451	<p>'<tt>value</tt>' shall be a pointer value that points to a value
				3452	that was allocated with the '<tt><a href="#i_malloc">malloc</a></tt>'
				3453	instruction.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3454
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3455	<h5>Semantics:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3456
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	3457	<p>Access to the memory pointed to by the pointer is no longer defined
Chris Lattner	0f103e1	2008-04-19 22:41:32 +0000	[diff] [blame]	3458	after this instruction executes. If the pointer is null, the operation
				3459	is a noop.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3460
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3461	<h5>Example:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3462
				3463	<pre>
Dan Gohman	7a5acb5	2009-01-04 23:49:44 +0000	[diff] [blame]	3464	%array = <a href="#i_malloc">malloc</a> [4 x i8] <i>; yields {[4 x i8]*}:array</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3465	free [4 x i8]* %array
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3466	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3467	</div>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3468
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3469	<!-- _______________________________________________________________________ -->
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3470	<div class="doc_subsubsection">
				3471	<a name="i_alloca">'<tt>alloca</tt>' Instruction</a>
				3472	</div>
				3473
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3474	<div class="doc_text">
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3475
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3476	<h5>Syntax:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3477
				3478	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3479	<result> = alloca <type>[, i32 <NumElements>][, align <alignment>] <i>; yields {type*}:result</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3480	</pre>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3481
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3482	<h5>Overview:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3483
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	3484	<p>The '<tt>alloca</tt>' instruction allocates memory on the stack frame of the
				3485	currently executing function, to be automatically released when this function
Christopher Lamb	55c6d4f	2007-12-17 01:00:21 +0000	[diff] [blame]	3486	returns to its caller. The object is always allocated in the generic address
				3487	space (address space zero).</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3488
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3489	<h5>Arguments:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3490
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	3491	<p>The '<tt>alloca</tt>' instruction allocates <tt>sizeof(<type>)*NumElements</tt>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3492	bytes of memory on the runtime stack, returning a pointer of the
Gabor Greif	dd1fc98	2008-02-09 22:24:34 +0000	[diff] [blame]	3493	appropriate type to the program. If "NumElements" is specified, it is the
				3494	number of elements allocated, otherwise "NumElements" is defaulted to be one.
Chris Lattner	1f17cce	2008-04-02 00:38:26 +0000	[diff] [blame]	3495	If a constant alignment is specified, the value result of the allocation is guaranteed
Gabor Greif	dd1fc98	2008-02-09 22:24:34 +0000	[diff] [blame]	3496	to be aligned to at least that boundary. If not specified, or if zero, the target
				3497	can choose to align the allocation on any convenient boundary.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3498
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3499	<p>'<tt>type</tt>' may be any sized type.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3500
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3501	<h5>Semantics:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3502
Bill Wendling	9ee6a31	2009-05-08 20:49:29 +0000	[diff] [blame]	3503	<p>Memory is allocated; a pointer is returned. The operation is undefined if
Chris Lattner	fe8519c	2008-04-19 21:01:16 +0000	[diff] [blame]	3504	there is insufficient stack space for the allocation. '<tt>alloca</tt>'d
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3505	memory is automatically released when the function returns. The '<tt>alloca</tt>'
				3506	instruction is commonly used to represent automatic variables that must
				3507	have an address available. When the function returns (either with the <tt><a
John Criswell	c932bef	2005-05-12 16:55:34 +0000	[diff] [blame]	3508	href="#i_ret">ret</a></tt> or <tt><a href="#i_unwind">unwind</a></tt>
Chris Lattner	1f17cce	2008-04-02 00:38:26 +0000	[diff] [blame]	3509	instructions), the memory is reclaimed. Allocating zero bytes
				3510	is legal, but the result is undefined.</p>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3511
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3512	<h5>Example:</h5>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3513
				3514	<pre>
Dan Gohman	7a5acb5	2009-01-04 23:49:44 +0000	[diff] [blame]	3515	%ptr = alloca i32 <i>; yields {i32*}:ptr</i>
				3516	%ptr = alloca i32, i32 4 <i>; yields {i32*}:ptr</i>
				3517	%ptr = alloca i32, i32 4, align 1024 <i>; yields {i32*}:ptr</i>
				3518	%ptr = alloca i32, align 1024 <i>; yields {i32*}:ptr</i>
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3519	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3520	</div>
Chris Lattner	54611b4	2005-11-06 08:02:57 +0000	[diff] [blame]	3521
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3522	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3523	<div class="doc_subsubsection"> <a name="i_load">'<tt>load</tt>'
				3524	Instruction</a> </div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3525	<div class="doc_text">
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3526	<h5>Syntax:</h5>
Christopher Lamb	bff5020	2007-04-21 08:16:25 +0000	[diff] [blame]	3527	<pre> <result> = load <ty>* <pointer>[, align <alignment>]<br> <result> = volatile load <ty>* <pointer>[, align <alignment>]<br></pre>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3528	<h5>Overview:</h5>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3529	<p>The '<tt>load</tt>' instruction is used to read from memory.</p>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3530	<h5>Arguments:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3531	<p>The argument to the '<tt>load</tt>' instruction specifies the memory
John Criswell	4c0cf7f	2005-10-24 16:17:18 +0000	[diff] [blame]	3532	address from which to load. The pointer must point to a <a
Chris Lattner	10ee965	2004-06-03 22:57:15 +0000	[diff] [blame]	3533	href="#t_firstclass">first class</a> type. If the <tt>load</tt> is
John Criswell	4c0cf7f	2005-10-24 16:17:18 +0000	[diff] [blame]	3534	marked as <tt>volatile</tt>, then the optimizer is not allowed to modify
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3535	the number or order of execution of this <tt>load</tt> with other
				3536	volatile <tt>load</tt> and <tt><a href="#i_store">store</a></tt>
				3537	instructions. </p>
Chris Lattner	2a1993f	2008-01-06 21:04:43 +0000	[diff] [blame]	3538	<p>
Chris Lattner	1f17cce	2008-04-02 00:38:26 +0000	[diff] [blame]	3539	The optional constant "align" argument specifies the alignment of the operation
Chris Lattner	2a1993f	2008-01-06 21:04:43 +0000	[diff] [blame]	3540	(that is, the alignment of the memory address). A value of 0 or an
				3541	omitted "align" argument means that the operation has the preferential
				3542	alignment for the target. It is the responsibility of the code emitter
				3543	to ensure that the alignment information is correct. Overestimating
				3544	the alignment results in an undefined behavior. Underestimating the
				3545	alignment may produce less efficient code. An alignment of 1 is always
				3546	safe.
				3547	</p>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3548	<h5>Semantics:</h5>
Duncan Sands	b1656c1	2009-03-22 11:33:16 +0000	[diff] [blame]	3549	<p>The location of memory pointed to is loaded. If the value being loaded
				3550	is of scalar type then the number of bytes read does not exceed the minimum
				3551	number of bytes needed to hold all bits of the type. For example, loading an
				3552	<tt>i24</tt> reads at most three bytes. When loading a value of a type like
				3553	<tt>i20</tt> with a size that is not an integral number of bytes, the result
				3554	is undefined if the value was not originally written using a store of the
				3555	same type.</p>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3556	<h5>Examples:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3557	<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3558	<a
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3559	href="#i_store">store</a> i32 3, i32* %ptr <i>; yields {void}</i>
				3560	%val = load i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3561	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3562	</div>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3563	<!-- _______________________________________________________________________ -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3564	<div class="doc_subsubsection"> <a name="i_store">'<tt>store</tt>'
				3565	Instruction</a> </div>
Reid Spencer	a89fb18	2006-11-09 21:18:01 +0000	[diff] [blame]	3566	<div class="doc_text">
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3567	<h5>Syntax:</h5>
Christopher Lamb	bff5020	2007-04-21 08:16:25 +0000	[diff] [blame]	3568	<pre> store <ty> <value>, <ty>* <pointer>[, align <alignment>] <i>; yields {void}</i>
				3569	volatile store <ty> <value>, <ty>* <pointer>[, align <alignment>] <i>; yields {void}</i>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3570	</pre>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3571	<h5>Overview:</h5>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3572	<p>The '<tt>store</tt>' instruction is used to write to memory.</p>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3573	<h5>Arguments:</h5>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3574	<p>There are two arguments to the '<tt>store</tt>' instruction: a value
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	3575	to store and an address at which to store it. The type of the '<tt><pointer></tt>'
Chris Lattner	1f17cce	2008-04-02 00:38:26 +0000	[diff] [blame]	3576	operand must be a pointer to the <a href="#t_firstclass">first class</a> type
				3577	of the '<tt><value></tt>'
John Criswell	4a3327e	2005-05-13 22:25:59 +0000	[diff] [blame]	3578	operand. If the <tt>store</tt> is marked as <tt>volatile</tt>, then the
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3579	optimizer is not allowed to modify the number or order of execution of
				3580	this <tt>store</tt> with other volatile <tt>load</tt> and <tt><a
				3581	href="#i_store">store</a></tt> instructions.</p>
Chris Lattner	2a1993f	2008-01-06 21:04:43 +0000	[diff] [blame]	3582	<p>
Chris Lattner	1f17cce	2008-04-02 00:38:26 +0000	[diff] [blame]	3583	The optional constant "align" argument specifies the alignment of the operation
Chris Lattner	2a1993f	2008-01-06 21:04:43 +0000	[diff] [blame]	3584	(that is, the alignment of the memory address). A value of 0 or an
				3585	omitted "align" argument means that the operation has the preferential
				3586	alignment for the target. It is the responsibility of the code emitter
				3587	to ensure that the alignment information is correct. Overestimating
				3588	the alignment results in an undefined behavior. Underestimating the
				3589	alignment may produce less efficient code. An alignment of 1 is always
				3590	safe.
				3591	</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3592	<h5>Semantics:</h5>
				3593	<p>The contents of memory are updated to contain '<tt><value></tt>'
Duncan Sands	b1656c1	2009-03-22 11:33:16 +0000	[diff] [blame]	3594	at the location specified by the '<tt><pointer></tt>' operand.
				3595	If '<tt><value></tt>' is of scalar type then the number of bytes
				3596	written does not exceed the minimum number of bytes needed to hold all
				3597	bits of the type. For example, storing an <tt>i24</tt> writes at most
				3598	three bytes. When writing a value of a type like <tt>i20</tt> with a
				3599	size that is not an integral number of bytes, it is unspecified what
				3600	happens to the extra bits that do not belong to the type, but they will
				3601	typically be overwritten.</p>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3602	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3603	<pre> %ptr = <a href="#i_alloca">alloca</a> i32 <i>; yields {i32*}:ptr</i>
Bill Wendling	8830ffe	2007-10-22 05:10:05 +0000	[diff] [blame]	3604	store i32 3, i32* %ptr <i>; yields {void}</i>
				3605	%val = <a href="#i_load">load</a> i32* %ptr <i>; yields {i32}:val = i32 3</i>
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3606	</pre>
Reid Spencer	443460a	2006-11-09 21:15:49 +0000	[diff] [blame]	3607	</div>
				3608
Chris Lattner	095735d	2002-05-06 03:03:22 +0000	[diff] [blame]	3609	<!-- _______________________________________________________________________ -->
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3610	<div class="doc_subsubsection">
				3611	<a name="i_getelementptr">'<tt>getelementptr</tt>' Instruction</a>
				3612	</div>
				3613
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3614	<div class="doc_text">
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	3615	<h5>Syntax:</h5>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3616	<pre>
Matthijs Kooijman	0e26827	2008-10-13 13:44:15 +0000	[diff] [blame]	3617	<result> = getelementptr <pty>* <ptrval>{, <ty> <idx>}*
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3618	</pre>
				3619
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	3620	<h5>Overview:</h5>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3621
				3622	<p>
				3623	The '<tt>getelementptr</tt>' instruction is used to get the address of a
Matthijs Kooijman	0e26827	2008-10-13 13:44:15 +0000	[diff] [blame]	3624	subelement of an aggregate data structure. It performs address calculation only
				3625	and does not access memory.</p>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3626
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	3627	<h5>Arguments:</h5>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3628
Matthijs Kooijman	0e26827	2008-10-13 13:44:15 +0000	[diff] [blame]	3629	<p>The first argument is always a pointer, and forms the basis of the
				3630	calculation. The remaining arguments are indices, that indicate which of the
				3631	elements of the aggregate object are indexed. The interpretation of each index
				3632	is dependent on the type being indexed into. The first index always indexes the
				3633	pointer value given as the first argument, the second index indexes a value of
				3634	the type pointed to (not necessarily the value directly pointed to, since the
				3635	first index can be non-zero), etc. The first type indexed into must be a pointer
				3636	value, subsequent types can be arrays, vectors and structs. Note that subsequent
				3637	types being indexed into can never be pointers, since that would require loading
				3638	the pointer before continuing calculation.</p>
				3639
				3640	<p>The type of each index argument depends on the type it is indexing into.
				3641	When indexing into a (packed) structure, only <tt>i32</tt> integer
				3642	<b>constants</b> are allowed. When indexing into an array, pointer or vector,
Sanjiv Gupta	1f8555a	2009-04-27 03:21:00 +0000	[diff] [blame]	3643	integers of any width are allowed (also non-constants).</p>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3644
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3645	<p>For example, let's consider a C code fragment and how it gets
				3646	compiled to LLVM:</p>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3647
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3648	<div class="doc_code">
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3649	<pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3650	struct RT {
				3651	char A;
Chris Lattner	a446f1b	2007-05-29 15:43:56 +0000	[diff] [blame]	3652	int B[10][20];
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3653	char C;
				3654	};
				3655	struct ST {
Chris Lattner	a446f1b	2007-05-29 15:43:56 +0000	[diff] [blame]	3656	int X;
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3657	double Y;
				3658	struct RT Z;
				3659	};
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3660
Chris Lattner	a446f1b	2007-05-29 15:43:56 +0000	[diff] [blame]	3661	int foo(struct ST s) {
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3662	return &s[1].Z.B[5][13];
				3663	}
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3664	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3665	</div>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3666
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3667	<p>The LLVM code generated by the GCC frontend is:</p>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3668
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3669	<div class="doc_code">
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3670	<pre>
Chris Lattner	bc08821	2009-01-11 20:53:49 +0000	[diff] [blame]	3671	%RT = <a href="#namedtypes">type</a> { i8 , [10 x [20 x i32]], i8 }
				3672	%ST = <a href="#namedtypes">type</a> { i32, double, %RT }
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3673
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3674	define i32* %foo(%ST* %s) {
				3675	entry:
				3676	%reg = getelementptr %ST* %s, i32 1, i32 2, i32 1, i32 5, i32 13
				3677	ret i32* %reg
				3678	}
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3679	</pre>
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	3680	</div>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3681
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	3682	<h5>Semantics:</h5>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3683
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3684	<p>In the example above, the first index is indexing into the '<tt>%ST*</tt>'
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3685	type, which is a pointer, yielding a '<tt>%ST</tt>' = '<tt>{ i32, double, %RT
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3686	}</tt>' type, a structure. The second index indexes into the third element of
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3687	the structure, yielding a '<tt>%RT</tt>' = '<tt>{ i8 , [10 x [20 x i32]],
				3688	i8 }</tt>' type, another structure. The third index indexes into the second
				3689	element of the structure, yielding a '<tt>[10 x [20 x i32]]</tt>' type, an
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3690	array. The two dimensions of the array are subscripted into, yielding an
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3691	'<tt>i32</tt>' type. The '<tt>getelementptr</tt>' instruction returns a pointer
				3692	to this element, thus computing a value of '<tt>i32*</tt>' type.</p>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3693
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	3694	<p>Note that it is perfectly legal to index partially through a
				3695	structure, returning a pointer to an inner element. Because of this,
				3696	the LLVM code for the given testcase is equivalent to:</p>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3697
				3698	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3699	define i32* %foo(%ST* %s) {
				3700	%t1 = getelementptr %ST* %s, i32 1 <i>; yields %ST*:%t1</i>
Jeff Cohen	5819f18	2007-04-22 01:17:39 +0000	[diff] [blame]	3701	%t2 = getelementptr %ST* %t1, i32 0, i32 2 <i>; yields %RT*:%t2</i>
				3702	%t3 = getelementptr %RT* %t2, i32 0, i32 1 <i>; yields [10 x [20 x i32]]*:%t3</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3703	%t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 <i>; yields [20 x i32]*:%t4</i>
				3704	%t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 <i>; yields i32*:%t5</i>
				3705	ret i32* %t5
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3706	}
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	3707	</pre>
Chris Lattner	c0ad71e	2005-06-24 17:22:57 +0000	[diff] [blame]	3708
Chris Lattner	dd28282	2009-03-09 20:55:18 +0000	[diff] [blame]	3709	<p>Note that it is undefined to access an array out of bounds: array
				3710	and pointer indexes must always be within the defined bounds of the
				3711	array type when accessed with an instruction that dereferences the
				3712	pointer (e.g. a load or store instruction). The one exception for
				3713	this rule is zero length arrays. These arrays are defined to be
				3714	accessible as variable length arrays, which requires access beyond the
				3715	zero'th element.</p>
Chris Lattner	c0ad71e	2005-06-24 17:22:57 +0000	[diff] [blame]	3716
Chris Lattner	6ab6672	2006-08-15 00:45:58 +0000	[diff] [blame]	3717	<p>The getelementptr instruction is often confusing. For some more insight
				3718	into how it works, see <a href="GetElementPtr.html">the getelementptr
				3719	FAQ</a>.</p>
				3720
Chris Lattner	590645f	2002-04-14 06:13:44 +0000	[diff] [blame]	3721	<h5>Example:</h5>
Chris Lattner	c0ad71e	2005-06-24 17:22:57 +0000	[diff] [blame]	3722
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3723	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3724	<i>; yields [12 x i8]*:aptr</i>
Matthijs Kooijman	0e26827	2008-10-13 13:44:15 +0000	[diff] [blame]	3725	%aptr = getelementptr {i32, [12 x i8]}* %saptr, i64 0, i32 1
				3726	<i>; yields i8*:vptr</i>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	3727	%vptr = getelementptr {i32, <2 x i8>}* %svptr, i64 0, i32 1, i32 1
Matthijs Kooijman	0e26827	2008-10-13 13:44:15 +0000	[diff] [blame]	3728	<i>; yields i8*:eptr</i>
				3729	%eptr = getelementptr [12 x i8]* %aptr, i64 0, i32 1
Sanjiv Gupta	0c155e6	2009-04-25 07:27:44 +0000	[diff] [blame]	3730	<i>; yields i32*:iptr</i>
Sanjiv Gupta	77abea0	2009-04-24 16:38:13 +0000	[diff] [blame]	3731	%iptr = getelementptr [10 x i32]* @arr, i16 0, i16 0
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3732	</pre>
Chris Lattner	33fd702	2004-04-05 01:30:49 +0000	[diff] [blame]	3733	</div>
Reid Spencer	443460a	2006-11-09 21:15:49 +0000	[diff] [blame]	3734
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	3735	<!-- ======================================================================= -->
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	3736	<div class="doc_subsection"> <a name="convertops">Conversion Operations</a>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3737	</div>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3738	<div class="doc_text">
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	3739	<p>The instructions in this category are the conversion instructions (casting)
				3740	which all take a single operand and a type. They perform various bit conversions
				3741	on the operand.</p>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	3742	</div>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	3743
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	3744	<!-- _______________________________________________________________________ -->
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	3745	<div class="doc_subsubsection">
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3746	<a name="i_trunc">'<tt>trunc .. to</tt>' Instruction</a>
				3747	</div>
				3748	<div class="doc_text">
				3749
				3750	<h5>Syntax:</h5>
				3751	<pre>
				3752	<result> = trunc <ty> <value> to <ty2> <i>; yields ty2</i>
				3753	</pre>
				3754
				3755	<h5>Overview:</h5>
				3756	<p>
				3757	The '<tt>trunc</tt>' instruction truncates its operand to the type <tt>ty2</tt>.
				3758	</p>
				3759
				3760	<h5>Arguments:</h5>
				3761	<p>
				3762	The '<tt>trunc</tt>' instruction takes a <tt>value</tt> to trunc, which must
				3763	be an <a href="#t_integer">integer</a> type, and a type that specifies the size
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	3764	and type of the result, which must be an <a href="#t_integer">integer</a>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3765	type. The bit size of <tt>value</tt> must be larger than the bit size of
				3766	<tt>ty2</tt>. Equal sized types are not allowed.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3767
				3768	<h5>Semantics:</h5>
				3769	<p>
				3770	The '<tt>trunc</tt>' instruction truncates the high order bits in <tt>value</tt>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3771	and converts the remaining bits to <tt>ty2</tt>. Since the source size must be
				3772	larger than the destination size, <tt>trunc</tt> cannot be a <i>no-op cast</i>.
				3773	It will always truncate bits.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3774
				3775	<h5>Example:</h5>
				3776	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3777	%X = trunc i32 257 to i8 <i>; yields i8:1</i>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	3778	%Y = trunc i32 123 to i1 <i>; yields i1:true</i>
				3779	%Y = trunc i32 122 to i1 <i>; yields i1:false</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3780	</pre>
				3781	</div>
				3782
				3783	<!-- _______________________________________________________________________ -->
				3784	<div class="doc_subsubsection">
				3785	<a name="i_zext">'<tt>zext .. to</tt>' Instruction</a>
				3786	</div>
				3787	<div class="doc_text">
				3788
				3789	<h5>Syntax:</h5>
				3790	<pre>
				3791	<result> = zext <ty> <value> to <ty2> <i>; yields ty2</i>
				3792	</pre>
				3793
				3794	<h5>Overview:</h5>
				3795	<p>The '<tt>zext</tt>' instruction zero extends its operand to type
				3796	<tt>ty2</tt>.</p>
				3797
				3798
				3799	<h5>Arguments:</h5>
				3800	<p>The '<tt>zext</tt>' instruction takes a value to cast, which must be of
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	3801	<a href="#t_integer">integer</a> type, and a type to cast it to, which must
				3802	also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3803	<tt>value</tt> must be smaller than the bit size of the destination type,
				3804	<tt>ty2</tt>.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3805
				3806	<h5>Semantics:</h5>
				3807	<p>The <tt>zext</tt> fills the high order bits of the <tt>value</tt> with zero
Chris Lattner	c87f3df	2007-05-24 19:13:27 +0000	[diff] [blame]	3808	bits until it reaches the size of the destination type, <tt>ty2</tt>.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3809
Reid Spencer	07c9c68	2007-01-12 15:46:11 +0000	[diff] [blame]	3810	<p>When zero extending from i1, the result will always be either 0 or 1.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3811
				3812	<h5>Example:</h5>
				3813	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3814	%X = zext i32 257 to i64 <i>; yields i64:257</i>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	3815	%Y = zext i1 true to i32 <i>; yields i32:1</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3816	</pre>
				3817	</div>
				3818
				3819	<!-- _______________________________________________________________________ -->
				3820	<div class="doc_subsubsection">
				3821	<a name="i_sext">'<tt>sext .. to</tt>' Instruction</a>
				3822	</div>
				3823	<div class="doc_text">
				3824
				3825	<h5>Syntax:</h5>
				3826	<pre>
				3827	<result> = sext <ty> <value> to <ty2> <i>; yields ty2</i>
				3828	</pre>
				3829
				3830	<h5>Overview:</h5>
				3831	<p>The '<tt>sext</tt>' sign extends <tt>value</tt> to the type <tt>ty2</tt>.</p>
				3832
				3833	<h5>Arguments:</h5>
				3834	<p>
				3835	The '<tt>sext</tt>' instruction takes a value to cast, which must be of
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	3836	<a href="#t_integer">integer</a> type, and a type to cast it to, which must
				3837	also be of <a href="#t_integer">integer</a> type. The bit size of the
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3838	<tt>value</tt> must be smaller than the bit size of the destination type,
				3839	<tt>ty2</tt>.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3840
				3841	<h5>Semantics:</h5>
				3842	<p>
				3843	The '<tt>sext</tt>' instruction performs a sign extension by copying the sign
				3844	bit (highest order bit) of the <tt>value</tt> until it reaches the bit size of
Chris Lattner	c87f3df	2007-05-24 19:13:27 +0000	[diff] [blame]	3845	the type <tt>ty2</tt>.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3846
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	3847	<p>When sign extending from i1, the extension always results in -1 or 0.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3848
				3849	<h5>Example:</h5>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3850	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3851	%X = sext i8 -1 to i16 <i>; yields i16 :65535</i>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	3852	%Y = sext i1 true to i32 <i>; yields i32:-1</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3853	</pre>
				3854	</div>
				3855
				3856	<!-- _______________________________________________________________________ -->
				3857	<div class="doc_subsubsection">
Reid Spencer	2e2740d	2006-11-09 21:48:10 +0000	[diff] [blame]	3858	<a name="i_fptrunc">'<tt>fptrunc .. to</tt>' Instruction</a>
				3859	</div>
				3860
				3861	<div class="doc_text">
				3862
				3863	<h5>Syntax:</h5>
				3864
				3865	<pre>
				3866	<result> = fptrunc <ty> <value> to <ty2> <i>; yields ty2</i>
				3867	</pre>
				3868
				3869	<h5>Overview:</h5>
				3870	<p>The '<tt>fptrunc</tt>' instruction truncates <tt>value</tt> to type
				3871	<tt>ty2</tt>.</p>
				3872
				3873
				3874	<h5>Arguments:</h5>
				3875	<p>The '<tt>fptrunc</tt>' instruction takes a <a href="#t_floating">floating
				3876	point</a> value to cast and a <a href="#t_floating">floating point</a> type to
				3877	cast it to. The size of <tt>value</tt> must be larger than the size of
				3878	<tt>ty2</tt>. This implies that <tt>fptrunc</tt> cannot be used to make a
				3879	<i>no-op cast</i>.</p>
				3880
				3881	<h5>Semantics:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3882	<p> The '<tt>fptrunc</tt>' instruction truncates a <tt>value</tt> from a larger
				3883	<a href="#t_floating">floating point</a> type to a smaller
				3884	<a href="#t_floating">floating point</a> type. If the value cannot fit within
				3885	the destination type, <tt>ty2</tt>, then the results are undefined.</p>
Reid Spencer	2e2740d	2006-11-09 21:48:10 +0000	[diff] [blame]	3886
				3887	<h5>Example:</h5>
				3888	<pre>
				3889	%X = fptrunc double 123.0 to float <i>; yields float:123.0</i>
				3890	%Y = fptrunc double 1.0E+300 to float <i>; yields undefined</i>
				3891	</pre>
				3892	</div>
				3893
				3894	<!-- _______________________________________________________________________ -->
				3895	<div class="doc_subsubsection">
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3896	<a name="i_fpext">'<tt>fpext .. to</tt>' Instruction</a>
				3897	</div>
				3898	<div class="doc_text">
				3899
				3900	<h5>Syntax:</h5>
				3901	<pre>
				3902	<result> = fpext <ty> <value> to <ty2> <i>; yields ty2</i>
				3903	</pre>
				3904
				3905	<h5>Overview:</h5>
				3906	<p>The '<tt>fpext</tt>' extends a floating point <tt>value</tt> to a larger
				3907	floating point value.</p>
				3908
				3909	<h5>Arguments:</h5>
				3910	<p>The '<tt>fpext</tt>' instruction takes a
				3911	<a href="#t_floating">floating point</a> <tt>value</tt> to cast,
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3912	and a <a href="#t_floating">floating point</a> type to cast it to. The source
				3913	type must be smaller than the destination type.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3914
				3915	<h5>Semantics:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3916	<p>The '<tt>fpext</tt>' instruction extends the <tt>value</tt> from a smaller
Duncan Sands	16f122e	2007-03-30 12:22:09 +0000	[diff] [blame]	3917	<a href="#t_floating">floating point</a> type to a larger
				3918	<a href="#t_floating">floating point</a> type. The <tt>fpext</tt> cannot be
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3919	used to make a <i>no-op cast</i> because it always changes bits. Use
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	3920	<tt>bitcast</tt> to make a <i>no-op cast</i> for a floating point cast.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3921
				3922	<h5>Example:</h5>
				3923	<pre>
				3924	%X = fpext float 3.1415 to double <i>; yields double:3.1415</i>
				3925	%Y = fpext float 1.0 to float <i>; yields float:1.0 (no-op)</i>
				3926	</pre>
				3927	</div>
				3928
				3929	<!-- _______________________________________________________________________ -->
				3930	<div class="doc_subsubsection">
Reid Spencer	2eadb53	2007-01-21 00:29:26 +0000	[diff] [blame]	3931	<a name="i_fptoui">'<tt>fptoui .. to</tt>' Instruction</a>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3932	</div>
				3933	<div class="doc_text">
				3934
				3935	<h5>Syntax:</h5>
				3936	<pre>
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	3937	<result> = fptoui <ty> <value> to <ty2> <i>; yields ty2</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3938	</pre>
				3939
				3940	<h5>Overview:</h5>
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	3941	<p>The '<tt>fptoui</tt>' converts a floating point <tt>value</tt> to its
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3942	unsigned integer equivalent of type <tt>ty2</tt>.
				3943	</p>
				3944
				3945	<h5>Arguments:</h5>
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	3946	<p>The '<tt>fptoui</tt>' instruction takes a value to cast, which must be a
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	3947	scalar or vector <a href="#t_floating">floating point</a> value, and a type
				3948	to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
				3949	type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
				3950	vector integer type with the same number of elements as <tt>ty</tt></p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3951
				3952	<h5>Semantics:</h5>
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	3953	<p> The '<tt>fptoui</tt>' instruction converts its
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3954	<a href="#t_floating">floating point</a> operand into the nearest (rounding
				3955	towards zero) unsigned integer value. If the value cannot fit in <tt>ty2</tt>,
				3956	the results are undefined.</p>
				3957
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3958	<h5>Example:</h5>
				3959	<pre>
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	3960	%X = fptoui double 123.0 to i32 <i>; yields i32:123</i>
Chris Lattner	5b95a17	2007-09-22 03:17:52 +0000	[diff] [blame]	3961	%Y = fptoui float 1.0E+300 to i1 <i>; yields undefined:1</i>
Reid Spencer	753163d	2007-07-31 14:40:14 +0000	[diff] [blame]	3962	%X = fptoui float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3963	</pre>
				3964	</div>
				3965
				3966	<!-- _______________________________________________________________________ -->
				3967	<div class="doc_subsubsection">
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3968	<a name="i_fptosi">'<tt>fptosi .. to</tt>' Instruction</a>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3969	</div>
				3970	<div class="doc_text">
				3971
				3972	<h5>Syntax:</h5>
				3973	<pre>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3974	<result> = fptosi <ty> <value> to <ty2> <i>; yields ty2</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3975	</pre>
				3976
				3977	<h5>Overview:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3978	<p>The '<tt>fptosi</tt>' instruction converts
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3979	<a href="#t_floating">floating point</a> <tt>value</tt> to type <tt>ty2</tt>.
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	3980	</p>
				3981
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	3982	<h5>Arguments:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3983	<p> The '<tt>fptosi</tt>' instruction takes a value to cast, which must be a
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	3984	scalar or vector <a href="#t_floating">floating point</a> value, and a type
				3985	to cast it to <tt>ty2</tt>, which must be an <a href="#t_integer">integer</a>
				3986	type. If <tt>ty</tt> is a vector floating point type, <tt>ty2</tt> must be a
				3987	vector integer type with the same number of elements as <tt>ty</tt></p>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	3988
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	3989	<h5>Semantics:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	3990	<p>The '<tt>fptosi</tt>' instruction converts its
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	3991	<a href="#t_floating">floating point</a> operand into the nearest (rounding
				3992	towards zero) signed integer value. If the value cannot fit in <tt>ty2</tt>,
				3993	the results are undefined.</p>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	3994
Chris Lattner	70de663	2001-07-09 00:26:23 +0000	[diff] [blame]	3995	<h5>Example:</h5>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	3996	<pre>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	3997	%X = fptosi double -123.0 to i32 <i>; yields i32:-123</i>
Chris Lattner	5b95a17	2007-09-22 03:17:52 +0000	[diff] [blame]	3998	%Y = fptosi float 1.0E-247 to i1 <i>; yields undefined:1</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	3999	%X = fptosi float 1.04E+17 to i8 <i>; yields undefined:1</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4000	</pre>
				4001	</div>
				4002
				4003	<!-- _______________________________________________________________________ -->
				4004	<div class="doc_subsubsection">
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4005	<a name="i_uitofp">'<tt>uitofp .. to</tt>' Instruction</a>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4006	</div>
				4007	<div class="doc_text">
				4008
				4009	<h5>Syntax:</h5>
				4010	<pre>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4011	<result> = uitofp <ty> <value> to <ty2> <i>; yields ty2</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4012	</pre>
				4013
				4014	<h5>Overview:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4015	<p>The '<tt>uitofp</tt>' instruction regards <tt>value</tt> as an unsigned
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4016	integer and converts that value to the <tt>ty2</tt> type.</p>
				4017
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4018	<h5>Arguments:</h5>
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	4019	<p>The '<tt>uitofp</tt>' instruction takes a value to cast, which must be a
				4020	scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
				4021	to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
				4022	type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
				4023	floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4024
				4025	<h5>Semantics:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4026	<p>The '<tt>uitofp</tt>' instruction interprets its operand as an unsigned
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4027	integer quantity and converts it to the corresponding floating point value. If
Jeff Cohen	beccb74	2007-04-22 14:56:37 +0000	[diff] [blame]	4028	the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4029
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4030	<h5>Example:</h5>
				4031	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	4032	%X = uitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4033	%Y = uitofp i8 -1 to double <i>; yields double:255.0</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4034	</pre>
				4035	</div>
				4036
				4037	<!-- _______________________________________________________________________ -->
				4038	<div class="doc_subsubsection">
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4039	<a name="i_sitofp">'<tt>sitofp .. to</tt>' Instruction</a>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4040	</div>
				4041	<div class="doc_text">
				4042
				4043	<h5>Syntax:</h5>
				4044	<pre>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4045	<result> = sitofp <ty> <value> to <ty2> <i>; yields ty2</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4046	</pre>
				4047
				4048	<h5>Overview:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4049	<p>The '<tt>sitofp</tt>' instruction regards <tt>value</tt> as a signed
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4050	integer and converts that value to the <tt>ty2</tt> type.</p>
				4051
				4052	<h5>Arguments:</h5>
Nate Begeman	d4d45c2	2007-11-17 03:58:34 +0000	[diff] [blame]	4053	<p>The '<tt>sitofp</tt>' instruction takes a value to cast, which must be a
				4054	scalar or vector <a href="#t_integer">integer</a> value, and a type to cast it
				4055	to <tt>ty2</tt>, which must be an <a href="#t_floating">floating point</a>
				4056	type. If <tt>ty</tt> is a vector integer type, <tt>ty2</tt> must be a vector
				4057	floating point type with the same number of elements as <tt>ty</tt></p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4058
				4059	<h5>Semantics:</h5>
Reid Spencer	51b0725	2006-11-09 23:03:26 +0000	[diff] [blame]	4060	<p>The '<tt>sitofp</tt>' instruction interprets its operand as a signed
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4061	integer quantity and converts it to the corresponding floating point value. If
Jeff Cohen	beccb74	2007-04-22 14:56:37 +0000	[diff] [blame]	4062	the value cannot fit in the floating point value, the results are undefined.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4063
				4064	<h5>Example:</h5>
				4065	<pre>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	4066	%X = sitofp i32 257 to float <i>; yields float:257.0</i>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4067	%Y = sitofp i8 -1 to double <i>; yields double:-1.0</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4068	</pre>
				4069	</div>
				4070
				4071	<!-- _______________________________________________________________________ -->
				4072	<div class="doc_subsubsection">
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4073	<a name="i_ptrtoint">'<tt>ptrtoint .. to</tt>' Instruction</a>
				4074	</div>
				4075	<div class="doc_text">
				4076
				4077	<h5>Syntax:</h5>
				4078	<pre>
				4079	<result> = ptrtoint <ty> <value> to <ty2> <i>; yields ty2</i>
				4080	</pre>
				4081
				4082	<h5>Overview:</h5>
				4083	<p>The '<tt>ptrtoint</tt>' instruction converts the pointer <tt>value</tt> to
				4084	the integer type <tt>ty2</tt>.</p>
				4085
				4086	<h5>Arguments:</h5>
				4087	<p>The '<tt>ptrtoint</tt>' instruction takes a <tt>value</tt> to cast, which
Duncan Sands	16f122e	2007-03-30 12:22:09 +0000	[diff] [blame]	4088	must be a <a href="#t_pointer">pointer</a> value, and a type to cast it to
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4089	<tt>ty2</tt>, which must be an <a href="#t_integer">integer</a> type.</p>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4090
				4091	<h5>Semantics:</h5>
				4092	<p>The '<tt>ptrtoint</tt>' instruction converts <tt>value</tt> to integer type
				4093	<tt>ty2</tt> by interpreting the pointer value as an integer and either
				4094	truncating or zero extending that value to the size of the integer type. If
				4095	<tt>value</tt> is smaller than <tt>ty2</tt> then a zero extension is done. If
				4096	<tt>value</tt> is larger than <tt>ty2</tt> then a truncation is done. If they
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4097	are the same size, then nothing is done (<i>no-op cast</i>) other than a type
				4098	change.</p>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4099
				4100	<h5>Example:</h5>
				4101	<pre>
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4102	%X = ptrtoint i32* %X to i8 <i>; yields truncation on 32-bit architecture</i>
				4103	%Y = ptrtoint i32* %x to i64 <i>; yields zero extension on 32-bit architecture</i>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4104	</pre>
				4105	</div>
				4106
				4107	<!-- _______________________________________________________________________ -->
				4108	<div class="doc_subsubsection">
				4109	<a name="i_inttoptr">'<tt>inttoptr .. to</tt>' Instruction</a>
				4110	</div>
				4111	<div class="doc_text">
				4112
				4113	<h5>Syntax:</h5>
				4114	<pre>
				4115	<result> = inttoptr <ty> <value> to <ty2> <i>; yields ty2</i>
				4116	</pre>
				4117
				4118	<h5>Overview:</h5>
				4119	<p>The '<tt>inttoptr</tt>' instruction converts an integer <tt>value</tt> to
				4120	a pointer type, <tt>ty2</tt>.</p>
				4121
				4122	<h5>Arguments:</h5>
Duncan Sands	16f122e	2007-03-30 12:22:09 +0000	[diff] [blame]	4123	<p>The '<tt>inttoptr</tt>' instruction takes an <a href="#t_integer">integer</a>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4124	value to cast, and a type to cast it to, which must be a
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4125	<a href="#t_pointer">pointer</a> type.</p>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4126
				4127	<h5>Semantics:</h5>
				4128	<p>The '<tt>inttoptr</tt>' instruction converts <tt>value</tt> to type
				4129	<tt>ty2</tt> by applying either a zero extension or a truncation depending on
				4130	the size of the integer <tt>value</tt>. If <tt>value</tt> is larger than the
				4131	size of a pointer then a truncation is done. If <tt>value</tt> is smaller than
				4132	the size of a pointer then a zero extension is done. If they are the same size,
				4133	nothing is done (<i>no-op cast</i>).</p>
				4134
				4135	<h5>Example:</h5>
				4136	<pre>
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4137	%X = inttoptr i32 255 to i32* <i>; yields zero extension on 64-bit architecture</i>
				4138	%X = inttoptr i32 255 to i32* <i>; yields no-op on 32-bit architecture</i>
				4139	%Y = inttoptr i64 0 to i32* <i>; yields truncation on 32-bit architecture</i>
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4140	</pre>
				4141	</div>
				4142
				4143	<!-- _______________________________________________________________________ -->
				4144	<div class="doc_subsubsection">
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	4145	<a name="i_bitcast">'<tt>bitcast .. to</tt>' Instruction</a>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4146	</div>
				4147	<div class="doc_text">
				4148
				4149	<h5>Syntax:</h5>
				4150	<pre>
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	4151	<result> = bitcast <ty> <value> to <ty2> <i>; yields ty2</i>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4152	</pre>
				4153
				4154	<h5>Overview:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4155
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	4156	<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4157	<tt>ty2</tt> without changing any bits.</p>
				4158
				4159	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4160
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	4161	<p>The '<tt>bitcast</tt>' instruction takes a value to cast, which must be
Dan Gohman	c05dca9	2008-09-08 16:45:59 +0000	[diff] [blame]	4162	a non-aggregate first class value, and a type to cast it to, which must also be
				4163	a non-aggregate <a href="#t_firstclass">first class</a> type. The bit sizes of
				4164	<tt>value</tt>
Reid Spencer	e3db84c	2007-01-09 20:08:58 +0000	[diff] [blame]	4165	and the destination type, <tt>ty2</tt>, must be identical. If the source
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4166	type is a pointer, the destination type must also be a pointer. This
				4167	instruction supports bitwise conversion of vectors to integers and to vectors
				4168	of other types (as long as they have the same size).</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4169
				4170	<h5>Semantics:</h5>
Reid Spencer	5b95064	2006-11-11 23:08:07 +0000	[diff] [blame]	4171	<p>The '<tt>bitcast</tt>' instruction converts <tt>value</tt> to type
Reid Spencer	b7344ff	2006-11-11 21:00:47 +0000	[diff] [blame]	4172	<tt>ty2</tt>. It is always a <i>no-op cast</i> because no bits change with
				4173	this conversion. The conversion is done as if the <tt>value</tt> had been
				4174	stored to memory and read back as type <tt>ty2</tt>. Pointer types may only be
				4175	converted to other pointer types with this instruction. To convert pointers to
				4176	other types, use the <a href="#i_inttoptr">inttoptr</a> or
				4177	<a href="#i_ptrtoint">ptrtoint</a> instructions first.</p>
Reid Spencer	59b6b7d	2006-11-08 01:11:31 +0000	[diff] [blame]	4178
				4179	<h5>Example:</h5>
				4180	<pre>
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4181	%X = bitcast i8 255 to i8 <i>; yields i8 :-1</i>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	4182	%Y = bitcast i32* %x to sint* <i>; yields sint*:%x</i>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4183	%Z = bitcast <2 x int> %V to i64; <i>; yields i64: %V</i>
Chris Lattner	70de663	2001-07-09 00:26:23 +0000	[diff] [blame]	4184	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4185	</div>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4186
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4187	<!-- ======================================================================= -->
				4188	<div class="doc_subsection"> <a name="otherops">Other Operations</a> </div>
				4189	<div class="doc_text">
				4190	<p>The instructions in this category are the "miscellaneous"
				4191	instructions, which defy better classification.</p>
				4192	</div>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4193
				4194	<!-- _______________________________________________________________________ -->
				4195	<div class="doc_subsubsection"><a name="i_icmp">'<tt>icmp</tt>' Instruction</a>
				4196	</div>
				4197	<div class="doc_text">
				4198	<h5>Syntax:</h5>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4199	<pre> <result> = icmp <cond> <ty> <op1>, <op2> <i>; yields {i1} or {<N x i1>}:result</i>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4200	</pre>
				4201	<h5>Overview:</h5>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4202	<p>The '<tt>icmp</tt>' instruction returns a boolean value or
				4203	a vector of boolean values based on comparison
				4204	of its two integer, integer vector, or pointer operands.</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4205	<h5>Arguments:</h5>
				4206	<p>The '<tt>icmp</tt>' instruction takes three operands. The first operand is
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4207	the condition code indicating the kind of comparison to perform. It is not
				4208	a value, just a keyword. The possible condition code are:
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4209	</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4210	<ol>
				4211	<li><tt>eq</tt>: equal</li>
				4212	<li><tt>ne</tt>: not equal </li>
				4213	<li><tt>ugt</tt>: unsigned greater than</li>
				4214	<li><tt>uge</tt>: unsigned greater or equal</li>
				4215	<li><tt>ult</tt>: unsigned less than</li>
				4216	<li><tt>ule</tt>: unsigned less or equal</li>
				4217	<li><tt>sgt</tt>: signed greater than</li>
				4218	<li><tt>sge</tt>: signed greater or equal</li>
				4219	<li><tt>slt</tt>: signed less than</li>
				4220	<li><tt>sle</tt>: signed less or equal</li>
				4221	</ol>
Chris Lattner	c0f423a	2007-01-15 01:54:13 +0000	[diff] [blame]	4222	<p>The remaining two arguments must be <a href="#t_integer">integer</a> or
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4223	<a href="#t_pointer">pointer</a>
				4224	or integer <a href="#t_vector">vector</a> typed.
				4225	They must also be identical types.</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4226	<h5>Semantics:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4227	<p>The '<tt>icmp</tt>' compares <tt>op1</tt> and <tt>op2</tt> according to
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4228	the condition code given as <tt>cond</tt>. The comparison performed always
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4229	yields either an <a href="#t_primitive"><tt>i1</tt></a> or vector of <tt>i1</tt> result, as follows:
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4230	</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4231	<ol>
				4232	<li><tt>eq</tt>: yields <tt>true</tt> if the operands are equal,
				4233	<tt>false</tt> otherwise. No sign interpretation is necessary or performed.
				4234	</li>
				4235	<li><tt>ne</tt>: yields <tt>true</tt> if the operands are unequal,
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4236	<tt>false</tt> otherwise. No sign interpretation is necessary or performed.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4237	<li><tt>ugt</tt>: interprets the operands as unsigned values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4238	<tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4239	<li><tt>uge</tt>: interprets the operands as unsigned values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4240	<tt>true</tt> if <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4241	<li><tt>ult</tt>: interprets the operands as unsigned values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4242	<tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4243	<li><tt>ule</tt>: interprets the operands as unsigned values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4244	<tt>true</tt> if <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4245	<li><tt>sgt</tt>: interprets the operands as signed values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4246	<tt>true</tt> if <tt>op1</tt> is greater than <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4247	<li><tt>sge</tt>: interprets the operands as signed values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4248	<tt>true</tt> if <tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4249	<li><tt>slt</tt>: interprets the operands as signed values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4250	<tt>true</tt> if <tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4251	<li><tt>sle</tt>: interprets the operands as signed values and yields
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4252	<tt>true</tt> if <tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4253	</ol>
				4254	<p>If the operands are <a href="#t_pointer">pointer</a> typed, the pointer
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4255	values are compared as if they were integers.</p>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4256	<p>If the operands are integer vectors, then they are compared
				4257	element by element. The result is an <tt>i1</tt> vector with
				4258	the same number of elements as the values being compared.
				4259	Otherwise, the result is an <tt>i1</tt>.
				4260	</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4261
				4262	<h5>Example:</h5>
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	4263	<pre> <result> = icmp eq i32 4, 5 <i>; yields: result=false</i>
				4264	<result> = icmp ne float* %X, %X <i>; yields: result=false</i>
				4265	<result> = icmp ult i16 4, 5 <i>; yields: result=true</i>
				4266	<result> = icmp sgt i16 4, 5 <i>; yields: result=false</i>
				4267	<result> = icmp ule i16 -4, 5 <i>; yields: result=false</i>
				4268	<result> = icmp sge i16 4, 5 <i>; yields: result=false</i>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4269	</pre>
Dan Gohman	a5127ab	2009-01-22 01:39:38 +0000	[diff] [blame]	4270
				4271	<p>Note that the code generator does not yet support vector types with
				4272	the <tt>icmp</tt> instruction.</p>
				4273
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4274	</div>
				4275
				4276	<!-- _______________________________________________________________________ -->
				4277	<div class="doc_subsubsection"><a name="i_fcmp">'<tt>fcmp</tt>' Instruction</a>
				4278	</div>
				4279	<div class="doc_text">
				4280	<h5>Syntax:</h5>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4281	<pre> <result> = fcmp <cond> <ty> <op1>, <op2> <i>; yields {i1} or {<N x i1>}:result</i>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4282	</pre>
				4283	<h5>Overview:</h5>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4284	<p>The '<tt>fcmp</tt>' instruction returns a boolean value
				4285	or vector of boolean values based on comparison
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4286	of its operands.</p>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4287	<p>
				4288	If the operands are floating point scalars, then the result
				4289	type is a boolean (<a href="#t_primitive"><tt>i1</tt></a>).
				4290	</p>
				4291	<p>If the operands are floating point vectors, then the result type
				4292	is a vector of boolean with the same number of elements as the
				4293	operands being compared.</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4294	<h5>Arguments:</h5>
				4295	<p>The '<tt>fcmp</tt>' instruction takes three operands. The first operand is
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4296	the condition code indicating the kind of comparison to perform. It is not
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4297	a value, just a keyword. The possible condition code are:</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4298	<ol>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4299	<li><tt>false</tt>: no comparison, always returns false</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4300	<li><tt>oeq</tt>: ordered and equal</li>
				4301	<li><tt>ogt</tt>: ordered and greater than </li>
				4302	<li><tt>oge</tt>: ordered and greater than or equal</li>
				4303	<li><tt>olt</tt>: ordered and less than </li>
				4304	<li><tt>ole</tt>: ordered and less than or equal</li>
				4305	<li><tt>one</tt>: ordered and not equal</li>
				4306	<li><tt>ord</tt>: ordered (no nans)</li>
				4307	<li><tt>ueq</tt>: unordered or equal</li>
				4308	<li><tt>ugt</tt>: unordered or greater than </li>
				4309	<li><tt>uge</tt>: unordered or greater than or equal</li>
				4310	<li><tt>ult</tt>: unordered or less than </li>
				4311	<li><tt>ule</tt>: unordered or less than or equal</li>
				4312	<li><tt>une</tt>: unordered or not equal</li>
				4313	<li><tt>uno</tt>: unordered (either nans)</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4314	<li><tt>true</tt>: no comparison, always returns true</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4315	</ol>
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4316	<p><i>Ordered</i> means that neither operand is a QNAN while
Reid Spencer	02e0d1d	2006-12-06 07:08:07 +0000	[diff] [blame]	4317	<i>unordered</i> means that either operand may be a QNAN.</p>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4318	<p>Each of <tt>val1</tt> and <tt>val2</tt> arguments must be
				4319	either a <a href="#t_floating">floating point</a> type
				4320	or a <a href="#t_vector">vector</a> of floating point type.
				4321	They must have identical types.</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4322	<h5>Semantics:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4323	<p>The '<tt>fcmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4324	according to the condition code given as <tt>cond</tt>.
				4325	If the operands are vectors, then the vectors are compared
				4326	element by element.
				4327	Each comparison performed
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4328	always yields an <a href="#t_primitive">i1</a> result, as follows:</p>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4329	<ol>
				4330	<li><tt>false</tt>: always yields <tt>false</tt>, regardless of operands.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4331	<li><tt>oeq</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4332	<tt>op1</tt> is equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4333	<li><tt>ogt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4334	<tt>op1</tt> is greather than <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4335	<li><tt>oge</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4336	<tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4337	<li><tt>olt</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4338	<tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4339	<li><tt>ole</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4340	<tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4341	<li><tt>one</tt>: yields <tt>true</tt> if both operands are not a QNAN and
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4342	<tt>op1</tt> is not equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4343	<li><tt>ord</tt>: yields <tt>true</tt> if both operands are not a QNAN.</li>
				4344	<li><tt>ueq</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4345	<tt>op1</tt> is equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4346	<li><tt>ugt</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4347	<tt>op1</tt> is greater than <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4348	<li><tt>uge</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4349	<tt>op1</tt> is greater than or equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4350	<li><tt>ult</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4351	<tt>op1</tt> is less than <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4352	<li><tt>ule</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4353	<tt>op1</tt> is less than or equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4354	<li><tt>une</tt>: yields <tt>true</tt> if either operand is a QNAN or
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4355	<tt>op1</tt> is not equal to <tt>op2</tt>.</li>
Reid Spencer	f69acf3	2006-11-19 03:00:14 +0000	[diff] [blame]	4356	<li><tt>uno</tt>: yields <tt>true</tt> if either operand is a QNAN.</li>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4357	<li><tt>true</tt>: always yields <tt>true</tt>, regardless of operands.</li>
				4358	</ol>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4359
				4360	<h5>Example:</h5>
				4361	<pre> <result> = fcmp oeq float 4.0, 5.0 <i>; yields: result=false</i>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4362	<result> = fcmp one float 4.0, 5.0 <i>; yields: result=true</i>
				4363	<result> = fcmp olt float 4.0, 5.0 <i>; yields: result=true</i>
				4364	<result> = fcmp ueq double 1.0, 2.0 <i>; yields: result=false</i>
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4365	</pre>
Dan Gohman	a5127ab	2009-01-22 01:39:38 +0000	[diff] [blame]	4366
				4367	<p>Note that the code generator does not yet support vector types with
				4368	the <tt>fcmp</tt> instruction.</p>
				4369
Reid Spencer	c828a0e	2006-11-18 21:50:54 +0000	[diff] [blame]	4370	</div>
				4371
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4372	<!-- _______________________________________________________________________ -->
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4373	<div class="doc_subsubsection">
				4374	<a name="i_vicmp">'<tt>vicmp</tt>' Instruction</a>
				4375	</div>
				4376	<div class="doc_text">
				4377	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4378	<pre> <result> = vicmp <cond> <ty> <op1>, <op2> <i>; yields {ty}:result</i>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4379	</pre>
				4380	<h5>Overview:</h5>
				4381	<p>The '<tt>vicmp</tt>' instruction returns an integer vector value based on
				4382	element-wise comparison of its two integer vector operands.</p>
				4383	<h5>Arguments:</h5>
				4384	<p>The '<tt>vicmp</tt>' instruction takes three operands. The first operand is
				4385	the condition code indicating the kind of comparison to perform. It is not
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4386	a value, just a keyword. The possible condition code are:</p>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4387	<ol>
				4388	<li><tt>eq</tt>: equal</li>
				4389	<li><tt>ne</tt>: not equal </li>
				4390	<li><tt>ugt</tt>: unsigned greater than</li>
				4391	<li><tt>uge</tt>: unsigned greater or equal</li>
				4392	<li><tt>ult</tt>: unsigned less than</li>
				4393	<li><tt>ule</tt>: unsigned less or equal</li>
				4394	<li><tt>sgt</tt>: signed greater than</li>
				4395	<li><tt>sge</tt>: signed greater or equal</li>
				4396	<li><tt>slt</tt>: signed less than</li>
				4397	<li><tt>sle</tt>: signed less or equal</li>
				4398	</ol>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4399	<p>The remaining two arguments must be <a href="#t_vector">vector</a> or
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4400	<a href="#t_integer">integer</a> typed. They must also be identical types.</p>
				4401	<h5>Semantics:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4402	<p>The '<tt>vicmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4403	according to the condition code given as <tt>cond</tt>. The comparison yields a
				4404	<a href="#t_vector">vector</a> of <a href="#t_integer">integer</a> result, of
				4405	identical type as the values being compared. The most significant bit in each
				4406	element is 1 if the element-wise comparison evaluates to true, and is 0
				4407	otherwise. All other bits of the result are undefined. The condition codes
				4408	are evaluated identically to the <a href="#i_icmp">'<tt>icmp</tt>'
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4409	instruction</a>.</p>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4410
				4411	<h5>Example:</h5>
				4412	<pre>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4413	<result> = vicmp eq <2 x i32> < i32 4, i32 0>, < i32 5, i32 0> <i>; yields: result=<2 x i32> < i32 0, i32 -1 ></i>
				4414	<result> = vicmp ult <2 x i8 > < i8 1, i8 2>, < i8 2, i8 2 > <i>; yields: result=<2 x i8> < i8 -1, i8 0 ></i>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4415	</pre>
				4416	</div>
				4417
				4418	<!-- _______________________________________________________________________ -->
				4419	<div class="doc_subsubsection">
				4420	<a name="i_vfcmp">'<tt>vfcmp</tt>' Instruction</a>
				4421	</div>
				4422	<div class="doc_text">
				4423	<h5>Syntax:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4424	<pre> <result> = vfcmp <cond> <ty> <op1>, <op2></pre>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4425	<h5>Overview:</h5>
				4426	<p>The '<tt>vfcmp</tt>' instruction returns an integer vector value based on
				4427	element-wise comparison of its two floating point vector operands. The output
				4428	elements have the same width as the input elements.</p>
				4429	<h5>Arguments:</h5>
				4430	<p>The '<tt>vfcmp</tt>' instruction takes three operands. The first operand is
				4431	the condition code indicating the kind of comparison to perform. It is not
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4432	a value, just a keyword. The possible condition code are:</p>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4433	<ol>
				4434	<li><tt>false</tt>: no comparison, always returns false</li>
				4435	<li><tt>oeq</tt>: ordered and equal</li>
				4436	<li><tt>ogt</tt>: ordered and greater than </li>
				4437	<li><tt>oge</tt>: ordered and greater than or equal</li>
				4438	<li><tt>olt</tt>: ordered and less than </li>
				4439	<li><tt>ole</tt>: ordered and less than or equal</li>
				4440	<li><tt>one</tt>: ordered and not equal</li>
				4441	<li><tt>ord</tt>: ordered (no nans)</li>
				4442	<li><tt>ueq</tt>: unordered or equal</li>
				4443	<li><tt>ugt</tt>: unordered or greater than </li>
				4444	<li><tt>uge</tt>: unordered or greater than or equal</li>
				4445	<li><tt>ult</tt>: unordered or less than </li>
				4446	<li><tt>ule</tt>: unordered or less than or equal</li>
				4447	<li><tt>une</tt>: unordered or not equal</li>
				4448	<li><tt>uno</tt>: unordered (either nans)</li>
				4449	<li><tt>true</tt>: no comparison, always returns true</li>
				4450	</ol>
				4451	<p>The remaining two arguments must be <a href="#t_vector">vector</a> of
				4452	<a href="#t_floating">floating point</a> typed. They must also be identical
				4453	types.</p>
				4454	<h5>Semantics:</h5>
Gabor Greif	0f75ad0	2008-08-07 21:46:00 +0000	[diff] [blame]	4455	<p>The '<tt>vfcmp</tt>' instruction compares <tt>op1</tt> and <tt>op2</tt>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4456	according to the condition code given as <tt>cond</tt>. The comparison yields a
				4457	<a href="#t_vector">vector</a> of <a href="#t_integer">integer</a> result, with
				4458	an identical number of elements as the values being compared, and each element
				4459	having identical with to the width of the floating point elements. The most
				4460	significant bit in each element is 1 if the element-wise comparison evaluates to
				4461	true, and is 0 otherwise. All other bits of the result are undefined. The
				4462	condition codes are evaluated identically to the
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4463	<a href="#i_fcmp">'<tt>fcmp</tt>' instruction</a>.</p>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4464
				4465	<h5>Example:</h5>
				4466	<pre>
Chris Lattner	0ae0209	2008-10-13 16:55:18 +0000	[diff] [blame]	4467	<i>; yields: result=<2 x i32> < i32 0, i32 -1 ></i>
				4468	<result> = vfcmp oeq <2 x float> < float 4, float 0 >, < float 5, float 0 >
				4469
				4470	<i>; yields: result=<2 x i64> < i64 -1, i64 0 ></i>
				4471	<result> = vfcmp ult <2 x double> < double 1, double 2 >, < double 2, double 2>
Nate Begeman	d219570	2008-05-12 19:01:56 +0000	[diff] [blame]	4472	</pre>
				4473	</div>
				4474
				4475	<!-- _______________________________________________________________________ -->
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4476	<div class="doc_subsubsection">
				4477	<a name="i_phi">'<tt>phi</tt>' Instruction</a>
				4478	</div>
				4479
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4480	<div class="doc_text">
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4481
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4482	<h5>Syntax:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4483
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4484	<pre> <result> = phi <ty> [ <val0>, <label0>], ...<br></pre>
				4485	<h5>Overview:</h5>
				4486	<p>The '<tt>phi</tt>' instruction is used to implement the φ node in
				4487	the SSA graph representing the function.</p>
				4488	<h5>Arguments:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4489
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4490	<p>The type of the incoming values is specified with the first type
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4491	field. After this, the '<tt>phi</tt>' instruction takes a list of pairs
				4492	as arguments, with one pair for each predecessor basic block of the
				4493	current block. Only values of <a href="#t_firstclass">first class</a>
				4494	type may be used as the value arguments to the PHI node. Only labels
				4495	may be used as the label arguments.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4496
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4497	<p>There must be no non-phi instructions between the start of a basic
				4498	block and the PHI instructions: i.e. PHI instructions must be first in
				4499	a basic block.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4500
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4501	<h5>Semantics:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4502
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4503	<p>At runtime, the '<tt>phi</tt>' instruction logically takes on the value
				4504	specified by the pair corresponding to the predecessor basic block that executed
				4505	just prior to the current block.</p>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4506
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4507	<h5>Example:</h5>
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4508	<pre>
				4509	Loop: ; Infinite loop that counts from 0 on up...
				4510	%indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]
				4511	%nextindvar = add i32 %indvar, 1
				4512	br label %Loop
				4513	</pre>
Reid Spencer	97c5fa4	2006-11-08 01:18:52 +0000	[diff] [blame]	4514	</div>
				4515
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4516	<!-- _______________________________________________________________________ -->
				4517	<div class="doc_subsubsection">
				4518	<a name="i_select">'<tt>select</tt>' Instruction</a>
				4519	</div>
				4520
				4521	<div class="doc_text">
				4522
				4523	<h5>Syntax:</h5>
				4524
				4525	<pre>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4526	<result> = select <i>selty</i> <cond>, <ty> <val1>, <ty> <val2> <i>; yields ty</i>
				4527
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4528	<i>selty</i> is either i1 or {<N x i1>}
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4529	</pre>
				4530
				4531	<h5>Overview:</h5>
				4532
				4533	<p>
				4534	The '<tt>select</tt>' instruction is used to choose one value based on a
				4535	condition, without branching.
				4536	</p>
				4537
				4538
				4539	<h5>Arguments:</h5>
				4540
				4541	<p>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4542	The '<tt>select</tt>' instruction requires an 'i1' value or
				4543	a vector of 'i1' values indicating the
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4544	condition, and two values of the same <a href="#t_firstclass">first class</a>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4545	type. If the val1/val2 are vectors and
				4546	the condition is a scalar, then entire vectors are selected, not
Chris Lattner	b54c30f	2008-05-20 20:48:21 +0000	[diff] [blame]	4547	individual elements.
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4548	</p>
				4549
				4550	<h5>Semantics:</h5>
				4551
				4552	<p>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4553	If the condition is an i1 and it evaluates to 1, the instruction returns the first
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	4554	value argument; otherwise, it returns the second value argument.
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4555	</p>
Dan Gohman	c579d97	2008-09-09 01:02:47 +0000	[diff] [blame]	4556	<p>
				4557	If the condition is a vector of i1, then the value arguments must
				4558	be vectors of the same size, and the selection is done element
				4559	by element.
				4560	</p>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4561
				4562	<h5>Example:</h5>
				4563
				4564	<pre>
Reid Spencer	36a1542	2007-01-12 03:35:51 +0000	[diff] [blame]	4565	%X = select i1 true, i8 17, i8 42 <i>; yields i8:17</i>
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4566	</pre>
Dan Gohman	a5127ab	2009-01-22 01:39:38 +0000	[diff] [blame]	4567
				4568	<p>Note that the code generator does not yet support conditions
				4569	with vector type.</p>
				4570
Chris Lattner	b53c28d	2004-03-12 05:50:16 +0000	[diff] [blame]	4571	</div>
				4572
Robert Bocchino	f72fdfe	2006-01-15 20:48:27 +0000	[diff] [blame]	4573
				4574	<!-- _______________________________________________________________________ -->
				4575	<div class="doc_subsubsection">
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4576	<a name="i_call">'<tt>call</tt>' Instruction</a>
				4577	</div>
				4578
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4579	<div class="doc_text">
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4580
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	4581	<h5>Syntax:</h5>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4582	<pre>
Devang Patel	0225623	2008-10-07 17:48:33 +0000	[diff] [blame]	4583	<result> = [tail] call [<a href="#callingconv">cconv</a>] [<a href="#paramattrs">ret attrs</a>] <ty> [<fnty>*] <fnptrval>(<function args>) [<a href="#fnattrs">fn attrs</a>]
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4584	</pre>
				4585
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	4586	<h5>Overview:</h5>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4587
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4588	<p>The '<tt>call</tt>' instruction represents a simple function call.</p>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4589
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	4590	<h5>Arguments:</h5>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4591
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4592	<p>This instruction requires several arguments:</p>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4593
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	4594	<ol>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4595	<li>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	4596	<p>The optional "tail" marker indicates whether the callee function accesses
				4597	any allocas or varargs in the caller. If the "tail" marker is present, the
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4598	function call is eligible for tail call optimization. Note that calls may
				4599	be marked "tail" even if they do not occur before a <a
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4600	href="#i_ret"><tt>ret</tt></a> instruction.</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4601	</li>
				4602	<li>
Duncan Sands	16f122e	2007-03-30 12:22:09 +0000	[diff] [blame]	4603	<p>The optional "cconv" marker indicates which <a href="#callingconv">calling
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	4604	convention</a> the call should use. If none is specified, the call defaults
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4605	to using C calling conventions.</p>
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	4606	</li>
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	4607
				4608	<li>
				4609	<p>The optional <a href="#paramattrs">Parameter Attributes</a> list for
				4610	return values. Only '<tt>zeroext</tt>', '<tt>signext</tt>',
				4611	and '<tt>inreg</tt>' attributes are valid here.</p>
				4612	</li>
				4613
Chris Lattner	0132aff	2005-05-06 22:57:40 +0000	[diff] [blame]	4614	<li>
Nick Lewycky	a9b13d5	2007-09-08 13:57:50 +0000	[diff] [blame]	4615	<p>'<tt>ty</tt>': the type of the call instruction itself which is also
				4616	the type of the return value. Functions that return no value are marked
				4617	<tt><a href="#t_void">void</a></tt>.</p>
				4618	</li>
				4619	<li>
				4620	<p>'<tt>fnty</tt>': shall be the signature of the pointer to function
				4621	value being invoked. The argument types must match the types implied by
				4622	this signature. This type can be omitted if the function is not varargs
				4623	and if the function type does not return a pointer to a function.</p>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4624	</li>
				4625	<li>
				4626	<p>'<tt>fnptrval</tt>': An LLVM value containing a pointer to a function to
				4627	be invoked. In most cases, this is a direct function invocation, but
				4628	indirect <tt>call</tt>s are just as possible, calling an arbitrary pointer
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	4629	to function value.</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4630	</li>
				4631	<li>
				4632	<p>'<tt>function args</tt>': argument list whose types match the
Reid Spencer	d845d16	2005-05-01 22:22:57 +0000	[diff] [blame]	4633	function signature argument types. All arguments must be of
				4634	<a href="#t_firstclass">first class</a> type. If the function signature
				4635	indicates the function accepts a variable number of arguments, the extra
				4636	arguments can be specified.</p>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4637	</li>
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	4638	<li>
Devang Patel	0225623	2008-10-07 17:48:33 +0000	[diff] [blame]	4639	<p>The optional <a href="#fnattrs">function attributes</a> list. Only
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	4640	'<tt>noreturn</tt>', '<tt>nounwind</tt>', '<tt>readonly</tt>' and
				4641	'<tt>readnone</tt>' attributes are valid here.</p>
				4642	</li>
Chris Lattner	a8292f3	2002-05-06 22:08:29 +0000	[diff] [blame]	4643	</ol>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4644
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	4645	<h5>Semantics:</h5>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4646
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4647	<p>The '<tt>call</tt>' instruction is used to cause control flow to
				4648	transfer to a specified function, with its incoming arguments bound to
				4649	the specified values. Upon a '<tt><a href="#i_ret">ret</a></tt>'
				4650	instruction in the called function, control flow continues with the
				4651	instruction after the function call, and the return value of the
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4652	function is bound to the result argument.</p>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4653
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	4654	<h5>Example:</h5>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4655
				4656	<pre>
Nick Lewycky	a9b13d5	2007-09-08 13:57:50 +0000	[diff] [blame]	4657	%retval = call i32 @test(i32 %argc)
Chris Lattner	fb7c88d	2008-03-21 17:24:17 +0000	[diff] [blame]	4658	call i32 (i8 , ...) @printf(i8 * %msg, i32 12, i8 42) <i>; yields i32</i>
				4659	%X = tail call i32 @foo() <i>; yields i32</i>
				4660	%Y = tail call <a href="#callingconv">fastcc</a> i32 @foo() <i>; yields i32</i>
				4661	call void %foo(i8 97 signext)
Devang Patel	d6cff51	2008-03-10 20:49:15 +0000	[diff] [blame]	4662
				4663	%struct.A = type { i32, i8 }
Devang Patel	7e9b05e	2008-10-06 18:50:38 +0000	[diff] [blame]	4664	%r = call %struct.A @foo() <i>; yields { 32, i8 }</i>
Dan Gohman	cc3132e	2008-10-04 19:00:07 +0000	[diff] [blame]	4665	%gr = extractvalue %struct.A %r, 0 <i>; yields i32</i>
				4666	%gr1 = extractvalue %struct.A %r, 1 <i>; yields i8</i>
Chris Lattner	6cbe8e9	2008-10-08 06:26:11 +0000	[diff] [blame]	4667	%Z = call void @foo() noreturn <i>; indicates that %foo never returns normally</i>
Matthijs Kooijman	eefa7df	2008-10-07 10:03:45 +0000	[diff] [blame]	4668	%ZZ = call zeroext i32 @bar() <i>; Return value is %zero extended</i>
Chris Lattner	e23c139	2005-05-06 05:47:36 +0000	[diff] [blame]	4669	</pre>
				4670
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4671	</div>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4672
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4673	<!-- _______________________________________________________________________ -->
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4674	<div class="doc_subsubsection">
Chris Lattner	3333747	2006-01-13 23:26:01 +0000	[diff] [blame]	4675	<a name="i_va_arg">'<tt>va_arg</tt>' Instruction</a>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4676	</div>
				4677
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4678	<div class="doc_text">
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4679
Chris Lattner	26ca62e	2003-10-18 05:51:36 +0000	[diff] [blame]	4680	<h5>Syntax:</h5>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4681
				4682	<pre>
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4683	<resultval> = va_arg <va_list*> <arglist>, <argty>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4684	</pre>
				4685
Chris Lattner	26ca62e	2003-10-18 05:51:36 +0000	[diff] [blame]	4686	<h5>Overview:</h5>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4687
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4688	<p>The '<tt>va_arg</tt>' instruction is used to access arguments passed through
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4689	the "variable argument" area of a function call. It is used to implement the
				4690	<tt>va_arg</tt> macro in C.</p>
				4691
Chris Lattner	26ca62e	2003-10-18 05:51:36 +0000	[diff] [blame]	4692	<h5>Arguments:</h5>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4693
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4694	<p>This instruction takes a <tt>va_list*</tt> value and the type of
				4695	the argument. It returns a value of the specified argument type and
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4696	increments the <tt>va_list</tt> to point to the next argument. The
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4697	actual type of <tt>va_list</tt> is target specific.</p>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4698
Chris Lattner	26ca62e	2003-10-18 05:51:36 +0000	[diff] [blame]	4699	<h5>Semantics:</h5>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4700
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4701	<p>The '<tt>va_arg</tt>' instruction loads an argument of the specified
				4702	type from the specified <tt>va_list</tt> and causes the
				4703	<tt>va_list</tt> to point to the next argument. For more information,
				4704	see the variable argument handling <a href="#int_varargs">Intrinsic
				4705	Functions</a>.</p>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4706
				4707	<p>It is legal for this instruction to be called in a function which does not
				4708	take a variable number of arguments, for example, the <tt>vfprintf</tt>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4709	function.</p>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4710
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4711	<p><tt>va_arg</tt> is an LLVM instruction instead of an <a
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	4712	href="#intrinsics">intrinsic function</a> because it takes a type as an
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4713	argument.</p>
				4714
Chris Lattner	26ca62e	2003-10-18 05:51:36 +0000	[diff] [blame]	4715	<h5>Example:</h5>
Chris Lattner	6a4a049	2004-09-27 21:51:25 +0000	[diff] [blame]	4716
				4717	<p>See the <a href="#int_varargs">variable argument processing</a> section.</p>
				4718
Dan Gohman	3065b61	2009-01-12 23:12:39 +0000	[diff] [blame]	4719	<p>Note that the code generator does not yet fully support va_arg
				4720	on many targets. Also, it does not currently support va_arg with
				4721	aggregate types on any target.</p>
				4722
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4723	</div>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4724
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4725	<!-- *********************************************************************** -->
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4726	<div class="doc_section"> <a name="intrinsics">Intrinsic Functions</a> </div>
				4727	<!-- *********************************************************************** -->
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4728
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4729	<div class="doc_text">
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4730
				4731	<p>LLVM supports the notion of an "intrinsic function". These functions have
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	4732	well known names and semantics and are required to follow certain restrictions.
				4733	Overall, these intrinsics represent an extension mechanism for the LLVM
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4734	language that does not require changing all of the transformations in LLVM when
Gabor Greif	a54634a	2007-07-06 22:07:22 +0000	[diff] [blame]	4735	adding to the language (or the bitcode reader/writer, the parser, etc...).</p>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4736
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	4737	<p>Intrinsic function names must all start with an "<tt>llvm.</tt>" prefix. This
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4738	prefix is reserved in LLVM for intrinsic names; thus, function names may not
				4739	begin with this prefix. Intrinsic functions must always be external functions:
				4740	you cannot define the body of intrinsic functions. Intrinsic functions may
				4741	only be used in call or invoke instructions: it is illegal to take the address
				4742	of an intrinsic function. Additionally, because intrinsic functions are part
				4743	of the LLVM language, it is required if any are added that they be documented
				4744	here.</p>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4745
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	4746	<p>Some intrinsic functions can be overloaded, i.e., the intrinsic represents
				4747	a family of functions that perform the same operation but on different data
				4748	types. Because LLVM can represent over 8 million different integer types,
				4749	overloading is used commonly to allow an intrinsic function to operate on any
				4750	integer type. One or more of the argument types or the result type can be
				4751	overloaded to accept any integer type. Argument types may also be defined as
				4752	exactly matching a previous argument's type or the result type. This allows an
				4753	intrinsic function which accepts multiple arguments, but needs all of them to
				4754	be of the same type, to only be overloaded with respect to a single argument or
				4755	the result.</p>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4756
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	4757	<p>Overloaded intrinsics will have the names of its overloaded argument types
				4758	encoded into its function name, each preceded by a period. Only those types
				4759	which are overloaded result in a name suffix. Arguments whose type is matched
				4760	against another type do not. For example, the <tt>llvm.ctpop</tt> function can
				4761	take an integer of any width and returns an integer of exactly the same integer
				4762	width. This leads to a family of functions such as
				4763	<tt>i8 @llvm.ctpop.i8(i8 %val)</tt> and <tt>i29 @llvm.ctpop.i29(i29 %val)</tt>.
				4764	Only one type, the return type, is overloaded, and only one type suffix is
				4765	required. Because the argument's type is matched against the return type, it
				4766	does not require its own name suffix.</p>
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	4767
				4768	<p>To learn how to add an intrinsic function, please see the
				4769	<a href="ExtendingLLVM.html">Extending LLVM Guide</a>.
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4770	</p>
				4771
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4772	</div>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4773
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4774	<!-- ======================================================================= -->
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4775	<div class="doc_subsection">
				4776	<a name="int_varargs">Variable Argument Handling Intrinsics</a>
				4777	</div>
				4778
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4779	<div class="doc_text">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4780
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4781	<p>Variable argument support is defined in LLVM with the <a
Chris Lattner	3333747	2006-01-13 23:26:01 +0000	[diff] [blame]	4782	href="#i_va_arg"><tt>va_arg</tt></a> instruction and these three
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4783	intrinsic functions. These functions are related to the similarly
				4784	named macros defined in the <tt><stdarg.h></tt> header file.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4785
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4786	<p>All of these functions operate on arguments that use a
				4787	target-specific value type "<tt>va_list</tt>". The LLVM assembly
				4788	language reference manual does not define what this type is, so all
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4789	transformations should be prepared to handle these functions regardless of
				4790	the type used.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4791
Chris Lattner	30b868d	2006-05-15 17:26:46 +0000	[diff] [blame]	4792	<p>This example shows how the <a href="#i_va_arg"><tt>va_arg</tt></a>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4793	instruction and the variable argument handling intrinsic functions are
				4794	used.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4795
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	4796	<div class="doc_code">
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4797	<pre>
Anton Korobeynikov	640bbe0	2007-03-21 23:58:04 +0000	[diff] [blame]	4798	define i32 @test(i32 %X, ...) {
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4799	; Initialize variable argument processing
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4800	%ap = alloca i8*
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4801	%ap2 = bitcast i8** %ap to i8*
Anton Korobeynikov	640bbe0	2007-03-21 23:58:04 +0000	[diff] [blame]	4802	call void @llvm.va_start(i8* %ap2)
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4803
				4804	; Read a single integer argument
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4805	%tmp = va_arg i8** %ap, i32
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4806
				4807	; Demonstrate usage of llvm.va_copy and llvm.va_end
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4808	%aq = alloca i8*
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4809	%aq2 = bitcast i8** %aq to i8*
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4810	call void @llvm.va_copy(i8* %aq2, i8* %ap2)
Anton Korobeynikov	640bbe0	2007-03-21 23:58:04 +0000	[diff] [blame]	4811	call void @llvm.va_end(i8* %aq2)
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4812
				4813	; Stop processing of arguments.
Anton Korobeynikov	640bbe0	2007-03-21 23:58:04 +0000	[diff] [blame]	4814	call void @llvm.va_end(i8* %ap2)
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	4815	ret i32 %tmp
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4816	}
Anton Korobeynikov	640bbe0	2007-03-21 23:58:04 +0000	[diff] [blame]	4817
				4818	declare void @llvm.va_start(i8*)
				4819	declare void @llvm.va_copy(i8, i8)
				4820	declare void @llvm.va_end(i8*)
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4821	</pre>
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4822	</div>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4823
Bill Wendling	3716c5d	2007-05-29 09:04:49 +0000	[diff] [blame]	4824	</div>
				4825
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4826	<!-- _______________________________________________________________________ -->
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4827	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4828	<a name="int_va_start">'<tt>llvm.va_start</tt>' Intrinsic</a>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4829	</div>
				4830
				4831
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4832	<div class="doc_text">
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4833	<h5>Syntax:</h5>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4834	<pre> declare void %llvm.va_start(i8* <arglist>)<br></pre>
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4835	<h5>Overview:</h5>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4836	<p>The '<tt>llvm.va_start</tt>' intrinsic initializes
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4837	<tt>*<arglist></tt> for subsequent use by <tt><a
				4838	href="#i_va_arg">va_arg</a></tt>.</p>
				4839
				4840	<h5>Arguments:</h5>
				4841
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4842	<p>The argument is a pointer to a <tt>va_list</tt> element to initialize.</p>
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4843
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4844	<h5>Semantics:</h5>
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4845
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	4846	<p>The '<tt>llvm.va_start</tt>' intrinsic works just like the <tt>va_start</tt>
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4847	macro available in C. In a target-dependent way, it initializes the
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4848	<tt>va_list</tt> element to which the argument points, so that the next call to
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4849	<tt>va_arg</tt> will produce the first variable argument passed to the function.
				4850	Unlike the C <tt>va_start</tt> macro, this intrinsic does not need to know the
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4851	last argument of the function as the compiler can figure that out.</p>
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4852
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4853	</div>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4854
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4855	<!-- _______________________________________________________________________ -->
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4856	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4857	<a name="int_va_end">'<tt>llvm.va_end</tt>' Intrinsic</a>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4858	</div>
				4859
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4860	<div class="doc_text">
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4861	<h5>Syntax:</h5>
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	4862	<pre> declare void @llvm.va_end(i8* <arglist>)<br></pre>
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4863	<h5>Overview:</h5>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4864
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4865	<p>The '<tt>llvm.va_end</tt>' intrinsic destroys <tt>*<arglist></tt>,
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4866	which has been initialized previously with <tt><a href="#int_va_start">llvm.va_start</a></tt>
Chris Lattner	48b383b0	2003-11-25 01:02:51 +0000	[diff] [blame]	4867	or <tt><a href="#i_va_copy">llvm.va_copy</a></tt>.</p>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4868
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4869	<h5>Arguments:</h5>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4870
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4871	<p>The argument is a pointer to a <tt>va_list</tt> to destroy.</p>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4872
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4873	<h5>Semantics:</h5>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4874
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4875	<p>The '<tt>llvm.va_end</tt>' intrinsic works just like the <tt>va_end</tt>
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4876	macro available in C. In a target-dependent way, it destroys the
				4877	<tt>va_list</tt> element to which the argument points. Calls to <a
				4878	href="#int_va_start"><tt>llvm.va_start</tt></a> and <a href="#int_va_copy">
				4879	<tt>llvm.va_copy</tt></a> must be matched exactly with calls to
				4880	<tt>llvm.va_end</tt>.</p>
Chris Lattner	db0790c	2007-01-08 07:55:15 +0000	[diff] [blame]	4881
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4882	</div>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4883
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4884	<!-- _______________________________________________________________________ -->
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4885	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4886	<a name="int_va_copy">'<tt>llvm.va_copy</tt>' Intrinsic</a>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4887	</div>
				4888
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4889	<div class="doc_text">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4890
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4891	<h5>Syntax:</h5>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4892
				4893	<pre>
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	4894	declare void @llvm.va_copy(i8* <destarglist>, i8* <srcarglist>)
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4895	</pre>
				4896
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4897	<h5>Overview:</h5>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4898
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4899	<p>The '<tt>llvm.va_copy</tt>' intrinsic copies the current argument position
				4900	from the source argument list to the destination argument list.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4901
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4902	<h5>Arguments:</h5>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4903
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4904	<p>The first argument is a pointer to a <tt>va_list</tt> element to initialize.
Andrew Lenharth	5305ea5	2005-06-22 20:38:11 +0000	[diff] [blame]	4905	The second argument is a pointer to a <tt>va_list</tt> element to copy from.</p>
Andrew Lenharth	5fb787c	2005-06-18 18:28:17 +0000	[diff] [blame]	4906
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4907
Chris Lattner	bd64b4e	2003-05-08 04:57:36 +0000	[diff] [blame]	4908	<h5>Semantics:</h5>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4909
Jeff Cohen	222a8a4	2007-04-29 01:07:00 +0000	[diff] [blame]	4910	<p>The '<tt>llvm.va_copy</tt>' intrinsic works just like the <tt>va_copy</tt>
				4911	macro available in C. In a target-dependent way, it copies the source
				4912	<tt>va_list</tt> element into the destination <tt>va_list</tt> element. This
				4913	intrinsic is necessary because the <tt><a href="#int_va_start">
				4914	llvm.va_start</a></tt> intrinsic may be arbitrarily complex and require, for
				4915	example, memory allocation.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4916
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	4917	</div>
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	4918
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	4919	<!-- ======================================================================= -->
				4920	<div class="doc_subsection">
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4921	<a name="int_gc">Accurate Garbage Collection Intrinsics</a>
				4922	</div>
				4923
				4924	<div class="doc_text">
				4925
				4926	<p>
				4927	LLVM support for <a href="GarbageCollection.html">Accurate Garbage
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	4928	Collection</a> (GC) requires the implementation and generation of these
				4929	intrinsics.
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4930	These intrinsics allow identification of <a href="#int_gcroot">GC roots on the
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4931	stack</a>, as well as garbage collector implementations that require <a
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4932	href="#int_gcread">read</a> and <a href="#int_gcwrite">write</a> barriers.
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4933	Front-ends for type-safe garbage collected languages should generate these
				4934	intrinsics to make use of the LLVM garbage collectors. For more details, see <a
				4935	href="GarbageCollection.html">Accurate Garbage Collection with LLVM</a>.
				4936	</p>
Christopher Lamb	55c6d4f	2007-12-17 01:00:21 +0000	[diff] [blame]	4937
				4938	<p>The garbage collection intrinsics only operate on objects in the generic
				4939	address space (address space zero).</p>
				4940
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4941	</div>
				4942
				4943	<!-- _______________________________________________________________________ -->
				4944	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4945	<a name="int_gcroot">'<tt>llvm.gcroot</tt>' Intrinsic</a>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4946	</div>
				4947
				4948	<div class="doc_text">
				4949
				4950	<h5>Syntax:</h5>
				4951
				4952	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	4953	declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4954	</pre>
				4955
				4956	<h5>Overview:</h5>
				4957
John Criswell	dfe6a86	2004-12-10 15:51:16 +0000	[diff] [blame]	4958	<p>The '<tt>llvm.gcroot</tt>' intrinsic declares the existence of a GC root to
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4959	the code generator, and allows some metadata to be associated with it.</p>
				4960
				4961	<h5>Arguments:</h5>
				4962
				4963	<p>The first argument specifies the address of a stack object that contains the
				4964	root pointer. The second pointer (which must be either a constant or a global
				4965	value address) contains the meta-data to be associated with the root.</p>
				4966
				4967	<h5>Semantics:</h5>
				4968
Chris Lattner	851b771	2008-04-24 05:59:56 +0000	[diff] [blame]	4969	<p>At runtime, a call to this intrinsic stores a null pointer into the "ptrloc"
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4970	location. At compile-time, the code generator generates information to allow
Gordon Henriksen	fb56bde	2007-12-25 02:31:26 +0000	[diff] [blame]	4971	the runtime to find the pointer at GC safe points. The '<tt>llvm.gcroot</tt>'
				4972	intrinsic may only be used in a function which <a href="#gc">specifies a GC
				4973	algorithm</a>.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4974
				4975	</div>
				4976
				4977
				4978	<!-- _______________________________________________________________________ -->
				4979	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	4980	<a name="int_gcread">'<tt>llvm.gcread</tt>' Intrinsic</a>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4981	</div>
				4982
				4983	<div class="doc_text">
				4984
				4985	<h5>Syntax:</h5>
				4986
				4987	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	4988	declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	4989	</pre>
				4990
				4991	<h5>Overview:</h5>
				4992
				4993	<p>The '<tt>llvm.gcread</tt>' intrinsic identifies reads of references from heap
				4994	locations, allowing garbage collector implementations that require read
				4995	barriers.</p>
				4996
				4997	<h5>Arguments:</h5>
				4998
Chris Lattner	f922807	2006-03-14 20:02:51 +0000	[diff] [blame]	4999	<p>The second argument is the address to read from, which should be an address
				5000	allocated from the garbage collector. The first object is a pointer to the
				5001	start of the referenced object, if needed by the language runtime (otherwise
				5002	null).</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	5003
				5004	<h5>Semantics:</h5>
				5005
				5006	<p>The '<tt>llvm.gcread</tt>' intrinsic has the same semantics as a load
				5007	instruction, but may be replaced with substantially more complex code by the
Gordon Henriksen	fb56bde	2007-12-25 02:31:26 +0000	[diff] [blame]	5008	garbage collector runtime, as needed. The '<tt>llvm.gcread</tt>' intrinsic
				5009	may only be used in a function which <a href="#gc">specifies a GC
				5010	algorithm</a>.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	5011
				5012	</div>
				5013
				5014
				5015	<!-- _______________________________________________________________________ -->
				5016	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5017	<a name="int_gcwrite">'<tt>llvm.gcwrite</tt>' Intrinsic</a>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	5018	</div>
				5019
				5020	<div class="doc_text">
				5021
				5022	<h5>Syntax:</h5>
				5023
				5024	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	5025	declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	5026	</pre>
				5027
				5028	<h5>Overview:</h5>
				5029
				5030	<p>The '<tt>llvm.gcwrite</tt>' intrinsic identifies writes of references to heap
				5031	locations, allowing garbage collector implementations that require write
				5032	barriers (such as generational or reference counting collectors).</p>
				5033
				5034	<h5>Arguments:</h5>
				5035
Chris Lattner	f922807	2006-03-14 20:02:51 +0000	[diff] [blame]	5036	<p>The first argument is the reference to store, the second is the start of the
				5037	object to store it to, and the third is the address of the field of Obj to
				5038	store to. If the runtime does not require a pointer to the object, Obj may be
				5039	null.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	5040
				5041	<h5>Semantics:</h5>
				5042
				5043	<p>The '<tt>llvm.gcwrite</tt>' intrinsic has the same semantics as a store
				5044	instruction, but may be replaced with substantially more complex code by the
Gordon Henriksen	fb56bde	2007-12-25 02:31:26 +0000	[diff] [blame]	5045	garbage collector runtime, as needed. The '<tt>llvm.gcwrite</tt>' intrinsic
				5046	may only be used in a function which <a href="#gc">specifies a GC
				5047	algorithm</a>.</p>
Chris Lattner	757528b0b	2004-05-23 21:06:01 +0000	[diff] [blame]	5048
				5049	</div>
				5050
				5051
				5052
				5053	<!-- ======================================================================= -->
				5054	<div class="doc_subsection">
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5055	<a name="int_codegen">Code Generator Intrinsics</a>
				5056	</div>
				5057
				5058	<div class="doc_text">
				5059	<p>
				5060	These intrinsics are provided by LLVM to expose special features that may only
				5061	be implemented with code generator support.
				5062	</p>
				5063
				5064	</div>
				5065
				5066	<!-- _______________________________________________________________________ -->
				5067	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5068	<a name="int_returnaddress">'<tt>llvm.returnaddress</tt>' Intrinsic</a>
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5069	</div>
				5070
				5071	<div class="doc_text">
				5072
				5073	<h5>Syntax:</h5>
				5074	<pre>
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5075	declare i8 *@llvm.returnaddress(i32 <level>)
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5076	</pre>
				5077
				5078	<h5>Overview:</h5>
				5079
				5080	<p>
Chris Lattner	c1fb426	2006-10-15 20:05:59 +0000	[diff] [blame]	5081	The '<tt>llvm.returnaddress</tt>' intrinsic attempts to compute a
				5082	target-specific value indicating the return address of the current function
				5083	or one of its callers.
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5084	</p>
				5085
				5086	<h5>Arguments:</h5>
				5087
				5088	<p>
				5089	The argument to this intrinsic indicates which function to return the address
				5090	for. Zero indicates the calling function, one indicates its caller, etc. The
				5091	argument is <b>required</b> to be a constant integer value.
				5092	</p>
				5093
				5094	<h5>Semantics:</h5>
				5095
				5096	<p>
				5097	The '<tt>llvm.returnaddress</tt>' intrinsic either returns a pointer indicating
				5098	the return address of the specified call frame, or zero if it cannot be
				5099	identified. The value returned by this intrinsic is likely to be incorrect or 0
				5100	for arguments other than zero, so it should only be used for debugging purposes.
				5101	</p>
				5102
				5103	<p>
				5104	Note that calling this intrinsic does not prevent function inlining or other
Chris Lattner	2e6eb5f	2005-03-07 20:30:51 +0000	[diff] [blame]	5105	aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5106	source-language caller.
				5107	</p>
				5108	</div>
				5109
				5110
				5111	<!-- _______________________________________________________________________ -->
				5112	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5113	<a name="int_frameaddress">'<tt>llvm.frameaddress</tt>' Intrinsic</a>
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5114	</div>
				5115
				5116	<div class="doc_text">
				5117
				5118	<h5>Syntax:</h5>
				5119	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	5120	declare i8 *@llvm.frameaddress(i32 <level>)
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5121	</pre>
				5122
				5123	<h5>Overview:</h5>
				5124
				5125	<p>
Chris Lattner	c1fb426	2006-10-15 20:05:59 +0000	[diff] [blame]	5126	The '<tt>llvm.frameaddress</tt>' intrinsic attempts to return the
				5127	target-specific frame pointer value for the specified stack frame.
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5128	</p>
				5129
				5130	<h5>Arguments:</h5>
				5131
				5132	<p>
				5133	The argument to this intrinsic indicates which function to return the frame
				5134	pointer for. Zero indicates the calling function, one indicates its caller,
				5135	etc. The argument is <b>required</b> to be a constant integer value.
				5136	</p>
				5137
				5138	<h5>Semantics:</h5>
				5139
				5140	<p>
				5141	The '<tt>llvm.frameaddress</tt>' intrinsic either returns a pointer indicating
				5142	the frame address of the specified call frame, or zero if it cannot be
				5143	identified. The value returned by this intrinsic is likely to be incorrect or 0
				5144	for arguments other than zero, so it should only be used for debugging purposes.
				5145	</p>
				5146
				5147	<p>
				5148	Note that calling this intrinsic does not prevent function inlining or other
Chris Lattner	2e6eb5f	2005-03-07 20:30:51 +0000	[diff] [blame]	5149	aggressive transformations, so the value returned may not be that of the obvious
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5150	source-language caller.
				5151	</p>
				5152	</div>
				5153
Chris Lattner	c8a2c22	2005-02-28 19:24:19 +0000	[diff] [blame]	5154	<!-- _______________________________________________________________________ -->
				5155	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5156	<a name="int_stacksave">'<tt>llvm.stacksave</tt>' Intrinsic</a>
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5157	</div>
				5158
				5159	<div class="doc_text">
				5160
				5161	<h5>Syntax:</h5>
				5162	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	5163	declare i8 *@llvm.stacksave()
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5164	</pre>
				5165
				5166	<h5>Overview:</h5>
				5167
				5168	<p>
				5169	The '<tt>llvm.stacksave</tt>' intrinsic is used to remember the current state of
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5170	the function stack, for use with <a href="#int_stackrestore">
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5171	<tt>llvm.stackrestore</tt></a>. This is useful for implementing language
				5172	features like scoped automatic variable sized arrays in C99.
				5173	</p>
				5174
				5175	<h5>Semantics:</h5>
				5176
				5177	<p>
				5178	This intrinsic returns a opaque pointer value that can be passed to <a
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5179	href="#int_stackrestore"><tt>llvm.stackrestore</tt></a>. When an
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5180	<tt>llvm.stackrestore</tt> intrinsic is executed with a value saved from
				5181	<tt>llvm.stacksave</tt>, it effectively restores the state of the stack to the
				5182	state it was in when the <tt>llvm.stacksave</tt> intrinsic executed. In
				5183	practice, this pops any <a href="#i_alloca">alloca</a> blocks from the stack
				5184	that were allocated after the <tt>llvm.stacksave</tt> was executed.
				5185	</p>
				5186
				5187	</div>
				5188
				5189	<!-- _______________________________________________________________________ -->
				5190	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5191	<a name="int_stackrestore">'<tt>llvm.stackrestore</tt>' Intrinsic</a>
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5192	</div>
				5193
				5194	<div class="doc_text">
				5195
				5196	<h5>Syntax:</h5>
				5197	<pre>
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5198	declare void @llvm.stackrestore(i8 * %ptr)
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5199	</pre>
				5200
				5201	<h5>Overview:</h5>
				5202
				5203	<p>
				5204	The '<tt>llvm.stackrestore</tt>' intrinsic is used to restore the state of
				5205	the function stack to the state it was in when the corresponding <a
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5206	href="#int_stacksave"><tt>llvm.stacksave</tt></a> intrinsic executed. This is
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5207	useful for implementing language features like scoped automatic variable sized
				5208	arrays in C99.
				5209	</p>
				5210
				5211	<h5>Semantics:</h5>
				5212
				5213	<p>
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5214	See the description for <a href="#int_stacksave"><tt>llvm.stacksave</tt></a>.
Chris Lattner	2f0f001	2006-01-13 02:03:13 +0000	[diff] [blame]	5215	</p>
				5216
				5217	</div>
				5218
				5219
				5220	<!-- _______________________________________________________________________ -->
				5221	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5222	<a name="int_prefetch">'<tt>llvm.prefetch</tt>' Intrinsic</a>
Chris Lattner	c8a2c22	2005-02-28 19:24:19 +0000	[diff] [blame]	5223	</div>
				5224
				5225	<div class="doc_text">
				5226
				5227	<h5>Syntax:</h5>
				5228	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	5229	declare void @llvm.prefetch(i8* <address>, i32 <rw>, i32 <locality>)
Chris Lattner	c8a2c22	2005-02-28 19:24:19 +0000	[diff] [blame]	5230	</pre>
				5231
				5232	<h5>Overview:</h5>
				5233
				5234
				5235	<p>
				5236	The '<tt>llvm.prefetch</tt>' intrinsic is a hint to the code generator to insert
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	5237	a prefetch instruction if supported; otherwise, it is a noop. Prefetches have
				5238	no
				5239	effect on the behavior of the program but can change its performance
Chris Lattner	ff85107	2005-02-28 19:47:14 +0000	[diff] [blame]	5240	characteristics.
Chris Lattner	c8a2c22	2005-02-28 19:24:19 +0000	[diff] [blame]	5241	</p>
				5242
				5243	<h5>Arguments:</h5>
				5244
				5245	<p>
				5246	<tt>address</tt> is the address to be prefetched, <tt>rw</tt> is the specifier
				5247	determining if the fetch should be for a read (0) or write (1), and
				5248	<tt>locality</tt> is a temporal locality specifier ranging from (0) - no
Chris Lattner	d3e641c	2005-03-07 20:31:38 +0000	[diff] [blame]	5249	locality, to (3) - extremely local keep in cache. The <tt>rw</tt> and
Chris Lattner	c8a2c22	2005-02-28 19:24:19 +0000	[diff] [blame]	5250	<tt>locality</tt> arguments must be constant integers.
				5251	</p>
				5252
				5253	<h5>Semantics:</h5>
				5254
				5255	<p>
				5256	This intrinsic does not modify the behavior of the program. In particular,
				5257	prefetches cannot trap and do not produce a value. On targets that support this
				5258	intrinsic, the prefetch can provide hints to the processor cache for better
				5259	performance.
				5260	</p>
				5261
				5262	</div>
				5263
Andrew Lenharth	b442791	2005-03-28 20:05:49 +0000	[diff] [blame]	5264	<!-- _______________________________________________________________________ -->
				5265	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5266	<a name="int_pcmarker">'<tt>llvm.pcmarker</tt>' Intrinsic</a>
Andrew Lenharth	b442791	2005-03-28 20:05:49 +0000	[diff] [blame]	5267	</div>
				5268
				5269	<div class="doc_text">
				5270
				5271	<h5>Syntax:</h5>
				5272	<pre>
Chris Lattner	1247773	2007-09-21 17:30:40 +0000	[diff] [blame]	5273	declare void @llvm.pcmarker(i32 <id>)
Andrew Lenharth	b442791	2005-03-28 20:05:49 +0000	[diff] [blame]	5274	</pre>
				5275
				5276	<h5>Overview:</h5>
				5277
				5278
				5279	<p>
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	5280	The '<tt>llvm.pcmarker</tt>' intrinsic is a method to export a Program Counter
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	5281	(PC) in a region of
				5282	code to simulators and other tools. The method is target specific, but it is
				5283	expected that the marker will use exported symbols to transmit the PC of the
				5284	marker.
				5285	The marker makes no guarantees that it will remain with any specific instruction
				5286	after optimizations. It is possible that the presence of a marker will inhibit
Chris Lattner	b40261e	2006-03-24 07:16:10 +0000	[diff] [blame]	5287	optimizations. The intended use is to be inserted after optimizations to allow
John Criswell	8819056	2005-05-16 16:17:45 +0000	[diff] [blame]	5288	correlations of simulation runs.
Andrew Lenharth	b442791	2005-03-28 20:05:49 +0000	[diff] [blame]	5289	</p>
				5290
				5291	<h5>Arguments:</h5>
				5292
				5293	<p>
				5294	<tt>id</tt> is a numerical id identifying the marker.
				5295	</p>
				5296
				5297	<h5>Semantics:</h5>
				5298
				5299	<p>
				5300	This intrinsic does not modify the behavior of the program. Backends that do not
				5301	support this intrinisic may ignore it.
				5302	</p>
				5303
				5304	</div>
				5305
Andrew Lenharth	01aa563	2005-11-11 16:47:30 +0000	[diff] [blame]	5306	<!-- _______________________________________________________________________ -->
				5307	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5308	<a name="int_readcyclecounter">'<tt>llvm.readcyclecounter</tt>' Intrinsic</a>
Andrew Lenharth	01aa563	2005-11-11 16:47:30 +0000	[diff] [blame]	5309	</div>
				5310
				5311	<div class="doc_text">
				5312
				5313	<h5>Syntax:</h5>
				5314	<pre>
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5315	declare i64 @llvm.readcyclecounter( )
Andrew Lenharth	01aa563	2005-11-11 16:47:30 +0000	[diff] [blame]	5316	</pre>
				5317
				5318	<h5>Overview:</h5>
				5319
				5320
				5321	<p>
				5322	The '<tt>llvm.readcyclecounter</tt>' intrinsic provides access to the cycle
				5323	counter register (or similar low latency, high accuracy clocks) on those targets
				5324	that support it. On X86, it should map to RDTSC. On Alpha, it should map to RPCC.
				5325	As the backing counters overflow quickly (on the order of 9 seconds on alpha), this
				5326	should only be used for small timings.
				5327	</p>
				5328
				5329	<h5>Semantics:</h5>
				5330
				5331	<p>
				5332	When directly supported, reading the cycle counter should not modify any memory.
				5333	Implementations are allowed to either return a application specific value or a
				5334	system wide value. On backends without support, this is lowered to a constant 0.
				5335	</p>
				5336
				5337	</div>
				5338
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5339	<!-- ======================================================================= -->
				5340	<div class="doc_subsection">
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5341	<a name="int_libc">Standard C Library Intrinsics</a>
				5342	</div>
				5343
				5344	<div class="doc_text">
				5345	<p>
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5346	LLVM provides intrinsics for a few important standard C library functions.
				5347	These intrinsics allow source-language front-ends to pass information about the
				5348	alignment of the pointer arguments to the code generator, providing opportunity
				5349	for more efficient code generation.
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5350	</p>
				5351
				5352	</div>
				5353
				5354	<!-- _______________________________________________________________________ -->
				5355	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5356	<a name="int_memcpy">'<tt>llvm.memcpy</tt>' Intrinsic</a>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5357	</div>
				5358
				5359	<div class="doc_text">
				5360
				5361	<h5>Syntax:</h5>
Chris Lattner	dd70834	2008-11-21 16:42:48 +0000	[diff] [blame]	5362	<p>This is an overloaded intrinsic. You can use llvm.memcpy on any integer bit
				5363	width. Not all targets support all bit widths however.</p>
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5364	<pre>
Chris Lattner	dd70834	2008-11-21 16:42:48 +0000	[diff] [blame]	5365	declare void @llvm.memcpy.i8(i8 * <dest>, i8 * <src>,
				5366	i8 <len>, i32 <align>)
				5367	declare void @llvm.memcpy.i16(i8 * <dest>, i8 * <src>,
				5368	i16 <len>, i32 <align>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5369	declare void @llvm.memcpy.i32(i8 * <dest>, i8 * <src>,
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5370	i32 <len>, i32 <align>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5371	declare void @llvm.memcpy.i64(i8 * <dest>, i8 * <src>,
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5372	i64 <len>, i32 <align>)
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5373	</pre>
				5374
				5375	<h5>Overview:</h5>
				5376
				5377	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5378	The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5379	location to the destination location.
				5380	</p>
				5381
				5382	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5383	Note that, unlike the standard libc function, the <tt>llvm.memcpy.*</tt>
				5384	intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5385	</p>
				5386
				5387	<h5>Arguments:</h5>
				5388
				5389	<p>
				5390	The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5391	the source. The third argument is an integer argument
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5392	specifying the number of bytes to copy, and the fourth argument is the alignment
				5393	of the source and destination locations.
				5394	</p>
				5395
Chris Lattner	4c67c48	2004-02-12 21:18:15 +0000	[diff] [blame]	5396	<p>
				5397	If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner	5316e5d	2006-03-04 00:02:10 +0000	[diff] [blame]	5398	the caller guarantees that both the source and destination pointers are aligned
				5399	to that boundary.
Chris Lattner	4c67c48	2004-02-12 21:18:15 +0000	[diff] [blame]	5400	</p>
				5401
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5402	<h5>Semantics:</h5>
				5403
				5404	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5405	The '<tt>llvm.memcpy.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner	fee1146	2004-02-12 17:01:32 +0000	[diff] [blame]	5406	location to the destination location, which are not allowed to overlap. It
				5407	copies "len" bytes of memory over. If the argument is known to be aligned to
				5408	some boundary, this can be specified as the fourth argument, otherwise it should
				5409	be set to 0 or 1.
				5410	</p>
				5411	</div>
				5412
				5413
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5414	<!-- _______________________________________________________________________ -->
				5415	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5416	<a name="int_memmove">'<tt>llvm.memmove</tt>' Intrinsic</a>
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5417	</div>
				5418
				5419	<div class="doc_text">
				5420
				5421	<h5>Syntax:</h5>
Chris Lattner	dd70834	2008-11-21 16:42:48 +0000	[diff] [blame]	5422	<p>This is an overloaded intrinsic. You can use llvm.memmove on any integer bit
				5423	width. Not all targets support all bit widths however.</p>
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5424	<pre>
Chris Lattner	dd70834	2008-11-21 16:42:48 +0000	[diff] [blame]	5425	declare void @llvm.memmove.i8(i8 * <dest>, i8 * <src>,
				5426	i8 <len>, i32 <align>)
				5427	declare void @llvm.memmove.i16(i8 * <dest>, i8 * <src>,
				5428	i16 <len>, i32 <align>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5429	declare void @llvm.memmove.i32(i8 * <dest>, i8 * <src>,
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5430	i32 <len>, i32 <align>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5431	declare void @llvm.memmove.i64(i8 * <dest>, i8 * <src>,
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5432	i64 <len>, i32 <align>)
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5433	</pre>
				5434
				5435	<h5>Overview:</h5>
				5436
				5437	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5438	The '<tt>llvm.memmove.*</tt>' intrinsics move a block of memory from the source
				5439	location to the destination location. It is similar to the
Chris Lattner	ec56402	2008-01-06 19:51:52 +0000	[diff] [blame]	5440	'<tt>llvm.memcpy</tt>' intrinsic but allows the two memory locations to overlap.
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5441	</p>
				5442
				5443	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5444	Note that, unlike the standard libc function, the <tt>llvm.memmove.*</tt>
				5445	intrinsics do not return a value, and takes an extra alignment argument.
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5446	</p>
				5447
				5448	<h5>Arguments:</h5>
				5449
				5450	<p>
				5451	The first argument is a pointer to the destination, the second is a pointer to
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5452	the source. The third argument is an integer argument
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5453	specifying the number of bytes to copy, and the fourth argument is the alignment
				5454	of the source and destination locations.
				5455	</p>
				5456
Chris Lattner	4c67c48	2004-02-12 21:18:15 +0000	[diff] [blame]	5457	<p>
				5458	If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner	5316e5d	2006-03-04 00:02:10 +0000	[diff] [blame]	5459	the caller guarantees that the source and destination pointers are aligned to
				5460	that boundary.
Chris Lattner	4c67c48	2004-02-12 21:18:15 +0000	[diff] [blame]	5461	</p>
				5462
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5463	<h5>Semantics:</h5>
				5464
				5465	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5466	The '<tt>llvm.memmove.*</tt>' intrinsics copy a block of memory from the source
Chris Lattner	f30152e	2004-02-12 18:10:10 +0000	[diff] [blame]	5467	location to the destination location, which may overlap. It
				5468	copies "len" bytes of memory over. If the argument is known to be aligned to
				5469	some boundary, this can be specified as the fourth argument, otherwise it should
				5470	be set to 0 or 1.
				5471	</p>
				5472	</div>
				5473
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	5474
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5475	<!-- _______________________________________________________________________ -->
				5476	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5477	<a name="int_memset">'<tt>llvm.memset.*</tt>' Intrinsics</a>
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5478	</div>
				5479
				5480	<div class="doc_text">
				5481
				5482	<h5>Syntax:</h5>
Chris Lattner	dd70834	2008-11-21 16:42:48 +0000	[diff] [blame]	5483	<p>This is an overloaded intrinsic. You can use llvm.memset on any integer bit
				5484	width. Not all targets support all bit widths however.</p>
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5485	<pre>
Chris Lattner	dd70834	2008-11-21 16:42:48 +0000	[diff] [blame]	5486	declare void @llvm.memset.i8(i8 * <dest>, i8 <val>,
				5487	i8 <len>, i32 <align>)
				5488	declare void @llvm.memset.i16(i8 * <dest>, i8 <val>,
				5489	i16 <len>, i32 <align>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5490	declare void @llvm.memset.i32(i8 * <dest>, i8 <val>,
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5491	i32 <len>, i32 <align>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5492	declare void @llvm.memset.i64(i8 * <dest>, i8 <val>,
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5493	i64 <len>, i32 <align>)
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5494	</pre>
				5495
				5496	<h5>Overview:</h5>
				5497
				5498	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5499	The '<tt>llvm.memset.*</tt>' intrinsics fill a block of memory with a particular
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5500	byte value.
				5501	</p>
				5502
				5503	<p>
				5504	Note that, unlike the standard libc function, the <tt>llvm.memset</tt> intrinsic
				5505	does not return a value, and takes an extra alignment argument.
				5506	</p>
				5507
				5508	<h5>Arguments:</h5>
				5509
				5510	<p>
				5511	The first argument is a pointer to the destination to fill, the second is the
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5512	byte value to fill it with, the third argument is an integer
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5513	argument specifying the number of bytes to fill, and the fourth argument is the
				5514	known alignment of destination location.
				5515	</p>
				5516
				5517	<p>
				5518	If the call to this intrinisic has an alignment value that is not 0 or 1, then
Chris Lattner	5316e5d	2006-03-04 00:02:10 +0000	[diff] [blame]	5519	the caller guarantees that the destination pointer is aligned to that boundary.
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5520	</p>
				5521
				5522	<h5>Semantics:</h5>
				5523
				5524	<p>
Chris Lattner	0c8b259	2006-03-03 00:07:20 +0000	[diff] [blame]	5525	The '<tt>llvm.memset.*</tt>' intrinsics fill "len" bytes of memory starting at
				5526	the
Chris Lattner	3649c3a	2004-02-14 04:08:35 +0000	[diff] [blame]	5527	destination location. If the argument is known to be aligned to some boundary,
				5528	this can be specified as the fourth argument, otherwise it should be set to 0 or
				5529	1.
				5530	</p>
				5531	</div>
				5532
				5533
Chris Lattner	3b4f437	2004-06-11 02:28:03 +0000	[diff] [blame]	5534	<!-- _______________________________________________________________________ -->
				5535	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5536	<a name="int_sqrt">'<tt>llvm.sqrt.*</tt>' Intrinsic</a>
Chris Lattner	8a8f2e5	2005-07-21 01:29:16 +0000	[diff] [blame]	5537	</div>
				5538
				5539	<div class="doc_text">
				5540
				5541	<h5>Syntax:</h5>
Dale Johannesen	dd89d27	2007-10-02 17:47:38 +0000	[diff] [blame]	5542	<p>This is an overloaded intrinsic. You can use <tt>llvm.sqrt</tt> on any
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5543	floating point or vector of floating point type. Not all targets support all
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5544	types however.</p>
Chris Lattner	8a8f2e5	2005-07-21 01:29:16 +0000	[diff] [blame]	5545	<pre>
Dale Johannesen	dd89d27	2007-10-02 17:47:38 +0000	[diff] [blame]	5546	declare float @llvm.sqrt.f32(float %Val)
				5547	declare double @llvm.sqrt.f64(double %Val)
				5548	declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
				5549	declare fp128 @llvm.sqrt.f128(fp128 %Val)
				5550	declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
Chris Lattner	8a8f2e5	2005-07-21 01:29:16 +0000	[diff] [blame]	5551	</pre>
				5552
				5553	<h5>Overview:</h5>
				5554
				5555	<p>
Reid Spencer	b4f9a6f	2006-01-16 21:12:35 +0000	[diff] [blame]	5556	The '<tt>llvm.sqrt</tt>' intrinsics return the sqrt of the specified operand,
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5557	returning the same value as the libm '<tt>sqrt</tt>' functions would. Unlike
Chris Lattner	8a8f2e5	2005-07-21 01:29:16 +0000	[diff] [blame]	5558	<tt>sqrt</tt> in libm, however, <tt>llvm.sqrt</tt> has undefined behavior for
Chris Lattner	00d7cb9	2008-01-29 07:00:44 +0000	[diff] [blame]	5559	negative numbers other than -0.0 (which allows for better optimization, because
				5560	there is no need to worry about errno being set). <tt>llvm.sqrt(-0.0)</tt> is
				5561	defined to return -0.0 like IEEE sqrt.
Chris Lattner	8a8f2e5	2005-07-21 01:29:16 +0000	[diff] [blame]	5562	</p>
				5563
				5564	<h5>Arguments:</h5>
				5565
				5566	<p>
				5567	The argument and return value are floating point numbers of the same type.
				5568	</p>
				5569
				5570	<h5>Semantics:</h5>
				5571
				5572	<p>
Dan Gohman	33988db	2007-07-16 14:37:41 +0000	[diff] [blame]	5573	This function returns the sqrt of the specified operand if it is a nonnegative
Chris Lattner	8a8f2e5	2005-07-21 01:29:16 +0000	[diff] [blame]	5574	floating point number.
				5575	</p>
				5576	</div>
				5577
Chris Lattner	33b73f9	2006-09-08 06:34:02 +0000	[diff] [blame]	5578	<!-- _______________________________________________________________________ -->
				5579	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5580	<a name="int_powi">'<tt>llvm.powi.*</tt>' Intrinsic</a>
Chris Lattner	33b73f9	2006-09-08 06:34:02 +0000	[diff] [blame]	5581	</div>
				5582
				5583	<div class="doc_text">
				5584
				5585	<h5>Syntax:</h5>
Dale Johannesen	dd89d27	2007-10-02 17:47:38 +0000	[diff] [blame]	5586	<p>This is an overloaded intrinsic. You can use <tt>llvm.powi</tt> on any
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5587	floating point or vector of floating point type. Not all targets support all
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5588	types however.</p>
Chris Lattner	33b73f9	2006-09-08 06:34:02 +0000	[diff] [blame]	5589	<pre>
Dale Johannesen	dd89d27	2007-10-02 17:47:38 +0000	[diff] [blame]	5590	declare float @llvm.powi.f32(float %Val, i32 %power)
				5591	declare double @llvm.powi.f64(double %Val, i32 %power)
				5592	declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
				5593	declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
				5594	declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
Chris Lattner	33b73f9	2006-09-08 06:34:02 +0000	[diff] [blame]	5595	</pre>
				5596
				5597	<h5>Overview:</h5>
				5598
				5599	<p>
				5600	The '<tt>llvm.powi.*</tt>' intrinsics return the first operand raised to the
				5601	specified (positive or negative) power. The order of evaluation of
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5602	multiplications is not defined. When a vector of floating point type is
				5603	used, the second argument remains a scalar integer value.
Chris Lattner	33b73f9	2006-09-08 06:34:02 +0000	[diff] [blame]	5604	</p>
				5605
				5606	<h5>Arguments:</h5>
				5607
				5608	<p>
				5609	The second argument is an integer power, and the first is a value to raise to
				5610	that power.
				5611	</p>
				5612
				5613	<h5>Semantics:</h5>
				5614
				5615	<p>
				5616	This function returns the first value raised to the second power with an
				5617	unspecified sequence of rounding operations.</p>
				5618	</div>
				5619
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5620	<!-- _______________________________________________________________________ -->
				5621	<div class="doc_subsubsection">
				5622	<a name="int_sin">'<tt>llvm.sin.*</tt>' Intrinsic</a>
				5623	</div>
				5624
				5625	<div class="doc_text">
				5626
				5627	<h5>Syntax:</h5>
				5628	<p>This is an overloaded intrinsic. You can use <tt>llvm.sin</tt> on any
				5629	floating point or vector of floating point type. Not all targets support all
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5630	types however.</p>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5631	<pre>
				5632	declare float @llvm.sin.f32(float %Val)
				5633	declare double @llvm.sin.f64(double %Val)
				5634	declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
				5635	declare fp128 @llvm.sin.f128(fp128 %Val)
				5636	declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
				5637	</pre>
				5638
				5639	<h5>Overview:</h5>
				5640
				5641	<p>
				5642	The '<tt>llvm.sin.*</tt>' intrinsics return the sine of the operand.
				5643	</p>
				5644
				5645	<h5>Arguments:</h5>
				5646
				5647	<p>
				5648	The argument and return value are floating point numbers of the same type.
				5649	</p>
				5650
				5651	<h5>Semantics:</h5>
				5652
				5653	<p>
				5654	This function returns the sine of the specified operand, returning the
				5655	same values as the libm <tt>sin</tt> functions would, and handles error
Dan Gohman	d0806a0	2007-10-17 18:05:13 +0000	[diff] [blame]	5656	conditions in the same way.</p>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5657	</div>
				5658
				5659	<!-- _______________________________________________________________________ -->
				5660	<div class="doc_subsubsection">
				5661	<a name="int_cos">'<tt>llvm.cos.*</tt>' Intrinsic</a>
				5662	</div>
				5663
				5664	<div class="doc_text">
				5665
				5666	<h5>Syntax:</h5>
				5667	<p>This is an overloaded intrinsic. You can use <tt>llvm.cos</tt> on any
				5668	floating point or vector of floating point type. Not all targets support all
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5669	types however.</p>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5670	<pre>
				5671	declare float @llvm.cos.f32(float %Val)
				5672	declare double @llvm.cos.f64(double %Val)
				5673	declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
				5674	declare fp128 @llvm.cos.f128(fp128 %Val)
				5675	declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
				5676	</pre>
				5677
				5678	<h5>Overview:</h5>
				5679
				5680	<p>
				5681	The '<tt>llvm.cos.*</tt>' intrinsics return the cosine of the operand.
				5682	</p>
				5683
				5684	<h5>Arguments:</h5>
				5685
				5686	<p>
				5687	The argument and return value are floating point numbers of the same type.
				5688	</p>
				5689
				5690	<h5>Semantics:</h5>
				5691
				5692	<p>
				5693	This function returns the cosine of the specified operand, returning the
				5694	same values as the libm <tt>cos</tt> functions would, and handles error
Dan Gohman	d0806a0	2007-10-17 18:05:13 +0000	[diff] [blame]	5695	conditions in the same way.</p>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5696	</div>
				5697
				5698	<!-- _______________________________________________________________________ -->
				5699	<div class="doc_subsubsection">
				5700	<a name="int_pow">'<tt>llvm.pow.*</tt>' Intrinsic</a>
				5701	</div>
				5702
				5703	<div class="doc_text">
				5704
				5705	<h5>Syntax:</h5>
				5706	<p>This is an overloaded intrinsic. You can use <tt>llvm.pow</tt> on any
				5707	floating point or vector of floating point type. Not all targets support all
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5708	types however.</p>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5709	<pre>
				5710	declare float @llvm.pow.f32(float %Val, float %Power)
				5711	declare double @llvm.pow.f64(double %Val, double %Power)
				5712	declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
				5713	declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
				5714	declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
				5715	</pre>
				5716
				5717	<h5>Overview:</h5>
				5718
				5719	<p>
				5720	The '<tt>llvm.pow.*</tt>' intrinsics return the first operand raised to the
				5721	specified (positive or negative) power.
				5722	</p>
				5723
				5724	<h5>Arguments:</h5>
				5725
				5726	<p>
				5727	The second argument is a floating point power, and the first is a value to
				5728	raise to that power.
				5729	</p>
				5730
				5731	<h5>Semantics:</h5>
				5732
				5733	<p>
				5734	This function returns the first value raised to the second power,
				5735	returning the
				5736	same values as the libm <tt>pow</tt> functions would, and handles error
Dan Gohman	d0806a0	2007-10-17 18:05:13 +0000	[diff] [blame]	5737	conditions in the same way.</p>
Dan Gohman	b6324c1	2007-10-15 20:30:11 +0000	[diff] [blame]	5738	</div>
				5739
Chris Lattner	33b73f9	2006-09-08 06:34:02 +0000	[diff] [blame]	5740
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5741	<!-- ======================================================================= -->
				5742	<div class="doc_subsection">
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5743	<a name="int_manip">Bit Manipulation Intrinsics</a>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5744	</div>
				5745
				5746	<div class="doc_text">
				5747	<p>
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5748	LLVM provides intrinsics for a few important bit manipulation operations.
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5749	These allow efficient code generation for some algorithms.
				5750	</p>
				5751
				5752	</div>
				5753
				5754	<!-- _______________________________________________________________________ -->
				5755	<div class="doc_subsubsection">
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5756	<a name="int_bswap">'<tt>llvm.bswap.*</tt>' Intrinsics</a>
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5757	</div>
				5758
				5759	<div class="doc_text">
				5760
				5761	<h5>Syntax:</h5>
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	5762	<p>This is an overloaded intrinsic function. You can use bswap on any integer
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5763	type that is an even number of bytes (i.e. BitWidth % 16 == 0).</p>
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5764	<pre>
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5765	declare i16 @llvm.bswap.i16(i16 <id>)
				5766	declare i32 @llvm.bswap.i32(i32 <id>)
				5767	declare i64 @llvm.bswap.i64(i64 <id>)
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5768	</pre>
				5769
				5770	<h5>Overview:</h5>
				5771
				5772	<p>
Reid Spencer	f361c4f	2007-04-02 02:25:19 +0000	[diff] [blame]	5773	The '<tt>llvm.bswap</tt>' family of intrinsics is used to byte swap integer
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	5774	values with an even number of bytes (positive multiple of 16 bits). These are
				5775	useful for performing operations on data that is not in the target's native
				5776	byte order.
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5777	</p>
				5778
				5779	<h5>Semantics:</h5>
				5780
				5781	<p>
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5782	The <tt>llvm.bswap.i16</tt> intrinsic returns an i16 value that has the high
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5783	and low byte of the input i16 swapped. Similarly, the <tt>llvm.bswap.i32</tt>
				5784	intrinsic returns an i32 value that has the four bytes of the input i32
				5785	swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the returned
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5786	i32 will have its bytes in 3, 2, 1, 0 order. The <tt>llvm.bswap.i48</tt>,
				5787	<tt>llvm.bswap.i64</tt> and other intrinsics extend this concept to
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	5788	additional even-byte lengths (6 bytes, 8 bytes and more, respectively).
Nate Begeman	0f223bb	2006-01-13 23:26:38 +0000	[diff] [blame]	5789	</p>
				5790
				5791	</div>
				5792
				5793	<!-- _______________________________________________________________________ -->
				5794	<div class="doc_subsubsection">
Reid Spencer	b4f9a6f	2006-01-16 21:12:35 +0000	[diff] [blame]	5795	<a name="int_ctpop">'<tt>llvm.ctpop.*</tt>' Intrinsic</a>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5796	</div>
				5797
				5798	<div class="doc_text">
				5799
				5800	<h5>Syntax:</h5>
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	5801	<p>This is an overloaded intrinsic. You can use llvm.ctpop on any integer bit
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5802	width. Not all targets support all bit widths however.</p>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5803	<pre>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	5804	declare i8 @llvm.ctpop.i8(i8 <src>)
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5805	declare i16 @llvm.ctpop.i16(i16 <src>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5806	declare i32 @llvm.ctpop.i32(i32 <src>)
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5807	declare i64 @llvm.ctpop.i64(i64 <src>)
				5808	declare i256 @llvm.ctpop.i256(i256 <src>)
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5809	</pre>
				5810
				5811	<h5>Overview:</h5>
				5812
				5813	<p>
Chris Lattner	069b5bd	2006-01-16 22:38:59 +0000	[diff] [blame]	5814	The '<tt>llvm.ctpop</tt>' family of intrinsics counts the number of bits set in a
				5815	value.
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5816	</p>
				5817
				5818	<h5>Arguments:</h5>
				5819
				5820	<p>
Chris Lattner	573f64e	2005-05-07 01:46:40 +0000	[diff] [blame]	5821	The only argument is the value to be counted. The argument may be of any
Reid Spencer	3e628eb9	2007-01-04 16:43:23 +0000	[diff] [blame]	5822	integer type. The return type must match the argument type.
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5823	</p>
				5824
				5825	<h5>Semantics:</h5>
				5826
				5827	<p>
				5828	The '<tt>llvm.ctpop</tt>' intrinsic counts the 1's in a variable.
				5829	</p>
				5830	</div>
				5831
				5832	<!-- _______________________________________________________________________ -->
				5833	<div class="doc_subsubsection">
Chris Lattner	b748c67	2006-01-16 22:34:14 +0000	[diff] [blame]	5834	<a name="int_ctlz">'<tt>llvm.ctlz.*</tt>' Intrinsic</a>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5835	</div>
				5836
				5837	<div class="doc_text">
				5838
				5839	<h5>Syntax:</h5>
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	5840	<p>This is an overloaded intrinsic. You can use <tt>llvm.ctlz</tt> on any
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5841	integer bit width. Not all targets support all bit widths however.</p>
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5842	<pre>
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5843	declare i8 @llvm.ctlz.i8 (i8 <src>)
				5844	declare i16 @llvm.ctlz.i16(i16 <src>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5845	declare i32 @llvm.ctlz.i32(i32 <src>)
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5846	declare i64 @llvm.ctlz.i64(i64 <src>)
				5847	declare i256 @llvm.ctlz.i256(i256 <src>)
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5848	</pre>
				5849
				5850	<h5>Overview:</h5>
				5851
				5852	<p>
Reid Spencer	b4f9a6f	2006-01-16 21:12:35 +0000	[diff] [blame]	5853	The '<tt>llvm.ctlz</tt>' family of intrinsic functions counts the number of
				5854	leading zeros in a variable.
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5855	</p>
				5856
				5857	<h5>Arguments:</h5>
				5858
				5859	<p>
Chris Lattner	573f64e	2005-05-07 01:46:40 +0000	[diff] [blame]	5860	The only argument is the value to be counted. The argument may be of any
Reid Spencer	3e628eb9	2007-01-04 16:43:23 +0000	[diff] [blame]	5861	integer type. The return type must match the argument type.
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5862	</p>
				5863
				5864	<h5>Semantics:</h5>
				5865
				5866	<p>
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5867	The '<tt>llvm.ctlz</tt>' intrinsic counts the leading (most significant) zeros
				5868	in a variable. If the src == 0 then the result is the size in bits of the type
Reid Spencer	b5ebf3d	2006-12-31 07:07:53 +0000	[diff] [blame]	5869	of src. For example, <tt>llvm.ctlz(i32 2) = 30</tt>.
Andrew Lenharth	1d46352	2005-05-03 18:01:48 +0000	[diff] [blame]	5870	</p>
				5871	</div>
Chris Lattner	3b4f437	2004-06-11 02:28:03 +0000	[diff] [blame]	5872
				5873
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5874
				5875	<!-- _______________________________________________________________________ -->
				5876	<div class="doc_subsubsection">
Chris Lattner	b748c67	2006-01-16 22:34:14 +0000	[diff] [blame]	5877	<a name="int_cttz">'<tt>llvm.cttz.*</tt>' Intrinsic</a>
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5878	</div>
				5879
				5880	<div class="doc_text">
				5881
				5882	<h5>Syntax:</h5>
Reid Spencer	4eefaab	2007-04-01 08:04:23 +0000	[diff] [blame]	5883	<p>This is an overloaded intrinsic. You can use <tt>llvm.cttz</tt> on any
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5884	integer bit width. Not all targets support all bit widths however.</p>
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5885	<pre>
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5886	declare i8 @llvm.cttz.i8 (i8 <src>)
				5887	declare i16 @llvm.cttz.i16(i16 <src>)
Anton Korobeynikov	be9c93c	2007-03-22 00:02:17 +0000	[diff] [blame]	5888	declare i32 @llvm.cttz.i32(i32 <src>)
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5889	declare i64 @llvm.cttz.i64(i64 <src>)
				5890	declare i256 @llvm.cttz.i256(i256 <src>)
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5891	</pre>
				5892
				5893	<h5>Overview:</h5>
				5894
				5895	<p>
Reid Spencer	b4f9a6f	2006-01-16 21:12:35 +0000	[diff] [blame]	5896	The '<tt>llvm.cttz</tt>' family of intrinsic functions counts the number of
				5897	trailing zeros.
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5898	</p>
				5899
				5900	<h5>Arguments:</h5>
				5901
				5902	<p>
				5903	The only argument is the value to be counted. The argument may be of any
Reid Spencer	3e628eb9	2007-01-04 16:43:23 +0000	[diff] [blame]	5904	integer type. The return type must match the argument type.
Chris Lattner	efa20fa	2005-05-15 19:39:26 +0000	[diff] [blame]	5905	</p>
				5906
				5907	<h5>Semantics:</h5>
				5908
				5909	<p>
				5910	The '<tt>llvm.cttz</tt>' intrinsic counts the trailing (least significant) zeros
				5911	in a variable. If the src == 0 then the result is the size in bits of the type
				5912	of src. For example, <tt>llvm.cttz(2) = 1</tt>.
				5913	</p>
				5914	</div>
				5915
Reid Spencer	8a5799f	2007-04-01 08:27:01 +0000	[diff] [blame]	5916	<!-- _______________________________________________________________________ -->
				5917	<div class="doc_subsubsection">
Reid Spencer	ea2945e	2007-04-10 02:51:31 +0000	[diff] [blame]	5918	<a name="int_part_select">'<tt>llvm.part.select.*</tt>' Intrinsic</a>
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5919	</div>
				5920
				5921	<div class="doc_text">
				5922
				5923	<h5>Syntax:</h5>
Reid Spencer	ea2945e	2007-04-10 02:51:31 +0000	[diff] [blame]	5924	<p>This is an overloaded intrinsic. You can use <tt>llvm.part.select</tt>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5925	on any integer bit width.</p>
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5926	<pre>
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5927	declare i17 @llvm.part.select.i17 (i17 %val, i32 %loBit, i32 %hiBit)
				5928	declare i29 @llvm.part.select.i29 (i29 %val, i32 %loBit, i32 %hiBit)
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5929	</pre>
				5930
				5931	<h5>Overview:</h5>
Reid Spencer	ea2945e	2007-04-10 02:51:31 +0000	[diff] [blame]	5932	<p>The '<tt>llvm.part.select</tt>' family of intrinsic functions selects a
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5933	range of bits from an integer value and returns them in the same bit width as
				5934	the original value.</p>
				5935
				5936	<h5>Arguments:</h5>
				5937	<p>The first argument, <tt>%val</tt> and the result may be integer types of
				5938	any bit width but they must have the same bit width. The second and third
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5939	arguments must be <tt>i32</tt> type since they specify only a bit index.</p>
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5940
				5941	<h5>Semantics:</h5>
Reid Spencer	ea2945e	2007-04-10 02:51:31 +0000	[diff] [blame]	5942	<p>The operation of the '<tt>llvm.part.select</tt>' intrinsic has two modes
Reid Spencer	96a5f02	2007-04-04 02:42:35 +0000	[diff] [blame]	5943	of operation: forwards and reverse. If <tt>%loBit</tt> is greater than
				5944	<tt>%hiBits</tt> then the intrinsic operates in reverse mode. Otherwise it
				5945	operates in forward mode.</p>
				5946	<p>In forward mode, this intrinsic is the equivalent of shifting <tt>%val</tt>
				5947	right by <tt>%loBit</tt> bits and then ANDing it with a mask with
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5948	only the <tt>%hiBit - %loBit</tt> bits set, as follows:</p>
				5949	<ol>
				5950	<li>The <tt>%val</tt> is shifted right (LSHR) by the number of bits specified
				5951	by <tt>%loBits</tt>. This normalizes the value to the low order bits.</li>
				5952	<li>The <tt>%loBits</tt> value is subtracted from the <tt>%hiBits</tt> value
				5953	to determine the number of bits to retain.</li>
				5954	<li>A mask of the retained bits is created by shifting a -1 value.</li>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5955	<li>The mask is ANDed with <tt>%val</tt> to produce the result.</li>
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5956	</ol>
Reid Spencer	70845c0	2007-05-14 16:14:57 +0000	[diff] [blame]	5957	<p>In reverse mode, a similar computation is made except that the bits are
				5958	returned in the reverse order. So, for example, if <tt>X</tt> has the value
				5959	<tt>i16 0x0ACF (101011001111)</tt> and we apply
				5960	<tt>part.select(i16 X, 8, 3)</tt> to it, we get back the value
				5961	<tt>i16 0x0026 (000000100110)</tt>.</p>
Reid Spencer	8bc7d95	2007-04-01 19:00:37 +0000	[diff] [blame]	5962	</div>
				5963
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	5964	<div class="doc_subsubsection">
				5965	<a name="int_part_set">'<tt>llvm.part.set.*</tt>' Intrinsic</a>
				5966	</div>
				5967
				5968	<div class="doc_text">
				5969
				5970	<h5>Syntax:</h5>
				5971	<p>This is an overloaded intrinsic. You can use <tt>llvm.part.set</tt>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	5972	on any integer bit width.</p>
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	5973	<pre>
Chandler Carruth	7132e00	2007-08-04 01:51:18 +0000	[diff] [blame]	5974	declare i17 @llvm.part.set.i17.i9 (i17 %val, i9 %repl, i32 %lo, i32 %hi)
				5975	declare i29 @llvm.part.set.i29.i9 (i29 %val, i9 %repl, i32 %lo, i32 %hi)
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	5976	</pre>
				5977
				5978	<h5>Overview:</h5>
				5979	<p>The '<tt>llvm.part.set</tt>' family of intrinsic functions replaces a range
				5980	of bits in an integer value with another integer value. It returns the integer
				5981	with the replaced bits.</p>
				5982
				5983	<h5>Arguments:</h5>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	5984	<p>The first argument, <tt>%val</tt>, and the result may be integer types of
				5985	any bit width, but they must have the same bit width. <tt>%val</tt> is the value
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	5986	whose bits will be replaced. The second argument, <tt>%repl</tt> may be an
				5987	integer of any bit width. The third and fourth arguments must be <tt>i32</tt>
				5988	type since they specify only a bit index.</p>
				5989
				5990	<h5>Semantics:</h5>
				5991	<p>The operation of the '<tt>llvm.part.set</tt>' intrinsic has two modes
				5992	of operation: forwards and reverse. If <tt>%lo</tt> is greater than
				5993	<tt>%hi</tt> then the intrinsic operates in reverse mode. Otherwise it
				5994	operates in forward mode.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	5995
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	5996	<p>For both modes, the <tt>%repl</tt> value is prepared for use by either
				5997	truncating it down to the size of the replacement area or zero extending it
				5998	up to that size.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	5999
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	6000	<p>In forward mode, the bits between <tt>%lo</tt> and <tt>%hi</tt> (inclusive)
				6001	are replaced with corresponding bits from <tt>%repl</tt>. That is the 0th bit
				6002	in <tt>%repl</tt> replaces the <tt>%lo</tt>th bit in <tt>%val</tt> and etc. up
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	6003	to the <tt>%hi</tt>th bit.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6004
Reid Spencer	146281c	2007-05-14 16:50:20 +0000	[diff] [blame]	6005	<p>In reverse mode, a similar computation is made except that the bits are
				6006	reversed. That is, the <tt>0</tt>th bit in <tt>%repl</tt> replaces the
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	6007	<tt>%hi</tt> bit in <tt>%val</tt> and etc. down to the <tt>%lo</tt>th bit.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6008
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	6009	<h5>Examples:</h5>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6010
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	6011	<pre>
Reid Spencer	c70afc3	2007-04-12 01:03:03 +0000	[diff] [blame]	6012	llvm.part.set(0xFFFF, 0, 4, 7) -> 0xFF0F
Reid Spencer	146281c	2007-05-14 16:50:20 +0000	[diff] [blame]	6013	llvm.part.set(0xFFFF, 0, 7, 4) -> 0xFF0F
				6014	llvm.part.set(0xFFFF, 1, 7, 4) -> 0xFF8F
				6015	llvm.part.set(0xFFFF, F, 8, 3) -> 0xFFE7
Reid Spencer	c70afc3	2007-04-12 01:03:03 +0000	[diff] [blame]	6016	llvm.part.set(0xFFFF, 0, 3, 8) -> 0xFE07
Reid Spencer	7972c47	2007-04-11 23:49:50 +0000	[diff] [blame]	6017	</pre>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6018
				6019	</div>
				6020
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6021	<!-- ======================================================================= -->
				6022	<div class="doc_subsection">
				6023	<a name="int_overflow">Arithmetic with Overflow Intrinsics</a>
				6024	</div>
				6025
				6026	<div class="doc_text">
				6027	<p>
				6028	LLVM provides intrinsics for some arithmetic with overflow operations.
				6029	</p>
				6030
				6031	</div>
				6032
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6033	<!-- _______________________________________________________________________ -->
				6034	<div class="doc_subsubsection">
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6035	<a name="int_sadd_overflow">'<tt>llvm.sadd.with.overflow.*</tt>' Intrinsics</a>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6036	</div>
				6037
				6038	<div class="doc_text">
				6039
				6040	<h5>Syntax:</h5>
				6041
				6042	<p>This is an overloaded intrinsic. You can use <tt>llvm.sadd.with.overflow</tt>
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6043	on any integer bit width.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6044
				6045	<pre>
				6046	declare {i16, i1} @llvm.sadd.with.overflow.i16(i16 %a, i16 %b)
				6047	declare {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
				6048	declare {i64, i1} @llvm.sadd.with.overflow.i64(i64 %a, i64 %b)
				6049	</pre>
				6050
				6051	<h5>Overview:</h5>
				6052
				6053	<p>The '<tt>llvm.sadd.with.overflow</tt>' family of intrinsic functions perform
				6054	a signed addition of the two arguments, and indicate whether an overflow
				6055	occurred during the signed summation.</p>
				6056
				6057	<h5>Arguments:</h5>
				6058
				6059	<p>The arguments (%a and %b) and the first element of the result structure may
				6060	be of integer types of any bit width, but they must have the same bit width. The
				6061	second element of the result structure must be of type <tt>i1</tt>. <tt>%a</tt>
				6062	and <tt>%b</tt> are the two values that will undergo signed addition.</p>
				6063
				6064	<h5>Semantics:</h5>
				6065
				6066	<p>The '<tt>llvm.sadd.with.overflow</tt>' family of intrinsic functions perform
				6067	a signed addition of the two variables. They return a structure — the
				6068	first element of which is the signed summation, and the second element of which
				6069	is a bit specifying if the signed summation resulted in an overflow.</p>
				6070
				6071	<h5>Examples:</h5>
				6072	<pre>
				6073	%res = call {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
				6074	%sum = extractvalue {i32, i1} %res, 0
				6075	%obit = extractvalue {i32, i1} %res, 1
				6076	br i1 %obit, label %overflow, label %normal
				6077	</pre>
				6078
				6079	</div>
				6080
				6081	<!-- _______________________________________________________________________ -->
				6082	<div class="doc_subsubsection">
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6083	<a name="int_uadd_overflow">'<tt>llvm.uadd.with.overflow.*</tt>' Intrinsics</a>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6084	</div>
				6085
				6086	<div class="doc_text">
				6087
				6088	<h5>Syntax:</h5>
				6089
				6090	<p>This is an overloaded intrinsic. You can use <tt>llvm.uadd.with.overflow</tt>
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6091	on any integer bit width.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6092
				6093	<pre>
				6094	declare {i16, i1} @llvm.uadd.with.overflow.i16(i16 %a, i16 %b)
				6095	declare {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
				6096	declare {i64, i1} @llvm.uadd.with.overflow.i64(i64 %a, i64 %b)
				6097	</pre>
				6098
				6099	<h5>Overview:</h5>
				6100
				6101	<p>The '<tt>llvm.uadd.with.overflow</tt>' family of intrinsic functions perform
				6102	an unsigned addition of the two arguments, and indicate whether a carry occurred
				6103	during the unsigned summation.</p>
				6104
				6105	<h5>Arguments:</h5>
				6106
				6107	<p>The arguments (%a and %b) and the first element of the result structure may
				6108	be of integer types of any bit width, but they must have the same bit width. The
				6109	second element of the result structure must be of type <tt>i1</tt>. <tt>%a</tt>
				6110	and <tt>%b</tt> are the two values that will undergo unsigned addition.</p>
				6111
				6112	<h5>Semantics:</h5>
				6113
				6114	<p>The '<tt>llvm.uadd.with.overflow</tt>' family of intrinsic functions perform
				6115	an unsigned addition of the two arguments. They return a structure — the
				6116	first element of which is the sum, and the second element of which is a bit
				6117	specifying if the unsigned summation resulted in a carry.</p>
				6118
				6119	<h5>Examples:</h5>
				6120	<pre>
				6121	%res = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
				6122	%sum = extractvalue {i32, i1} %res, 0
				6123	%obit = extractvalue {i32, i1} %res, 1
				6124	br i1 %obit, label %carry, label %normal
				6125	</pre>
				6126
				6127	</div>
				6128
				6129	<!-- _______________________________________________________________________ -->
				6130	<div class="doc_subsubsection">
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6131	<a name="int_ssub_overflow">'<tt>llvm.ssub.with.overflow.*</tt>' Intrinsics</a>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6132	</div>
				6133
				6134	<div class="doc_text">
				6135
				6136	<h5>Syntax:</h5>
				6137
				6138	<p>This is an overloaded intrinsic. You can use <tt>llvm.ssub.with.overflow</tt>
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6139	on any integer bit width.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6140
				6141	<pre>
				6142	declare {i16, i1} @llvm.ssub.with.overflow.i16(i16 %a, i16 %b)
				6143	declare {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
				6144	declare {i64, i1} @llvm.ssub.with.overflow.i64(i64 %a, i64 %b)
				6145	</pre>
				6146
				6147	<h5>Overview:</h5>
				6148
				6149	<p>The '<tt>llvm.ssub.with.overflow</tt>' family of intrinsic functions perform
				6150	a signed subtraction of the two arguments, and indicate whether an overflow
				6151	occurred during the signed subtraction.</p>
				6152
				6153	<h5>Arguments:</h5>
				6154
				6155	<p>The arguments (%a and %b) and the first element of the result structure may
				6156	be of integer types of any bit width, but they must have the same bit width. The
				6157	second element of the result structure must be of type <tt>i1</tt>. <tt>%a</tt>
				6158	and <tt>%b</tt> are the two values that will undergo signed subtraction.</p>
				6159
				6160	<h5>Semantics:</h5>
				6161
				6162	<p>The '<tt>llvm.ssub.with.overflow</tt>' family of intrinsic functions perform
				6163	a signed subtraction of the two arguments. They return a structure — the
				6164	first element of which is the subtraction, and the second element of which is a bit
				6165	specifying if the signed subtraction resulted in an overflow.</p>
				6166
				6167	<h5>Examples:</h5>
				6168	<pre>
				6169	%res = call {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
				6170	%sum = extractvalue {i32, i1} %res, 0
				6171	%obit = extractvalue {i32, i1} %res, 1
				6172	br i1 %obit, label %overflow, label %normal
				6173	</pre>
				6174
				6175	</div>
				6176
				6177	<!-- _______________________________________________________________________ -->
				6178	<div class="doc_subsubsection">
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6179	<a name="int_usub_overflow">'<tt>llvm.usub.with.overflow.*</tt>' Intrinsics</a>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6180	</div>
				6181
				6182	<div class="doc_text">
				6183
				6184	<h5>Syntax:</h5>
				6185
				6186	<p>This is an overloaded intrinsic. You can use <tt>llvm.usub.with.overflow</tt>
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6187	on any integer bit width.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6188
				6189	<pre>
				6190	declare {i16, i1} @llvm.usub.with.overflow.i16(i16 %a, i16 %b)
				6191	declare {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
				6192	declare {i64, i1} @llvm.usub.with.overflow.i64(i64 %a, i64 %b)
				6193	</pre>
				6194
				6195	<h5>Overview:</h5>
				6196
				6197	<p>The '<tt>llvm.usub.with.overflow</tt>' family of intrinsic functions perform
				6198	an unsigned subtraction of the two arguments, and indicate whether an overflow
				6199	occurred during the unsigned subtraction.</p>
				6200
				6201	<h5>Arguments:</h5>
				6202
				6203	<p>The arguments (%a and %b) and the first element of the result structure may
				6204	be of integer types of any bit width, but they must have the same bit width. The
				6205	second element of the result structure must be of type <tt>i1</tt>. <tt>%a</tt>
				6206	and <tt>%b</tt> are the two values that will undergo unsigned subtraction.</p>
				6207
				6208	<h5>Semantics:</h5>
				6209
				6210	<p>The '<tt>llvm.usub.with.overflow</tt>' family of intrinsic functions perform
				6211	an unsigned subtraction of the two arguments. They return a structure — the
				6212	first element of which is the subtraction, and the second element of which is a bit
				6213	specifying if the unsigned subtraction resulted in an overflow.</p>
				6214
				6215	<h5>Examples:</h5>
				6216	<pre>
				6217	%res = call {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
				6218	%sum = extractvalue {i32, i1} %res, 0
				6219	%obit = extractvalue {i32, i1} %res, 1
				6220	br i1 %obit, label %overflow, label %normal
				6221	</pre>
				6222
				6223	</div>
				6224
				6225	<!-- _______________________________________________________________________ -->
				6226	<div class="doc_subsubsection">
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6227	<a name="int_smul_overflow">'<tt>llvm.smul.with.overflow.*</tt>' Intrinsics</a>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6228	</div>
				6229
				6230	<div class="doc_text">
				6231
				6232	<h5>Syntax:</h5>
				6233
				6234	<p>This is an overloaded intrinsic. You can use <tt>llvm.smul.with.overflow</tt>
Bill Wendling	fd2bd72	2009-02-08 04:04:40 +0000	[diff] [blame]	6235	on any integer bit width.</p>
Bill Wendling	f4d7062	2009-02-08 01:40:31 +0000	[diff] [blame]	6236
				6237	<pre>
				6238	declare {i16, i1} @llvm.smul.with.overflow.i16(i16 %a, i16 %b)
				6239	declare {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
				6240	declare {i64, i1} @llvm.smul.with.overflow.i64(i64 %a, i64 %b)
				6241	</pre>
				6242
				6243	<h5>Overview:</h5>
				6244
				6245	<p>The '<tt>llvm.smul.with.overflow</tt>' family of intrinsic functions perform
				6246	a signed multiplication of the two arguments, and indicate whether an overflow
				6247	occurred during the signed multiplication.</p>
				6248
				6249	<h5>Arguments:</h5>
				6250
				6251	<p>The arguments (%a and %b) and the first element of the result structure may
				6252	be of integer types of any bit width, but they must have the same bit width. The
				6253	second element of the result structure must be of type <tt>i1</tt>. <tt>%a</tt>
				6254	and <tt>%b</tt> are the two values that will undergo signed multiplication.</p>
				6255
				6256	<h5>Semantics:</h5>
				6257
				6258	<p>The '<tt>llvm.smul.with.overflow</tt>' family of intrinsic functions perform
				6259	a signed multiplication of the two arguments. They return a structure —
				6260	the first element of which is the multiplication, and the second element of
				6261	which is a bit specifying if the signed multiplication resulted in an
				6262	overflow.</p>
				6263
				6264	<h5>Examples:</h5>
				6265	<pre>
				6266	%res = call {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
				6267	%sum = extractvalue {i32, i1} %res, 0
				6268	%obit = extractvalue {i32, i1} %res, 1
				6269	br i1 %obit, label %overflow, label %normal
				6270	</pre>
				6271
Reid Spencer	5bf54c8	2007-04-11 23:23:49 +0000	[diff] [blame]	6272	</div>
				6273
Bill Wendling	b9a7327	2009-02-08 23:00:09 +0000	[diff] [blame]	6274	<!-- _______________________________________________________________________ -->
				6275	<div class="doc_subsubsection">
				6276	<a name="int_umul_overflow">'<tt>llvm.umul.with.overflow.*</tt>' Intrinsics</a>
				6277	</div>
				6278
				6279	<div class="doc_text">
				6280
				6281	<h5>Syntax:</h5>
				6282
				6283	<p>This is an overloaded intrinsic. You can use <tt>llvm.umul.with.overflow</tt>
				6284	on any integer bit width.</p>
				6285
				6286	<pre>
				6287	declare {i16, i1} @llvm.umul.with.overflow.i16(i16 %a, i16 %b)
				6288	declare {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
				6289	declare {i64, i1} @llvm.umul.with.overflow.i64(i64 %a, i64 %b)
				6290	</pre>
				6291
				6292	<h5>Overview:</h5>
				6293
				6294	<p><i><b>Warning:</b> '<tt>llvm.umul.with.overflow</tt>' is badly broken. It is
				6295	actively being fixed, but it should not currently be used!</i></p>
				6296
				6297	<p>The '<tt>llvm.umul.with.overflow</tt>' family of intrinsic functions perform
				6298	a unsigned multiplication of the two arguments, and indicate whether an overflow
				6299	occurred during the unsigned multiplication.</p>
				6300
				6301	<h5>Arguments:</h5>
				6302
				6303	<p>The arguments (%a and %b) and the first element of the result structure may
				6304	be of integer types of any bit width, but they must have the same bit width. The
				6305	second element of the result structure must be of type <tt>i1</tt>. <tt>%a</tt>
				6306	and <tt>%b</tt> are the two values that will undergo unsigned
				6307	multiplication.</p>
				6308
				6309	<h5>Semantics:</h5>
				6310
				6311	<p>The '<tt>llvm.umul.with.overflow</tt>' family of intrinsic functions perform
				6312	an unsigned multiplication of the two arguments. They return a structure —
				6313	the first element of which is the multiplication, and the second element of
				6314	which is a bit specifying if the unsigned multiplication resulted in an
				6315	overflow.</p>
				6316
				6317	<h5>Examples:</h5>
				6318	<pre>
				6319	%res = call {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
				6320	%sum = extractvalue {i32, i1} %res, 0
				6321	%obit = extractvalue {i32, i1} %res, 1
				6322	br i1 %obit, label %overflow, label %normal
				6323	</pre>
				6324
				6325	</div>
				6326
Chris Lattner	941515c	2004-01-06 05:31:32 +0000	[diff] [blame]	6327	<!-- ======================================================================= -->
				6328	<div class="doc_subsection">
				6329	<a name="int_debugger">Debugger Intrinsics</a>
				6330	</div>
				6331
				6332	<div class="doc_text">
				6333	<p>
				6334	The LLVM debugger intrinsics (which all start with <tt>llvm.dbg.</tt> prefix),
				6335	are described in the <a
				6336	href="SourceLevelDebugging.html#format_common_intrinsics">LLVM Source Level
				6337	Debugging</a> document.
				6338	</p>
				6339	</div>
				6340
				6341
Jim Laskey	2211f49	2007-03-14 19:31:19 +0000	[diff] [blame]	6342	<!-- ======================================================================= -->
				6343	<div class="doc_subsection">
				6344	<a name="int_eh">Exception Handling Intrinsics</a>
				6345	</div>
				6346
				6347	<div class="doc_text">
				6348	<p> The LLVM exception handling intrinsics (which all start with
				6349	<tt>llvm.eh.</tt> prefix), are described in the <a
				6350	href="ExceptionHandling.html#format_common_intrinsics">LLVM Exception
				6351	Handling</a> document. </p>
				6352	</div>
				6353
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	6354	<!-- ======================================================================= -->
				6355	<div class="doc_subsection">
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6356	<a name="int_trampoline">Trampoline Intrinsic</a>
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6357	</div>
				6358
				6359	<div class="doc_text">
				6360	<p>
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6361	This intrinsic makes it possible to excise one parameter, marked with
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6362	the <tt>nest</tt> attribute, from a function. The result is a callable
				6363	function pointer lacking the nest parameter - the caller does not need
				6364	to provide a value for it. Instead, the value to use is stored in
				6365	advance in a "trampoline", a block of memory usually allocated
				6366	on the stack, which also contains code to splice the nest value into the
				6367	argument list. This is used to implement the GCC nested function address
				6368	extension.
				6369	</p>
				6370	<p>
				6371	For example, if the function is
				6372	<tt>i32 f(i8* nest %c, i32 %x, i32 %y)</tt> then the resulting function
Bill Wendling	252570f	2007-09-22 09:23:55 +0000	[diff] [blame]	6373	pointer has signature <tt>i32 (i32, i32)*</tt>. It can be created as follows:</p>
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6374	<pre>
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6375	%tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
				6376	%tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
				6377	%p = call i8* @llvm.init.trampoline( i8* %tramp1, i8* bitcast (i32 (i8* nest , i32, i32)* @f to i8), i8 %nval )
				6378	%fp = bitcast i8* %p to i32 (i32, i32)*
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6379	</pre>
Bill Wendling	252570f	2007-09-22 09:23:55 +0000	[diff] [blame]	6380	<p>The call <tt>%val = call i32 %fp( i32 %x, i32 %y )</tt> is then equivalent
				6381	to <tt>%val = call i32 %f( i8* %nval, i32 %x, i32 %y )</tt>.</p>
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6382	</div>
				6383
				6384	<!-- _______________________________________________________________________ -->
				6385	<div class="doc_subsubsection">
				6386	<a name="int_it">'<tt>llvm.init.trampoline</tt>' Intrinsic</a>
				6387	</div>
				6388	<div class="doc_text">
				6389	<h5>Syntax:</h5>
				6390	<pre>
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6391	declare i8* @llvm.init.trampoline(i8* <tramp>, i8* <func>, i8* <nval>)
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6392	</pre>
				6393	<h5>Overview:</h5>
				6394	<p>
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6395	This fills the memory pointed to by <tt>tramp</tt> with code
				6396	and returns a function pointer suitable for executing it.
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6397	</p>
				6398	<h5>Arguments:</h5>
				6399	<p>
				6400	The <tt>llvm.init.trampoline</tt> intrinsic takes three arguments, all
				6401	pointers. The <tt>tramp</tt> argument must point to a sufficiently large
				6402	and sufficiently aligned block of memory; this memory is written to by the
Duncan Sands	f2bcd37	2007-08-22 23:39:54 +0000	[diff] [blame]	6403	intrinsic. Note that the size and the alignment are target-specific - LLVM
				6404	currently provides no portable way of determining them, so a front-end that
				6405	generates this intrinsic needs to have some target-specific knowledge.
				6406	The <tt>func</tt> argument must hold a function bitcast to an <tt>i8*</tt>.
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6407	</p>
				6408	<h5>Semantics:</h5>
				6409	<p>
				6410	The block of memory pointed to by <tt>tramp</tt> is filled with target
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6411	dependent code, turning it into a function. A pointer to this function is
				6412	returned, but needs to be bitcast to an
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6413	<a href="#int_trampoline">appropriate function pointer type</a>
Duncan Sands	86e0119	2007-09-11 14:10:23 +0000	[diff] [blame]	6414	before being called. The new function's signature is the same as that of
				6415	<tt>func</tt> with any arguments marked with the <tt>nest</tt> attribute
				6416	removed. At most one such <tt>nest</tt> argument is allowed, and it must be
				6417	of pointer type. Calling the new function is equivalent to calling
				6418	<tt>func</tt> with the same argument list, but with <tt>nval</tt> used for the
				6419	missing <tt>nest</tt> argument. If, after calling
				6420	<tt>llvm.init.trampoline</tt>, the memory pointed to by <tt>tramp</tt> is
				6421	modified, then the effect of any later call to the returned function pointer is
				6422	undefined.
Duncan Sands	644f917	2007-07-27 12:58:54 +0000	[diff] [blame]	6423	</p>
				6424	</div>
				6425
				6426	<!-- ======================================================================= -->
				6427	<div class="doc_subsection">
Andrew Lenharth	9b254ee	2008-02-16 01:24:58 +0000	[diff] [blame]	6428	<a name="int_atomics">Atomic Operations and Synchronization Intrinsics</a>
				6429	</div>
				6430
				6431	<div class="doc_text">
				6432	<p>
				6433	These intrinsic functions expand the "universal IR" of LLVM to represent
				6434	hardware constructs for atomic operations and memory synchronization. This
				6435	provides an interface to the hardware, not an interface to the programmer. It
Chris Lattner	67c37d1	2008-08-05 18:29:16 +0000	[diff] [blame]	6436	is aimed at a low enough level to allow any programming models or APIs
				6437	(Application Programming Interfaces) which
Andrew Lenharth	9b254ee	2008-02-16 01:24:58 +0000	[diff] [blame]	6438	need atomic behaviors to map cleanly onto it. It is also modeled primarily on
				6439	hardware behavior. Just as hardware provides a "universal IR" for source
				6440	languages, it also provides a starting point for developing a "universal"
				6441	atomic operation and synchronization IR.
				6442	</p>
				6443	<p>
				6444	These do <em>not</em> form an API such as high-level threading libraries,
				6445	software transaction memory systems, atomic primitives, and intrinsic
				6446	functions as found in BSD, GNU libc, atomic_ops, APR, and other system and
				6447	application libraries. The hardware interface provided by LLVM should allow
				6448	a clean implementation of all of these APIs and parallel programming models.
				6449	No one model or paradigm should be selected above others unless the hardware
				6450	itself ubiquitously does so.
				6451
				6452	</p>
				6453	</div>
				6454
				6455	<!-- _______________________________________________________________________ -->
				6456	<div class="doc_subsubsection">
				6457	<a name="int_memory_barrier">'<tt>llvm.memory.barrier</tt>' Intrinsic</a>
				6458	</div>
				6459	<div class="doc_text">
				6460	<h5>Syntax:</h5>
				6461	<pre>
				6462	declare void @llvm.memory.barrier( i1 <ll>, i1 <ls>, i1 <sl>, i1 <ss>,
				6463	i1 <device> )
				6464
				6465	</pre>
				6466	<h5>Overview:</h5>
				6467	<p>
				6468	The <tt>llvm.memory.barrier</tt> intrinsic guarantees ordering between
				6469	specific pairs of memory access types.
				6470	</p>
				6471	<h5>Arguments:</h5>
				6472	<p>
				6473	The <tt>llvm.memory.barrier</tt> intrinsic requires five boolean arguments.
				6474	The first four arguments enables a specific barrier as listed below. The fith
				6475	argument specifies that the barrier applies to io or device or uncached memory.
				6476
				6477	</p>
				6478	<ul>
				6479	<li><tt>ll</tt>: load-load barrier</li>
				6480	<li><tt>ls</tt>: load-store barrier</li>
				6481	<li><tt>sl</tt>: store-load barrier</li>
				6482	<li><tt>ss</tt>: store-store barrier</li>
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	6483	<li><tt>device</tt>: barrier applies to device and uncached memory also.</li>
Andrew Lenharth	9b254ee	2008-02-16 01:24:58 +0000	[diff] [blame]	6484	</ul>
				6485	<h5>Semantics:</h5>
				6486	<p>
				6487	This intrinsic causes the system to enforce some ordering constraints upon
				6488	the loads and stores of the program. This barrier does not indicate
				6489	<em>when</em> any events will occur, it only enforces an <em>order</em> in
				6490	which they occur. For any of the specified pairs of load and store operations
				6491	(f.ex. load-load, or store-load), all of the first operations preceding the
				6492	barrier will complete before any of the second operations succeeding the
				6493	barrier begin. Specifically the semantics for each pairing is as follows:
				6494	</p>
				6495	<ul>
				6496	<li><tt>ll</tt>: All loads before the barrier must complete before any load
				6497	after the barrier begins.</li>
				6498
				6499	<li><tt>ls</tt>: All loads before the barrier must complete before any
				6500	store after the barrier begins.</li>
				6501	<li><tt>ss</tt>: All stores before the barrier must complete before any
				6502	store after the barrier begins.</li>
				6503	<li><tt>sl</tt>: All stores before the barrier must complete before any
				6504	load after the barrier begins.</li>
				6505	</ul>
				6506	<p>
				6507	These semantics are applied with a logical "and" behavior when more than one
				6508	is enabled in a single memory barrier intrinsic.
				6509	</p>
				6510	<p>
				6511	Backends may implement stronger barriers than those requested when they do not
				6512	support as fine grained a barrier as requested. Some architectures do not
				6513	need all types of barriers and on such architectures, these become noops.
				6514	</p>
				6515	<h5>Example:</h5>
				6516	<pre>
				6517	%ptr = malloc i32
				6518	store i32 4, %ptr
				6519
				6520	%result1 = load i32* %ptr <i>; yields {i32}:result1 = 4</i>
				6521	call void @llvm.memory.barrier( i1 false, i1 true, i1 false, i1 false )
				6522	<i>; guarantee the above finishes</i>
				6523	store i32 8, %ptr <i>; before this begins</i>
				6524	</pre>
				6525	</div>
				6526
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6527	<!-- _______________________________________________________________________ -->
				6528	<div class="doc_subsubsection">
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6529	<a name="int_atomic_cmp_swap">'<tt>llvm.atomic.cmp.swap.*</tt>' Intrinsic</a>
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6530	</div>
				6531	<div class="doc_text">
				6532	<h5>Syntax:</h5>
				6533	<p>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6534	This is an overloaded intrinsic. You can use <tt>llvm.atomic.cmp.swap</tt> on
				6535	any integer bit width and for different address spaces. Not all targets
				6536	support all bit widths however.</p>
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6537
				6538	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6539	declare i8 @llvm.atomic.cmp.swap.i8.p0i8( i8* <ptr>, i8 <cmp>, i8 <val> )
				6540	declare i16 @llvm.atomic.cmp.swap.i16.p0i16( i16* <ptr>, i16 <cmp>, i16 <val> )
				6541	declare i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* <ptr>, i32 <cmp>, i32 <val> )
				6542	declare i64 @llvm.atomic.cmp.swap.i64.p0i64( i64* <ptr>, i64 <cmp>, i64 <val> )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6543
				6544	</pre>
				6545	<h5>Overview:</h5>
				6546	<p>
				6547	This loads a value in memory and compares it to a given value. If they are
				6548	equal, it stores a new value into the memory.
				6549	</p>
				6550	<h5>Arguments:</h5>
				6551	<p>
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6552	The <tt>llvm.atomic.cmp.swap</tt> intrinsic takes three arguments. The result as
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6553	well as both <tt>cmp</tt> and <tt>val</tt> must be integer values with the
				6554	same bit width. The <tt>ptr</tt> argument must be a pointer to a value of
				6555	this integer type. While any bit width integer may be used, targets may only
				6556	lower representations they support in hardware.
				6557
				6558	</p>
				6559	<h5>Semantics:</h5>
				6560	<p>
				6561	This entire intrinsic must be executed atomically. It first loads the value
				6562	in memory pointed to by <tt>ptr</tt> and compares it with the value
				6563	<tt>cmp</tt>. If they are equal, <tt>val</tt> is stored into the memory. The
				6564	loaded value is yielded in all cases. This provides the equivalent of an
				6565	atomic compare-and-swap operation within the SSA framework.
				6566	</p>
				6567	<h5>Examples:</h5>
				6568
				6569	<pre>
				6570	%ptr = malloc i32
				6571	store i32 4, %ptr
				6572
				6573	%val1 = add i32 4, 4
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6574	%result1 = call i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* %ptr, i32 4, %val1 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6575	<i>; yields {i32}:result1 = 4</i>
				6576	%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
				6577	%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
				6578
				6579	%val2 = add i32 1, 1
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6580	%result2 = call i32 @llvm.atomic.cmp.swap.i32.p0i32( i32* %ptr, i32 5, %val2 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6581	<i>; yields {i32}:result2 = 8</i>
				6582	%stored2 = icmp eq i32 %result2, 5 <i>; yields {i1}:stored2 = false</i>
				6583
				6584	%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 8</i>
				6585	</pre>
				6586	</div>
				6587
				6588	<!-- _______________________________________________________________________ -->
				6589	<div class="doc_subsubsection">
				6590	<a name="int_atomic_swap">'<tt>llvm.atomic.swap.*</tt>' Intrinsic</a>
				6591	</div>
				6592	<div class="doc_text">
				6593	<h5>Syntax:</h5>
				6594
				6595	<p>
				6596	This is an overloaded intrinsic. You can use <tt>llvm.atomic.swap</tt> on any
				6597	integer bit width. Not all targets support all bit widths however.</p>
				6598	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6599	declare i8 @llvm.atomic.swap.i8.p0i8( i8* <ptr>, i8 <val> )
				6600	declare i16 @llvm.atomic.swap.i16.p0i16( i16* <ptr>, i16 <val> )
				6601	declare i32 @llvm.atomic.swap.i32.p0i32( i32* <ptr>, i32 <val> )
				6602	declare i64 @llvm.atomic.swap.i64.p0i64( i64* <ptr>, i64 <val> )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6603
				6604	</pre>
				6605	<h5>Overview:</h5>
				6606	<p>
				6607	This intrinsic loads the value stored in memory at <tt>ptr</tt> and yields
				6608	the value from memory. It then stores the value in <tt>val</tt> in the memory
				6609	at <tt>ptr</tt>.
				6610	</p>
				6611	<h5>Arguments:</h5>
				6612
				6613	<p>
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6614	The <tt>llvm.atomic.swap</tt> intrinsic takes two arguments. Both the
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6615	<tt>val</tt> argument and the result must be integers of the same bit width.
				6616	The first argument, <tt>ptr</tt>, must be a pointer to a value of this
				6617	integer type. The targets may only lower integer representations they
				6618	support.
				6619	</p>
				6620	<h5>Semantics:</h5>
				6621	<p>
				6622	This intrinsic loads the value pointed to by <tt>ptr</tt>, yields it, and
				6623	stores <tt>val</tt> back into <tt>ptr</tt> atomically. This provides the
				6624	equivalent of an atomic swap operation within the SSA framework.
				6625
				6626	</p>
				6627	<h5>Examples:</h5>
				6628	<pre>
				6629	%ptr = malloc i32
				6630	store i32 4, %ptr
				6631
				6632	%val1 = add i32 4, 4
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6633	%result1 = call i32 @llvm.atomic.swap.i32.p0i32( i32* %ptr, i32 %val1 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6634	<i>; yields {i32}:result1 = 4</i>
				6635	%stored1 = icmp eq i32 %result1, 4 <i>; yields {i1}:stored1 = true</i>
				6636	%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 8</i>
				6637
				6638	%val2 = add i32 1, 1
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6639	%result2 = call i32 @llvm.atomic.swap.i32.p0i32( i32* %ptr, i32 %val2 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6640	<i>; yields {i32}:result2 = 8</i>
				6641
				6642	%stored2 = icmp eq i32 %result2, 8 <i>; yields {i1}:stored2 = true</i>
				6643	%memval2 = load i32* %ptr <i>; yields {i32}:memval2 = 2</i>
				6644	</pre>
				6645	</div>
				6646
				6647	<!-- _______________________________________________________________________ -->
				6648	<div class="doc_subsubsection">
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6649	<a name="int_atomic_load_add">'<tt>llvm.atomic.load.add.*</tt>' Intrinsic</a>
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6650
				6651	</div>
				6652	<div class="doc_text">
				6653	<h5>Syntax:</h5>
				6654	<p>
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6655	This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.add</tt> on any
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6656	integer bit width. Not all targets support all bit widths however.</p>
				6657	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6658	declare i8 @llvm.atomic.load.add.i8..p0i8( i8* <ptr>, i8 <delta> )
				6659	declare i16 @llvm.atomic.load.add.i16..p0i16( i16* <ptr>, i16 <delta> )
				6660	declare i32 @llvm.atomic.load.add.i32..p0i32( i32* <ptr>, i32 <delta> )
				6661	declare i64 @llvm.atomic.load.add.i64..p0i64( i64* <ptr>, i64 <delta> )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6662
				6663	</pre>
				6664	<h5>Overview:</h5>
				6665	<p>
				6666	This intrinsic adds <tt>delta</tt> to the value stored in memory at
				6667	<tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.
				6668	</p>
				6669	<h5>Arguments:</h5>
				6670	<p>
				6671
				6672	The intrinsic takes two arguments, the first a pointer to an integer value
				6673	and the second an integer value. The result is also an integer value. These
				6674	integer types can have any bit width, but they must all have the same bit
				6675	width. The targets may only lower integer representations they support.
				6676	</p>
				6677	<h5>Semantics:</h5>
				6678	<p>
				6679	This intrinsic does a series of operations atomically. It first loads the
				6680	value stored at <tt>ptr</tt>. It then adds <tt>delta</tt>, stores the result
				6681	to <tt>ptr</tt>. It yields the original value stored at <tt>ptr</tt>.
				6682	</p>
				6683
				6684	<h5>Examples:</h5>
				6685	<pre>
				6686	%ptr = malloc i32
				6687	store i32 4, %ptr
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6688	%result1 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 4 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6689	<i>; yields {i32}:result1 = 4</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6690	%result2 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 2 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6691	<i>; yields {i32}:result2 = 8</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6692	%result3 = call i32 @llvm.atomic.load.add.i32.p0i32( i32* %ptr, i32 5 )
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6693	<i>; yields {i32}:result3 = 10</i>
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6694	%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 15</i>
Andrew Lenharth	9552894	2008-02-21 06:45:13 +0000	[diff] [blame]	6695	</pre>
				6696	</div>
				6697
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6698	<!-- _______________________________________________________________________ -->
				6699	<div class="doc_subsubsection">
				6700	<a name="int_atomic_load_sub">'<tt>llvm.atomic.load.sub.*</tt>' Intrinsic</a>
				6701
				6702	</div>
				6703	<div class="doc_text">
				6704	<h5>Syntax:</h5>
				6705	<p>
				6706	This is an overloaded intrinsic. You can use <tt>llvm.atomic.load.sub</tt> on
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6707	any integer bit width and for different address spaces. Not all targets
				6708	support all bit widths however.</p>
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6709	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6710	declare i8 @llvm.atomic.load.sub.i8.p0i32( i8* <ptr>, i8 <delta> )
				6711	declare i16 @llvm.atomic.load.sub.i16.p0i32( i16* <ptr>, i16 <delta> )
				6712	declare i32 @llvm.atomic.load.sub.i32.p0i32( i32* <ptr>, i32 <delta> )
				6713	declare i64 @llvm.atomic.load.sub.i64.p0i32( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6714
				6715	</pre>
				6716	<h5>Overview:</h5>
				6717	<p>
				6718	This intrinsic subtracts <tt>delta</tt> to the value stored in memory at
				6719	<tt>ptr</tt>. It yields the original value at <tt>ptr</tt>.
				6720	</p>
				6721	<h5>Arguments:</h5>
				6722	<p>
				6723
				6724	The intrinsic takes two arguments, the first a pointer to an integer value
				6725	and the second an integer value. The result is also an integer value. These
				6726	integer types can have any bit width, but they must all have the same bit
				6727	width. The targets may only lower integer representations they support.
				6728	</p>
				6729	<h5>Semantics:</h5>
				6730	<p>
				6731	This intrinsic does a series of operations atomically. It first loads the
				6732	value stored at <tt>ptr</tt>. It then subtracts <tt>delta</tt>, stores the
				6733	result to <tt>ptr</tt>. It yields the original value stored at <tt>ptr</tt>.
				6734	</p>
				6735
				6736	<h5>Examples:</h5>
				6737	<pre>
				6738	%ptr = malloc i32
				6739	store i32 8, %ptr
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6740	%result1 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 4 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6741	<i>; yields {i32}:result1 = 8</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6742	%result2 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 2 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6743	<i>; yields {i32}:result2 = 4</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6744	%result3 = call i32 @llvm.atomic.load.sub.i32.p0i32( i32* %ptr, i32 5 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6745	<i>; yields {i32}:result3 = 2</i>
				6746	%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = -3</i>
				6747	</pre>
				6748	</div>
				6749
				6750	<!-- _______________________________________________________________________ -->
				6751	<div class="doc_subsubsection">
				6752	<a name="int_atomic_load_and">'<tt>llvm.atomic.load.and.*</tt>' Intrinsic</a><br>
				6753	<a name="int_atomic_load_nand">'<tt>llvm.atomic.load.nand.*</tt>' Intrinsic</a><br>
				6754	<a name="int_atomic_load_or">'<tt>llvm.atomic.load.or.*</tt>' Intrinsic</a><br>
				6755	<a name="int_atomic_load_xor">'<tt>llvm.atomic.load.xor.*</tt>' Intrinsic</a><br>
				6756
				6757	</div>
				6758	<div class="doc_text">
				6759	<h5>Syntax:</h5>
				6760	<p>
				6761	These are overloaded intrinsics. You can use <tt>llvm.atomic.load_and</tt>,
				6762	<tt>llvm.atomic.load_nand</tt>, <tt>llvm.atomic.load_or</tt>, and
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6763	<tt>llvm.atomic.load_xor</tt> on any integer bit width and for different
				6764	address spaces. Not all targets support all bit widths however.</p>
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6765	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6766	declare i8 @llvm.atomic.load.and.i8.p0i8( i8* <ptr>, i8 <delta> )
				6767	declare i16 @llvm.atomic.load.and.i16.p0i16( i16* <ptr>, i16 <delta> )
				6768	declare i32 @llvm.atomic.load.and.i32.p0i32( i32* <ptr>, i32 <delta> )
				6769	declare i64 @llvm.atomic.load.and.i64.p0i64( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6770
				6771	</pre>
				6772
				6773	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6774	declare i8 @llvm.atomic.load.or.i8.p0i8( i8* <ptr>, i8 <delta> )
				6775	declare i16 @llvm.atomic.load.or.i16.p0i16( i16* <ptr>, i16 <delta> )
				6776	declare i32 @llvm.atomic.load.or.i32.p0i32( i32* <ptr>, i32 <delta> )
				6777	declare i64 @llvm.atomic.load.or.i64.p0i64( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6778
				6779	</pre>
				6780
				6781	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6782	declare i8 @llvm.atomic.load.nand.i8.p0i32( i8* <ptr>, i8 <delta> )
				6783	declare i16 @llvm.atomic.load.nand.i16.p0i32( i16* <ptr>, i16 <delta> )
				6784	declare i32 @llvm.atomic.load.nand.i32.p0i32( i32* <ptr>, i32 <delta> )
				6785	declare i64 @llvm.atomic.load.nand.i64.p0i32( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6786
				6787	</pre>
				6788
				6789	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6790	declare i8 @llvm.atomic.load.xor.i8.p0i32( i8* <ptr>, i8 <delta> )
				6791	declare i16 @llvm.atomic.load.xor.i16.p0i32( i16* <ptr>, i16 <delta> )
				6792	declare i32 @llvm.atomic.load.xor.i32.p0i32( i32* <ptr>, i32 <delta> )
				6793	declare i64 @llvm.atomic.load.xor.i64.p0i32( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6794
				6795	</pre>
				6796	<h5>Overview:</h5>
				6797	<p>
				6798	These intrinsics bitwise the operation (and, nand, or, xor) <tt>delta</tt> to
				6799	the value stored in memory at <tt>ptr</tt>. It yields the original value
				6800	at <tt>ptr</tt>.
				6801	</p>
				6802	<h5>Arguments:</h5>
				6803	<p>
				6804
				6805	These intrinsics take two arguments, the first a pointer to an integer value
				6806	and the second an integer value. The result is also an integer value. These
				6807	integer types can have any bit width, but they must all have the same bit
				6808	width. The targets may only lower integer representations they support.
				6809	</p>
				6810	<h5>Semantics:</h5>
				6811	<p>
				6812	These intrinsics does a series of operations atomically. They first load the
				6813	value stored at <tt>ptr</tt>. They then do the bitwise operation
				6814	<tt>delta</tt>, store the result to <tt>ptr</tt>. They yield the original
				6815	value stored at <tt>ptr</tt>.
				6816	</p>
				6817
				6818	<h5>Examples:</h5>
				6819	<pre>
				6820	%ptr = malloc i32
				6821	store i32 0x0F0F, %ptr
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6822	%result0 = call i32 @llvm.atomic.load.nand.i32.p0i32( i32* %ptr, i32 0xFF )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6823	<i>; yields {i32}:result0 = 0x0F0F</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6824	%result1 = call i32 @llvm.atomic.load.and.i32.p0i32( i32* %ptr, i32 0xFF )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6825	<i>; yields {i32}:result1 = 0xFFFFFFF0</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6826	%result2 = call i32 @llvm.atomic.load.or.i32.p0i32( i32* %ptr, i32 0F )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6827	<i>; yields {i32}:result2 = 0xF0</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6828	%result3 = call i32 @llvm.atomic.load.xor.i32.p0i32( i32* %ptr, i32 0F )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6829	<i>; yields {i32}:result3 = FF</i>
				6830	%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = F0</i>
				6831	</pre>
				6832	</div>
				6833
				6834
				6835	<!-- _______________________________________________________________________ -->
				6836	<div class="doc_subsubsection">
				6837	<a name="int_atomic_load_max">'<tt>llvm.atomic.load.max.*</tt>' Intrinsic</a><br>
				6838	<a name="int_atomic_load_min">'<tt>llvm.atomic.load.min.*</tt>' Intrinsic</a><br>
				6839	<a name="int_atomic_load_umax">'<tt>llvm.atomic.load.umax.*</tt>' Intrinsic</a><br>
				6840	<a name="int_atomic_load_umin">'<tt>llvm.atomic.load.umin.*</tt>' Intrinsic</a><br>
				6841
				6842	</div>
				6843	<div class="doc_text">
				6844	<h5>Syntax:</h5>
				6845	<p>
				6846	These are overloaded intrinsics. You can use <tt>llvm.atomic.load_max</tt>,
				6847	<tt>llvm.atomic.load_min</tt>, <tt>llvm.atomic.load_umax</tt>, and
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6848	<tt>llvm.atomic.load_umin</tt> on any integer bit width and for different
				6849	address spaces. Not all targets
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6850	support all bit widths however.</p>
				6851	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6852	declare i8 @llvm.atomic.load.max.i8.p0i8( i8* <ptr>, i8 <delta> )
				6853	declare i16 @llvm.atomic.load.max.i16.p0i16( i16* <ptr>, i16 <delta> )
				6854	declare i32 @llvm.atomic.load.max.i32.p0i32( i32* <ptr>, i32 <delta> )
				6855	declare i64 @llvm.atomic.load.max.i64.p0i64( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6856
				6857	</pre>
				6858
				6859	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6860	declare i8 @llvm.atomic.load.min.i8.p0i8( i8* <ptr>, i8 <delta> )
				6861	declare i16 @llvm.atomic.load.min.i16.p0i16( i16* <ptr>, i16 <delta> )
				6862	declare i32 @llvm.atomic.load.min.i32..p0i32( i32* <ptr>, i32 <delta> )
				6863	declare i64 @llvm.atomic.load.min.i64..p0i64( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6864
				6865	</pre>
				6866
				6867	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6868	declare i8 @llvm.atomic.load.umax.i8.p0i8( i8* <ptr>, i8 <delta> )
				6869	declare i16 @llvm.atomic.load.umax.i16.p0i16( i16* <ptr>, i16 <delta> )
				6870	declare i32 @llvm.atomic.load.umax.i32.p0i32( i32* <ptr>, i32 <delta> )
				6871	declare i64 @llvm.atomic.load.umax.i64.p0i64( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6872
				6873	</pre>
				6874
				6875	<pre>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6876	declare i8 @llvm.atomic.load.umin.i8..p0i8( i8* <ptr>, i8 <delta> )
				6877	declare i16 @llvm.atomic.load.umin.i16.p0i16( i16* <ptr>, i16 <delta> )
				6878	declare i32 @llvm.atomic.load.umin.i32..p0i32( i32* <ptr>, i32 <delta> )
				6879	declare i64 @llvm.atomic.load.umin.i64..p0i64( i64* <ptr>, i64 <delta> )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6880
				6881	</pre>
				6882	<h5>Overview:</h5>
				6883	<p>
				6884	These intrinsics takes the signed or unsigned minimum or maximum of
				6885	<tt>delta</tt> and the value stored in memory at <tt>ptr</tt>. It yields the
				6886	original value at <tt>ptr</tt>.
				6887	</p>
				6888	<h5>Arguments:</h5>
				6889	<p>
				6890
				6891	These intrinsics take two arguments, the first a pointer to an integer value
				6892	and the second an integer value. The result is also an integer value. These
				6893	integer types can have any bit width, but they must all have the same bit
				6894	width. The targets may only lower integer representations they support.
				6895	</p>
				6896	<h5>Semantics:</h5>
				6897	<p>
				6898	These intrinsics does a series of operations atomically. They first load the
				6899	value stored at <tt>ptr</tt>. They then do the signed or unsigned min or max
				6900	<tt>delta</tt> and the value, store the result to <tt>ptr</tt>. They yield
				6901	the original value stored at <tt>ptr</tt>.
				6902	</p>
				6903
				6904	<h5>Examples:</h5>
				6905	<pre>
				6906	%ptr = malloc i32
				6907	store i32 7, %ptr
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6908	%result0 = call i32 @llvm.atomic.load.min.i32.p0i32( i32* %ptr, i32 -2 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6909	<i>; yields {i32}:result0 = 7</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6910	%result1 = call i32 @llvm.atomic.load.max.i32.p0i32( i32* %ptr, i32 8 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6911	<i>; yields {i32}:result1 = -2</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6912	%result2 = call i32 @llvm.atomic.load.umin.i32.p0i32( i32* %ptr, i32 10 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6913	<i>; yields {i32}:result2 = 8</i>
Mon P Wang	2c839d4	2008-07-30 04:36:53 +0000	[diff] [blame]	6914	%result3 = call i32 @llvm.atomic.load.umax.i32.p0i32( i32* %ptr, i32 30 )
Mon P Wang	6a49037	2008-06-25 08:15:39 +0000	[diff] [blame]	6915	<i>; yields {i32}:result3 = 8</i>
				6916	%memval1 = load i32* %ptr <i>; yields {i32}:memval1 = 30</i>
				6917	</pre>
				6918	</div>
Andrew Lenharth	9b254ee	2008-02-16 01:24:58 +0000	[diff] [blame]	6919
				6920	<!-- ======================================================================= -->
				6921	<div class="doc_subsection">
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	6922	<a name="int_general">General Intrinsics</a>
				6923	</div>
				6924
				6925	<div class="doc_text">
				6926	<p> This class of intrinsics is designed to be generic and has
				6927	no specific purpose. </p>
				6928	</div>
				6929
				6930	<!-- _______________________________________________________________________ -->
				6931	<div class="doc_subsubsection">
				6932	<a name="int_var_annotation">'<tt>llvm.var.annotation</tt>' Intrinsic</a>
				6933	</div>
				6934
				6935	<div class="doc_text">
				6936
				6937	<h5>Syntax:</h5>
				6938	<pre>
Tanya Lattner	bed1d4d	2007-06-18 23:42:37 +0000	[diff] [blame]	6939	declare void @llvm.var.annotation(i8* <val>, i8* <str>, i8* <str>, i32 <int> )
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	6940	</pre>
				6941
				6942	<h5>Overview:</h5>
				6943
				6944	<p>
				6945	The '<tt>llvm.var.annotation</tt>' intrinsic
				6946	</p>
				6947
				6948	<h5>Arguments:</h5>
				6949
				6950	<p>
Tanya Lattner	bed1d4d	2007-06-18 23:42:37 +0000	[diff] [blame]	6951	The first argument is a pointer to a value, the second is a pointer to a
				6952	global string, the third is a pointer to a global string which is the source
				6953	file name, and the last argument is the line number.
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	6954	</p>
				6955
				6956	<h5>Semantics:</h5>
				6957
				6958	<p>
Anton Korobeynikov	06cbb65	2008-01-15 22:31:34 +0000	[diff] [blame]	6959	This intrinsic allows annotation of local variables with arbitrary strings.
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	6960	This can be useful for special purpose optimizations that want to look for these
Anton Korobeynikov	06cbb65	2008-01-15 22:31:34 +0000	[diff] [blame]	6961	annotations. These have no other defined use, they are ignored by code
				6962	generation and optimization.
				6963	</p>
Tanya Lattner	cb1b960	2007-06-15 20:50:54 +0000	[diff] [blame]	6964	</div>
				6965
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	6966	<!-- _______________________________________________________________________ -->
				6967	<div class="doc_subsubsection">
Tanya Lattner	0186a65	2007-09-21 23:57:59 +0000	[diff] [blame]	6968	<a name="int_annotation">'<tt>llvm.annotation.*</tt>' Intrinsic</a>
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	6969	</div>
				6970
				6971	<div class="doc_text">
				6972
				6973	<h5>Syntax:</h5>
Tanya Lattner	23dbd57	2007-09-21 23:56:27 +0000	[diff] [blame]	6974	<p>This is an overloaded intrinsic. You can use '<tt>llvm.annotation</tt>' on
				6975	any integer bit width.
				6976	</p>
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	6977	<pre>
Tanya Lattner	cf3e26f	2007-09-22 00:03:01 +0000	[diff] [blame]	6978	declare i8 @llvm.annotation.i8(i8 <val>, i8* <str>, i8* <str>, i32 <int> )
				6979	declare i16 @llvm.annotation.i16(i16 <val>, i8* <str>, i8* <str>, i32 <int> )
				6980	declare i32 @llvm.annotation.i32(i32 <val>, i8* <str>, i8* <str>, i32 <int> )
				6981	declare i64 @llvm.annotation.i64(i64 <val>, i8* <str>, i8* <str>, i32 <int> )
				6982	declare i256 @llvm.annotation.i256(i256 <val>, i8* <str>, i8* <str>, i32 <int> )
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	6983	</pre>
				6984
				6985	<h5>Overview:</h5>
Tanya Lattner	23dbd57	2007-09-21 23:56:27 +0000	[diff] [blame]	6986
				6987	<p>
				6988	The '<tt>llvm.annotation</tt>' intrinsic.
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	6989	</p>
				6990
				6991	<h5>Arguments:</h5>
				6992
				6993	<p>
				6994	The first argument is an integer value (result of some expression),
				6995	the second is a pointer to a global string, the third is a pointer to a global
				6996	string which is the source file name, and the last argument is the line number.
Tanya Lattner	23dbd57	2007-09-21 23:56:27 +0000	[diff] [blame]	6997	It returns the value of the first argument.
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	6998	</p>
				6999
				7000	<h5>Semantics:</h5>
				7001
				7002	<p>
				7003	This intrinsic allows annotations to be put on arbitrary expressions
				7004	with arbitrary strings. This can be useful for special purpose optimizations
				7005	that want to look for these annotations. These have no other defined use, they
				7006	are ignored by code generation and optimization.
Dan Gohman	ef9462f	2008-10-14 16:51:45 +0000	[diff] [blame]	7007	</p>
Tanya Lattner	293c037	2007-09-21 22:59:12 +0000	[diff] [blame]	7008	</div>
Jim Laskey	2211f49	2007-03-14 19:31:19 +0000	[diff] [blame]	7009
Anton Korobeynikov	06cbb65	2008-01-15 22:31:34 +0000	[diff] [blame]	7010	<!-- _______________________________________________________________________ -->
				7011	<div class="doc_subsubsection">
				7012	<a name="int_trap">'<tt>llvm.trap</tt>' Intrinsic</a>
				7013	</div>
				7014
				7015	<div class="doc_text">
				7016
				7017	<h5>Syntax:</h5>
				7018	<pre>
				7019	declare void @llvm.trap()
				7020	</pre>
				7021
				7022	<h5>Overview:</h5>
				7023
				7024	<p>
				7025	The '<tt>llvm.trap</tt>' intrinsic
				7026	</p>
				7027
				7028	<h5>Arguments:</h5>
				7029
				7030	<p>
				7031	None
				7032	</p>
				7033
				7034	<h5>Semantics:</h5>
				7035
				7036	<p>
				7037	This intrinsics is lowered to the target dependent trap instruction. If the
				7038	target does not have a trap instruction, this intrinsic will be lowered to the
				7039	call of the abort() function.
				7040	</p>
				7041	</div>
				7042
Bill Wendling	1431331	2008-11-19 05:56:17 +0000	[diff] [blame]	7043	<!-- _______________________________________________________________________ -->
				7044	<div class="doc_subsubsection">
Misha Brukman	50de2b2	2008-11-22 23:55:29 +0000	[diff] [blame]	7045	<a name="int_stackprotector">'<tt>llvm.stackprotector</tt>' Intrinsic</a>
Bill Wendling	1431331	2008-11-19 05:56:17 +0000	[diff] [blame]	7046	</div>
				7047	<div class="doc_text">
				7048	<h5>Syntax:</h5>
				7049	<pre>
				7050	declare void @llvm.stackprotector( i8* <guard>, i8** <slot> )
				7051
				7052	</pre>
				7053	<h5>Overview:</h5>
				7054	<p>
				7055	The <tt>llvm.stackprotector</tt> intrinsic takes the <tt>guard</tt> and stores
				7056	it onto the stack at <tt>slot</tt>. The stack slot is adjusted to ensure that
				7057	it is placed on the stack before local variables.
				7058	</p>
				7059	<h5>Arguments:</h5>
				7060	<p>
				7061	The <tt>llvm.stackprotector</tt> intrinsic requires two pointer arguments. The
				7062	first argument is the value loaded from the stack guard
				7063	<tt>@__stack_chk_guard</tt>. The second variable is an <tt>alloca</tt> that
				7064	has enough space to hold the value of the guard.
				7065	</p>
				7066	<h5>Semantics:</h5>
				7067	<p>
				7068	This intrinsic causes the prologue/epilogue inserter to force the position of
				7069	the <tt>AllocaInst</tt> stack slot to be before local variables on the
				7070	stack. This is to ensure that if a local variable on the stack is overwritten,
				7071	it will destroy the value of the guard. When the function exits, the guard on
				7072	the stack is checked against the original guard. If they're different, then
				7073	the program aborts by calling the <tt>__stack_chk_fail()</tt> function.
				7074	</p>
				7075	</div>
				7076
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	7077	<!-- *********************************************************************** -->
Chris Lattner	2f7c963	2001-06-06 20:29:01 +0000	[diff] [blame]	7078	<hr>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	7079	<address>
				7080	<a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman	86242e1	2008-12-11 17:34:48 +0000	[diff] [blame]	7081	src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	7082	<a href="http://validator.w3.org/check/referer"><img
Misha Brukman	86242e1	2008-12-11 17:34:48 +0000	[diff] [blame]	7083	src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	7084
				7085	<a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
Reid Spencer	ca05854	2006-03-14 05:39:39 +0000	[diff] [blame]	7086	<a href="http://llvm.org">The LLVM Compiler Infrastructure</a><br>
Misha Brukman	c501f55	2004-03-01 17:47:27 +0000	[diff] [blame]	7087	Last modified: $Date$
				7088	</address>
Chris Lattner	b8f816e	2008-01-04 04:33:49 +0000	[diff] [blame]	7089
Misha Brukman	7630785	2003-11-08 01:05:38 +0000	[diff] [blame]	7090	</body>
				7091	</html>